You are on page 1of 174

<Insert Picture Here>

Oracle Business Intelligence Suite Enterprise Edition (BI EE) Samples Sales Content Guide (V1.3)
Oracle Business Intelligence Product Management Apr, 2009

Contents
1. Introduction
2. Dashboards and Reports Samples
A. B. Dashboards Overview Dashboard Details
01 History and Trends 02 Tiering and Toppers 03 Data Distribution 04 Dimensional Analysis

.......... Slide 5 ... Slide 14


.... .... .... .... 12 23 44 62

C.

Answers features & Segment example

... Slide 79

3. Samples Repository Overview


A. B. C. Overview How to Demo selected RPD features Logical Aggregations details ... Slide 83 ... Slide 91 ... Slide 109 ... Slide 150 ... Slide 169

4. Cloning Samples with CAF V1 Utility 5. Switching/Updating Datasets

Samples Sales Content Overview


A Sample Oracle BI application and data set called Sample Sales provided to illustrate functional capabilities of Oracle BI EE and numerous best practices

Web Presentation : Typical reports examples


Showcasing specific Answers features Answers formula building, Layouts Presentation Variables, Filtering, Navigations.

Repository : Examples of Metadata constructs


Intended to demonstrate design patterns and to showcase RPD modeling best practices Numerous logical aggregations Time series and Rolling X Months Variations, Time Span Variations and Compounded Variations Multi Physical Sourcing Dimensional Snow flaking Canonical Time Construct (Multi LTS Facts tables) Data security, projects, segmentation metadata, etc

OTN External Link


http://www.oracle.com/technology/products/bi/enterprise-edition.html

2.A Dashboards Overview


Overview of Samples Dashboards Content

Sample Sales Dashboards List


00 Overview 01 History and Trends
1 - History 2a - Seasonality 2b - Seasonality. Day 2c - Seasonality Qtr 3a - Trending Comparative 3b - Trending Day 3c - Trending Qtr 4 - Trend Lines

03 Data Distribution
13 - Boxplot 14 - Distribution 15 - Comparative Distribution 16 - Scatter 17 - Variability 18 - Standard Deviation 19 Statistical Process Control (SPC) 20 Deviants

02 Tiering and Toppers


5 - Eighty Twenty 6 - Tiering 7 - TopNs 8 - TopNs Heatmap 9 - TopNs History 10 - TopNs Frequency 11 - N-tiling 12 - Rank Changes

04 Dimensional Analysis
21- Indexing 22- Waterfall 23 - Waterfall Multidimensional 24 Benchmark 25 Index to Average 26 Index to Average Trended 27 Profile 28 Details

Overview Page
Index page listing all available Dashboards, allows quick navigation to any dashboards

01 History and Trends Dashboard

Yearly, quarterly, and monthly values and averages for a given metric

Shows pattern of monthly values by year, over multiple years. Leverages daily granular data from DB

Comparative monthly trend charts for several individuals on a single metric.

Comparative quarterly trend charts for several individuals on a single metric.

Shows pattern of monthly values by year, over multiple years.

Shows pattern of quarterly values by year, over multiple years

Comparative monthly trend charts for several individuals on a single metric. Leverages daily granular data from DB.

Displays trend lines and variations in the data by automatically grouping data points into homogeneous clusters of data individuals

02 Tiering and Toppers Dashboard

Contribution measure of upper tier of a specific population for a metric.

Measures how Top N individuals contribute to total aggregation of population, for two distinct metrics.

Shows historical amplitude of Top N and Bottom N value layers and gives a visual indication of how values for Toppers and Bottomers evolve over time comparatively.

Comparative analysis of dimension individuals for several metrics, ordered in descending way of values of a specific metric. 'How are my top revenue deciles contributing to profit ?'

Tiers total of a metric in clusters (tiers) of equal values, with descending order of individuals.

Combined set of views of Top Individuals for two distinct dimensions on a single metric. Includes bi-dimensional toppers matrix.

Shows the ranking variations on a metric for each individual in a dimension. Historical information on Top and Bottom N individuals, frequency of individuals qualifying for Tops and Bottoms groups.

03 Data Distribution Dashboard

Comparative graphical summary of a set of data. For each value in a dimension, it shows measures of central, average, dispersion and skewness.

Comparative representations of statistical distribution for a selected population, over a dimension. 'How do sales order size distribute every year ?'

Displays Top, Mid and Bottom tiles of a population, with extreme ranking individuals, range, IQR and basic descriptive measurements

Automatically groups data points into user selected number of clusters and highlights those data points that fall outside the control limits

Simple statistical discrete distribution of a population for one metric

Graphical summary of a set of data. Displays measures of central median, dispersion and skewness.

Comparative standard deviation analysis on a metric for a selected set of population, with a selected grain for analysis.

Identifies and highlights top sequences of consecutive deviant data points that fall outside user selected control limits

04 Dimensional Analysis Dashboard

Indexed representation of a metric for dimension individuals.

Shows how an initial value is increased and decreased by a series of intermediate values, breaking down details of dimension individuals for each intermediate value.

Performance of dimension individuals indexed against average of all individuals in report.

Profiles selected population by several attributes

Shows how an initial value is increased and decreased by a series of intermediate values.

Relative performance of individuals in a dimension, benchmarked against a fixed individual.

Shows trended comparative Tabular list of detailed records for performance for individuals from a selected population dimension, benchmarked against Average of individuals in this dimension

2.B Dashboard Details


01 History and Trends
Analyze trended information, history and seasonality

01 History
This report shows the monthly, quarterly and yearly historical values and averages for a given metric. The metric to be analyzed can be

chosen from the list of options available in the Select Metric dropdown.

01 History Help
1.1 Purpose: This report presents the value of a metric over a selected period for a time dimension. It displays aggregated values for each period in this interval: total yearly, quarterly, monthly and weekly, as well as average quarterly, monthly and weekly. The purpose of This report is to help understand the behavior of a given measure over time. The charts break down the information by Year, Quarter, Month, and Week and provide a snapshot of long term trends in the value of the measurement. It also provides an understanding of how a given period in particular relates to the total trend, as well as how periodical averages are impacted.

1.2 Usage: Select the metric to analyze from Select Metric drop down list of choices at top of page and select Go. If necessary, enter a division factor, for example, 1000 in Divide Metric Value by prompt at top of page and select Go. Filter the scope of the report as required by using Page Filter Prompts at bottom of page. Clicking on the blue colored numbers within tables will give you access to more detailed navigation reports.

02 Seasonality by Month (2a), Day(2b), Quarter(2c)


This report displays several views comparing how a specific metric value is evolving month to month over multiple years. It shows pattern of monthly values by year, for selected history. Report 2b (by day) provides ability to select the time object to analyze on.

02 Seasonality by Month Help


1.1 Purpose: This report displays several views to compare how a specific metric is evolving month to month over multiple years. It shows pattern of monthly values by year, for selected history. The analysis helps to understand monthly seasonality across years and how the values balance across periods within a year. It computes month values as a percentage of total year value in a way facilitating comparison between years. It also provides a cumulative representation of the full year, indicating the overall pace to completion of the year total. The analysis is especially useful in forecast and performance measurement processes.

1.2 Usage: Select the metric to analyze from Select Metric drop down list of choices at top of page and select Go . If necessary, enter a division factor. For example, enter 1000 in Divide Metric Value by prompt at top of the page and select Go button. Filter the scope of the report as required by using Page Filter Prompts at bottom of page. Clicking on the blue colored numbers within tables give you access to more detailed navigation reports.

02 Seasonality : Alternate Views


2b Seasonality accessing Day Grain data : provides the flexibility to select the time object to analyze on.

2b Seasonality at Quarter level

03 Trending comparative Month (3a), Day(3b), Qter(3c)


This reports display historical representations of the selected metric, presented individually in selected dimensions. It also provides a quick visual comparison of year on year evolution of metric values by month or quarter for distinct individuals in a dimension. Report 3b (by day) provides ability to select the time object to analyze on.

03 Trending Help
1.1 Purpose: This report displays historical representations of the selected metric, presented individually in selected dimensions. It also provides a quick visual comparison of year on year evolution of metric values by month for distinct individuals in a dimension. This analysis helps understand the behavior of a given metric over time for each year, in comparison to individual dimensions. It gives insight on how individuals of a dimension perform differently over time, for a single metric. The chart displays details of business issues over time which may remain unnoticed if we look only at aggregate time value or aggregate dimensional values.

1.2 Usage: From drop down list of choices at top of page, select a metric to analyze from Select Metric

prompt and select a dimension from Select Dimension prompt and select Go. The dimension
values will each generate a series of three charts (a row of charts on the page) . Note: This report generates a chart for every value in the dimension selected. Choosing a dimension with a large number of values might lead to difficulties in generating the chart and the report not working properly. If necessary, enter a division factor , for example 1000, in Divide Metric Value by prompt at top of the page and select Go. Filter the scope of the report as required by using Page Filter Prompts at bottom of page. To view tables with detailed data, select the option from Select information to display : drop down.

03 Trending Comparative : Alternate Views


3b Trending accessing Day Grain data : provides the flexibility to select the time object to analyze on. Detailed table view (in all trending reports) Shows detailed chart values in multi blocks pivot tables

3b Trending Quarter

04 Trend Lines
This report identifies and displays trending lines in metric values over time. This analysis automatically groups data points into user selected number of clusters and then determines linear trends in metric values in each cluster.

04 Trend Lines Help


1.1 Purpose:
This report identifies and displays trending lines in metric values over time. This analysis automatically groups data points into user selected number of clusters and then calculates linear trending algorithm for each cluster. A trend line is a momentum indicator, it measures the rate of increase or decrease in the metric value over time. Clusters breaking can alert to any acceleration or deceleration of the trend. 1.2 Usage: From drop down list of choices at the top of page, select a metric to analyze from "Select Metric" prompt and select a dimension from "Select Dimension" prompt and select Go. If required, user may also adjust Control zone bandwidth (expressed in # of standard deviations). The number of sigmas expressed will increase the control band above and beyond the average line. User may adjust No of clusters : the number of clusters will define how many breaking points will be inferred from the detailed data. For more detailed analysis on any particular cluster, user may select corresponding down arrow icon in second column on table beside the bubble chart Filter the scope of the report as required by using "Page Filter Prompts" at bottom of page.

2.B Dashboard Details


02 Tiering and Toppers
Analysis of top and bottom rankers, tiers, deciles, eighty-twenty paretos and historical variations of top layers of individuals.

05 Eighty Twenty
This report provides several dynamic views to help measure how the upper tier of a specific population set contributes in descending order of value. Filters at the top of the page enable users to set that percent limit of value; the report then renders the corresponding percent of population that makes up that value.

05 Eighty Twenty Help


1.1 Purpose: This report provides several dynamic views to help measure how the upper tier of a specific population set contributes in descending order of value. The filters at the top enable users to set that percent limit of value; the report then renders the corresponding percent of population that makes up that value. For example, setting the filter to "80" marks up the records in the entire

population that make up 80% of the value.


This report provides users insight on where to focus for a particular analysis. It answers the question "What part of the population should I focus on to be sure to address the most significant part of the value/ problem ?

1.2 Usage: From drop down list of choices at top of page, select a metric to analyze from "Select Metric" prompt and select a dimension from "Select Dimension" prompt and select Go. Set the '%' limit to desired value on the prompt at the top of the page and hit "Go" button. Filter the scope of the report as required by using "Page Filter Prompts" at bottom of page. To view tables with detailed data, select the option from "Select information to display :" drop down

05 Eighty Twenty : Alternate Views


Detailed table view Shows detailed values by buckets, in multi blocks pivot tables

06 Tiering
This report provides a visual distribution of the population after grouping the value of a metric into a given number of tiers. It ranks individuals in descending order and then groups them in buckets of equal value (not equal counts). The report then displays the counts per tiers and how other metrics distribute according to this tier split up. The number of tiers in the report is dynamically set by the user.

06 Tiering Help
1.1 Purpose: This report provides a visual distribution of the population after grouping the value of a metric into a given number of tiers. It ranks individuals in descending order and then groups them in buckets of equal value (not equal counts).

The report then displays the counts per tiers and how other
metrics distribute according to this tier split up. The number of tiers in the report is dynamically set by the user. This report is very useful to gather a high level idea of how a value distributes across a population. For example, how many customer make up first third of my revenue, vs. how many in the second third, vs. how many in the last third. Also, how is my profit for the population that composes the first tier of my revenue? This report will visually display answers to these questions and can also provide with details in tabular form.

1.2 Usage: From drop down list of choices at top of page, select a metric to analyze from "Select Metric 1" prompt and select a second metric from "Select Metric 2" prompt. Select a dimension from "Select Dimension" prompt and select Go. Set the number of tiers to desired value on the prompt at the top of the page and hit "Go" button. Filter the scope of the report as required by using "Page Filter Prompts" at bottom of page. To view tables with detailed data, select the option from "Select information to display :" drop down.

06 Tiering : Alternate Views


Detailed table view Shows detailed values by tier, in multi blocks formatted pivot table

07 TopNs
This report provides views of top individuals in the selected dimension along with an aggregation of all the non Top N individuals, per distinct metrics values for selected metrics. Filtering on the reports limits it to show those individuals that belong to the top N list for at least one of the metrics considered.

07 TopNs Help
1.1 Purpose: This report provides views of top individuals in the selected dimension along with an aggregation of all the non Top N individuals, per distinct metrics values for selected metrics. Filtering on the reports limits it to show those individuals

that belong to the top N list for at least one of the metrics
considered. This report is useful to identify the top performers on a given metric, and to aggregate their importance relative to the total population. This dashboard helps users to understand the impact of top performers in the context of the whole business and may contribute towards business decisions and actions on the top individuals.

1.2 Usage: From drop down list of choices at top of page, select a metric to analyze from "Select Metric 1" prompt and select a second metric from "Select Metric 2" prompt. Select a dimension from "Select Dimension" prompt and select Go. Set the 'N' limit to desired value on the prompt at the top of the page and hit "Go" button. Filter the scope of the report as required by using "Page Filter Prompts" at bottom of page.

08 TopNs Heatmap
This report uses color to represent top individuals for selected dimensions in a two-dimensional table, showing the value of a selected metric. The heat map matrix displays all individuals in selected dimensions and the overall rank of their crossing metric value. It is color coded so as to quickly distinguish between leader, top 10s, 20s, 30s, 40s, 50s, 60s individuals and the rest.

08 TopNs Heatmap Help


1.1 Purpose: This report uses color to represent top individuals for selected dimensions in a two-dimensional table, showing the value of a selected metric. The heat map matrix displays all individuals in selected dimensions and the overall rank of their crossing metric value. It is color coded so as to quickly distinguish between leader, top 10s, 20s, 30s, 40s, 50s, 60s individuals and the rest. Heatmaps give an edge in identifying critical issues, allocating resources, finding trends or relationships and creating quicker insight on large data sets. Trends, anomalies, distributions and relationship can be seen in context and at multiple levels at once. The report will help to quickly identify the top performers on a given metric, in a cross dimensional context. and to aggregate their importance relative to the total population. It This dashboard helps users to understand the impact of top performers in the context of the whole business and may contribute towards business decisions and actions on the top individuals. For example, what are my top selling region - products intersections ? How is the distribution of region / product. top sellers spread on the population, any visible correlation ? 1.2 Usage: From drop down list of choices at the top of page, select two dimensions from "Select Dimension 1" and "Select Dimension 2" prompts. These dimensions are used respectively as columns and rows of the heatmap. It is important that you select two different dimensions there in order for the heatmap to be meaningful. Set N limit to limit the number of toppers displayed and select Go, Select the metric to analyze from "Select Metric 1" prompt and select Go. Filter the scope of the report as required by using Page Filter Prompts at the bottom of page.

09 TopNs History
This page displays historical amplitude of Top N and Bottom N layers and gives a visual indication of how values for Toppers and Bottomers evolve over time comparatively. For each month, the report indicates how much of the total value was represented by Top and Bottom Ns, what was the floor/limit to qualify for toppers, as well as how much in percentage each group represented to the total value.

09 TopNs History Help


1.1 Purpose: This page displays historical amplitude of Top N and Bottom N value layers and gives a visual indication of how values for Toppers and Bottomers evolve over time comparatively. For each month in the range selected, this report indicates how much of the total value was represented by Top and Bottom Ns, what was the floor/limit to qualify for toppers, as well as how much in

percentage each group represented to the total value.

1.2 Usage: From drop down list of choices at top of page, select the metric to analyze from Select Metric 1 prompt and select Go. From the top of the page, select Dimension 1 and Dimension 2 and select Go . From the

top of the page, select the Top and Bottom N value in the Set N (Top & Bottom) and select
Go .This value sets the maximum Rank limit for individuals to qualify for Top N and Bottom N ranges calculations. Filter the scope of the report as required by using Page Filter Prompts at bottom of page.

10 TopNs Frequency
This report displays historical information about Top N and Bottom N individuals. For each month in the range selected, this analysis will indicate which individuals were the Top or Bottom N, and, for each, how many times each of them made it in the Top or Bottom N during the range of time selected.

10 TopNs Frequency Help


1.1 Purpose: This report displays historical information about Top N and Bottom N individuals. For each month in the range

selected, this analysis will indicate which individuals were


the Top or Bottom N, and, for each, how many times each of them made it in the Top or Bottom N during the range of time selected. The information on this report is useful to understand who are the individuals that regularly make it in top or bottom N monthly podium, over a period of many months. There can be lower business risks with a population where names on monthly top N positions are regularly rotating, versus a situation where the list of monthly toppers over a long period of time is very short. Similarly Bottomers that never move out of the bottom zone are indicative of no relative business improvements and need consideration.

1.2 Usage: From drop down list of choices at top of page, select the metric to analyze from Select Metric prompt and select Go. From the top of the page, select Dimension 1 and Dimension 2 and select Go . From the top of the page, select the Top N value in the Set N(Top & Bottom) and select Go . The value entered here is set as the rank limit for Top and Bottom N qualifiers in the report Filter the scope of the report as required by using Page Filter Prompts at bottom of page.

10 TopNs Frequency : Alternate Views


Detailed records view Shows detailed TopNs records values by rank

11 Ntiling
This report shows how the different tiles of individuals on metric 1 value contribute to metric 2 values. It arranges individuals in ascending order of metric 1 value and groups them into user selected number of tiles. It then displays the exact same tiles and how they relate to the metric 2 values. A color code visually indicates if contribution of individuals in a tile is relatively higher or lesser toward metric 2 than towards metric 1.

11 Ntiling Help
1.1 Purpose: This report shows how the different tiles of individuals on metric 1 value contribute to metric 2 values. It arranges the individuals in ascending order of the metric 1 value and groups them into user selected number of tiles. It then displays the exact same tiles and how they relate to the metric 2 values. A color code visually indicates if contribution of individuals in a tile is relatively greater or lesser toward metric 2 than towards metric 1. This analysis is useful to understand how the same set of individuals' contributions to different metrics vary. For example, you can get answers to the following questions using this dashboard: Are my top revenue customers making up most of my profits ? How do small customers contribute to my costs ? Are my top costing plants making the most of my revenue ? ... This analysis can lead to interesting conclusions such as the customers in the highest revenue decile may not be the top most profitable customers. 1.2 Usage: From drop down list of choices at top of page, select two metrics to analyze from "Select Metric 1" and "Select Metric 2" prompts; select the dimension and select Go . The metric 1 is the one used for the tiling of the dimension individuals.

At the top of the page, set number of tiles (for example, 10) and select Go . This determines how
many bars display in the charts. Filter the scope of the report as required by using Page Filter Prompts at bottom of page. To view tables with detailed data, select the option from "Select information to display :" drop down

11 Ntiling : Alternate Views


Detailed Table view Shows detailed individual values by Ntiles, organized in multi-blocks pivot table

12 Rank Changes
This report displays information on Top individuals in a dimension with a condition upon amplitude of variations in their ranking on a

measurement, from one month to another one. The filtering at the top of the report provides users with the flexibility to reduce the scope
of the analysis to only top items with a minimum variation differential over time.

12 Rank Changes Help


1.1 Purpose: This report displays information on Top individuals in a dimension with a condition upon amplitude of variations in their ranking on a measurement, from one month to another one. The LV filtering on the report provides users with the flexibility to reduce the scope of the analysis to only top

items with a minimum variation differential over time.


1.2 Usage: From drop down list of choices at top of page, select metrics to analyze from Select Metric , Select Period Ago Metric 1 and Select Period Ago Metric 2 prompts and select Go. From the top of the page, Select Dimension and select Go . From the top of the page, set the Top N Limit value in the Set Top N Limit drop down and select Go .This allows users to fix the limit of top positions they elect for displaying in the report. This filter applies to each period in the report set. For example a limit set to 'top 5' means to show any individual that made it at least once in the top 5, either in current month, the month before or quarter ago month. Again, the month ago and quarter ago can be replaced by any period ago metric. The report will then display the rank variations for this population. From the top of the page, set the Rank Variation Limit value in the Set Rk Var Limit drop down and select Go .This allows the user to fix the minimum absolute value for rank variations he wants to see in the report. For example, setting this value to 3 results in items that have increased or decreased their rank by at least 3 positions between current and last month, or between current and quarter ago month. Any lower rank variations will not show. This limit filtering is applied in addition to the filter rule set in the 'Set Top N Limit' top page dashboard prompt block.

2.B Dashboard Details


03 Data Distribution
Various data distribution representations and simple statistical analysis

13 Boxplot
This report displays a boxplot whisker diagram comparing the spread of detailed data point values between individuals of a dimension. It depicts a set of values for each dimension individual through seven number summaries: smallest observation (Bottom), lower decile (10% mark), lower quartile and upper quartile (IQR), Median and Average, upper decile (90% mark), and largest observation (Top).

13 Boxplot Help
1.1 Purpose: This report displays a boxplot whisker diagram comparing the spread of detailed data point values between individuals of a dimension. It depicts a set of values for each dimension individual through seven number summaries: smallest observation (Bottom), lower decile (10% mark), lower quartile and upper quartile (IQR), Median and Average, upper decile (90% mark), and largest observation (Top). Boxplots are useful to display differences between populations datasets without any assumptions of the underlying statistical distribution. The spacing between the different parts of the box indicate the degree of dispersion (spread) and skewness in the data, and identify outliers. A boxplot report provides users with immediate visual insight on where to focus for a

particular individual of a dimension


1.2 Usage: From drop down list of choices at top of page, select two dimensions to do the comparative analysis on. Dimension 1 defines the comparative individuals (X axis values on boxplot charts). Dimension 2 defines the grain for each dispersion analysis. Use an aggregated column for dimension 1, so not to have too many values on X axis for the charts, and a

granular column for dimension 2 (dispersion analysis requires a dataset with multiple
datapoints to render meaningful charts). From drop down list of choices at top of page, select the values to run Boxplot analysis on, select Go. If necessary, enter a division factor (For example, 1000) in Divide Metric Value by prompt at top of the page and select Go.

14 Distribution
This report describes basic statistical discrete distribution views of a selected population. It lets the user dynamically define the number of buckets to use for statistical distribution, as well as the grain in the population, and provides several dynamic representations of the results.

14 Distribution Help
1.1 Purpose: This report describes basic statistical discrete distribution views of a selected population. It lets the user dynamically define the number of buckets to use for statistical distribution, as well as the grain in the population, and provides several dynamic representations in the results. This report is useful to understand how the individuals of a population are distributed between the minimum and the maximum values, and to suggest the probabilities of where an individual may fall in a specific bucket. The report applies to numerous business situations: distribution of order values, distribution of call times, distribution of salaries, and so on. It allows visualizing skewness of a given population versus typical distribution.

1.2 Usage:

From drop down list of choices at the top of page, select a metric to analyze from "Select
Metric" prompt and select a dimension from "Select Dimension" prompt and select Go. Set the number of bins to desired value on the prompt at the top of the page and hit "Go" button. If necessary, enter a division factor (For example, 1000) in "Divide Metric Value by" prompt at top of the page and select Go. Filter the scope of the report as required by using "Page Filter Prompts" at bottom of page. To view tables with detailed data, select the option from "Select information to display :" drop down.

14 Distribution : Alternate Views


Detailed Table view Shows drillable detailed individual values by bin

15 Comparative Distribution
This report provides a comparative representation of simple statistical distributions, by individuals, for a selected population. It allows user to see how a metric comparatively distributes between different categories. The report lets the user dynamically define the number of buckets to use for statistical distribution, as well as the grain of the population.

15 Comparative Distribution Help


1.1 Purpose: This page provides a comparative representation of simple statistical distributions, by individuals, for a selected

population. It allows you to see how a metric comparatively


distributes between different categories. The report lets the user dynamically define the number of buckets to use for statistical distribution, as well as the grain of the population. This report highlights how the spread of metric values distribution changes from one value of a dimension to

another, for example, from one year to another.


This is done by displaying the structural distribution changes in the population of events. For example, the evolution of mix of order values from large to small, change from one region to another for the distribution of call durations, salaries, and so on. 1.2 Usage: From drop down list of choices at top of page, select two dimensions for a comparative analysis. Dimension 1 defines the comparative individuals, how many rows of charts in the page, and dimension 2 defines the grain for each distribution. Use an aggregated column for dimension 1 to avoid having too many charts on the page and a granular column for dimension 2. Distribution analysis requires a dataset with multiple data points to render meaningful charts. From drop down list of choices at top of page, select the to run distribution analysis on, select Go. From Set # of Bins drop down, enter the number of buckets to include in distribution charts. This defines the grain of the intervals in the distribution analysis. A value of 10 means that the chart is splitting individuals into 10 buckets of equal range between minimum and maximum value of the population (# of columns in the bar chart). Select Go.

16 Scatter
This report provides a graphical summary of a set of data. Individual values are represented by the position of the point in the chart space. It displays measures of central median, dispersion and skewness. It also identifies top, bottom values and interquartile range (IQR).

16 Scatter Help
1.1 Purpose: This report provides a graphical summary of a set of data. Individual values are represented by the position of the point in the chart space. It displays measures of central median, dispersion and skewness. It also identifies top, bottom values and interquartile range (IQR). Scatter charts are typically used to compare distinct

values across categories and visualize metric values for


individuals in a selected dimension. It also provides an understanding of how dispersed those values are and where extreme values compare with the rest of the population.

1.2 Usage: From drop down list of choices at top of page, select a metric to analyze from "Select Metric" prompt and select a dimension from "Select Dimension" prompt and select Go. Filter the scope of the report as required by using Page Filter Prompts at bottom of page.

17 Variability
This report shows how an initial value is increased and decreased by a series of intermediate values and, details of dimension individuals for each intermediate value. It displays the top and bottom Ntiles of the selected dimension individuals by selected metric.

17 Variability Help
1.1 Purpose: This report shows how an initial value is increased and decreased by a series of intermediate values and, details of dimension individuals for each intermediate value. It displays the top and bottom Ntiles of the selected dimension individuals by selected metric. This analysis is helpful in understanding the distribution of metric value for individuals in selected dimension.

1.2 Usage: From drop down list of choices at top of page, select a metric to analyze from "Select Metric" prompt and select a dimension from "Select Dimension" prompt and select Go. Set the number of Percentiles to display and hit "Go" button. Filter the scope of the report as required by using "Page Filter Prompts" at bottom of page.

18 Standard Deviation
This report provides a comparative analysis of standard deviation on a metric, for a selected set of dimensions. For each individual of dimension 1, reports displays standard deviation of metric values along dimension 2 (grain). It also identifies individuals with highest variability and highest standard deviation.

18 Standard Deviation Help


1.1 Purpose: This report provides a comparative analysis of standard deviation on a metric, for a selected set of dimensions. For each

individual of dimension 1, measurement of standard deviation of


metric values along dimension 2 (grain) will be displayed. In addition, the report also presents comparison of average and median values, and identifies the dimension 1 individual with highest variability and highest standard deviation. This analysis is useful in identifying extent of variation in metric values in dimension 2 individuals and to compare that between different dimension 1 individuals. For example, what is the variability in customer revenue for brand X versus brand Y ? You can gain an understanding of overall volatility of business metrics, and identify which individuals have the highest variability (volatility) in their businesses. Variability can correlate with a level of risk in the business, or indicate potential area for processes optimization. For example, the report can help understand which individual has the highest diversity in the size of its orders, which accounts have the highest variation in their balances over months... and help identify candidates for improvement Variability indicator is a percentage of standard deviation over average value for the population. It's expressed as a percentage. Variability = 100% means that Std Dev = Avg value of population. 1.2 Usage: From drop down list of choices at top of page, select two dimensions from "Select Dimension 1" and "Select Dimension 2" prompts. Dimension 1 is represented in X axis for the charts, Dimension 2 is the grain detail upon which Standard Deviation is calculated for each of the X axis individuals. Select the metric to analyze from "Select Metric 1" prompt and select Go.

19 Statistical Process Control (SPC)


This report displays a configurable Control Chart analysis (also known as the Shewhart chart or process-behavior chart). This report is used to determine whether a business process is in a state of statistical control or not. The chart identifies most homogenous periods of data over time and indicates if the process is within control (between Lower and upper control limit) or not. If the chart indicates that the process is not in control, the pattern it reveals can help determine the source of variation to be eliminated to bring the process back into control.

19 SPC Help
1.1 Purpose: This report displays a configurable template version of a Control Chart. The control chart, also known as the Shewhart chart or process-behavior chart, in statistical process control is a tool used to determine whether a manufacturing or business process is in a state of statistical control or not. If the chart indicates that the process is currently under control then it can be used with confidence to predict the future performance of the process. If the chart indicates that the process being monitored is not in control, the pattern it reveals can help determine the source of variation to be eliminated to bring the process back into control. On a practical level the control chart can be seen as part of an objective disciplined approach that facilitates the decision as to whether process performance warrants attention or not. The control chart is one of the seven basic tools of quality control. This analysis automatically groups data points into user selected number of clusters and then highlights those data points that fall outside the control limits. Control Limits are determined by user selected value for control band (no of sigmas). For each cluster, a control zone marked with their UCL(Upper Control Limit), Average, and Lower Control Limit (LCL) (Lower Control Limit). The boundaries of clusters are also shown by break points.

1.2 Usage: From drop down list of choices at the top of page, select a metric to analyze from "Select Metric" prompt and select a dimension from "Select Dimension" prompt and select Go. If required, user may also adjust Control zone bandwidth (expressed in # of standard deviations). The number of sigmas expressed will increase the control band above and beyond the average line. User may adjust No of clusters : the number of clusters will define how many breaking points will be inferred from the detailed data.

20 Deviants
This report displays how metric values vary over time and highlights top sequences of consecutive deviant data points. Consecutive deviants

are two or more consecutive data points which are above and below control zone (where metric value is outside control limits of a Control
Chart). The control chart is a tool used to determine whether a manufacturing or business process is in a state of statistical control or not.

20 Deviants Help
1.1 Purpose: This report displays how metric values vary over time and highlights top sequences of consecutive deviant data points. Consecutive deviants are two or more consecutive data points which are above and below control zone (where metric value is outside control limits of a Control Chart). The control chart, also known as the Shewhart chart or process-behavior chart, in statistical process control is a tool used to determine whether a manufacturing or business process is in a state of statistical control or not. Consecutive deviants in a control chart indicates that the process being monitored is not in control, the pattern it reveals can help determine the source of variation to be eliminated to bring the process back into control. On a practical level the control chart and deviants can be seen as part of an objective disciplined approach that facilitates the decision as to whether process performance warrants attention or not. The control chart is one of the seven basic tools of quality control. 1.2 Usage: From drop down list of choices at the top of page, select a metric to analyze from "Select Metric" prompt and select a dimension from "Select Dimension" prompt and select Go. If required, user may also adjust # of top sequences by setting the prompts at the top of the page. Increasing # of top sequences value will let the chart identify more of the top sequences. # of top sequences set to 1 means that only the longest sequence of deviants (the one with the most individuals) will be marked. # of top sequences set to 3 means that only the top 3 longest sequences of deviants (the ones with the top 3 most individuals) will be marked (that could result in more than 3 sequences). If required, user may also adjust Control zone bandwidth (expressed in # of standard deviations). The number of sigmas expressed will increase the control band above and beyond the average line. For more detailed analysis on any particular set of consecutive deviants, user may select corresponding down arrow icon in third column on table beside the bubble chart Filter the scope of the report as required by using "Page Filter Prompts" at bottom of page.

2.B Dashboard Details


04 Dimensional Analysis
Dimensional analysis layouts and techniques, and other detailed reports

21 Indexing
This report provides a comparison of several dimension values over a time period using indexed line charts, as opposed to absolute value line charts. The analysis turns absolute values into indexes and makes comparison between trended values intuitive. It allows users to select a value from the X axis of "Actual Values" chart, for example, use Month, as the index basis point. Indexed information allows users to compare the pattern of evolution of values in a more insightful manner than when using absolute values. Regardless of how far apart the absolute values may be from one another, indexes allow them to be represented in a framed format with comparisons making more visual sense.

21 Indexing Help
1.1 Purpose: This report provides a comparison of several dimension values over a time period using indexed line charts, as opposed to absolute value line charts,

This analysis turns absolute values into indexes and makes


comparison between trended values intuitive. It allows users to select a value from the X axis of "Actual Values" chart, for example, use Month, as the index basis point. Indexed information allows users to compare the pattern of evolution of values in a more insightful manner than when using absolute values. Regardless of how far apart the absolute values may be from one another, indexes allow them to be represented in a framed format with comparisons making more visual sense. 1.2 Usage: From drop down list of choices at top of page, select a metric to analyze from Select Metric prompt and select a dimension from Select Dimension prompt and select Go. This dimension is displayed as lines of different colors on the chart. From the top of the page, select Dimension 2 which is to be used for Indexed charting. The values from this dimension form the distinct values of X axis for the charts. Select Go From the top of the page, select a value for index base and select Go . The selected value defines the starting base for the index (value 100). Filter the scope of the report as required by using Page Filter Prompts at bottom of page. Use the drop down in the report to switch tabular display between indexed metrics and actual values

22 Waterfall
This dashboard shows how individuals in a selected dimension contribute to total value of a selected metric, cumulatively. The report uses a

waterfall chart model to show individual's metric values in the form of floating columns leading up to total value of the metric. Height of the
column is proportional to value of metric for that individual.

22 Waterfall Help
1.1 Purpose: This report shows how individuals in a selected dimension contribute to total value of a selected metric, cumulatively. The report uses a waterfall chart model to show individual's metric values in the form of floating columns leading up to total value of the metric. Height of the column is proportional to value of metric for that individual. This analysis can be used to visually compare different individual's contribution towards the total value of metric and

immediately appreciate the gaps between contributions.


It can also be used to visually explain what happened or in the case of a forecast - what may happen. For example, to view how the revenue fluctuated over the months and which month

contributed the most to the total revenue.

1.2 Usage: From drop down list of choices at top of page, select a metric to analyze from Select Metric prompt and select a dimension from Select Dimension prompt and select Go. The individuals of the dimension selected represent the different bars in the bar chart. It is better usage to select

aggregate dimensions in this chart, as opposed to very granular dimensions.


If necessary, enter a division factor (For example, 1000) in Divide Metric Value by prompt at top of the page and select Go. Filter the scope of the report as required by using Page Filter Prompts at bottom of page. Click on the dimension individuals in the table to navigate to more detailed report.

23 Waterfall Multidim
This dashboard shows how individuals of two selected dimensions contribute to total value of a selected metric, cumulatively. The report uses

a waterfall chart model showing metric values in the form of floating columns leading up to total value of the metric. X axis indicates individuals
from one dimension, while height of the bars in each column is proportional to value of metric for individuals in from the other dimension.

23 Waterfall Multidim Help


1.1 Purpose: This report shows how individuals of two selected dimensions contribute to total value of a selected metric, cumulatively. The report uses a waterfall chart model showing metric values in the form of floating columns leading up to total value of the metric. X axis indicates individuals from one dimension, while height of the bars in each column is proportional to value of metric for individuals in from the other dimension. This analysis can be used to visually compare different individuals contribution towards the total value of metric and immediately appreciate the gaps between contributions. It can also be used to visually explain what happened or in the case of a forecast - what may happen. For

example, to view how the regional revenue fluctuated over the months and which month/region
contributed the most to the total revenue.

1.2 Usage: From drop down list of choices at top of page, select a metric to analyze from Select Metric prompt and select the dimensions from Select Dimension 1 and Select Dimension 2 prompt

and select Go. The individuals of the Dimension 2 selected represent the different bars in the
bar chart. Within each bar, the distribution by individuals in Dimension 1 is shown by different colors. It is better usage to select aggregate dimensions in this chart, as opposed to very granular dimensions. Filter the scope of the report as required by using Page Filter Prompts at bottom of page.

24 Benchmark
This dashboard shows comparative performance of two selected metrics for individuals on a dimension, benchmarked dynamically against a user selected individual from the dimension. The report allows users to compare performance on metrics between individuals of a dimension. It presents intuitive view of how individuals perform relative to each other, and relative to a selected individual elected as "benchmark" value. The chart easily points out inconsistent performance gaps.

24 Benchmark Help
1.1 Purpose: This report shows comparative performance of two selected metrics for individuals on a dimension, benchmarked dynamically against a user selected individual from the dimension. The report allows users to quickly compare performance on metrics between individuals of a dimension. It presents clear and intuitive view of how individuals perform relative to each other, and relative to a selected individual elected as "benchmark" value. The chart easily points out inconsistent performance gaps. For example, setting the bench on region A, we can quickly visualize which regions are performing above A both for revenue and margin, or which one is better than A in revenue but worse in margin, and so on.

1.2 Usage: From drop down list of choices at top of page, select two metrics to analyze from Select Metric 1 and Select Metric 2 prompts and select Go. From the top of the page, select Dimension 1 and select Go . From the top of the page, select the bench value in the Set Bench Value drop down and select Go . The value selected in this drop down is the one set as bench (base 100 for indexes), to which other individuals are indexed and compared. Filter the scope of the report as required by using Page Filter Prompts at bottom of page

25 Index to Avg
This dashboard shows comparative performance for individuals on a dimension, for several metrics, indexed against the average performance of all the individuals in the report. The report compares performance on several metrics. Each individual of a dimension and the average of the group. It provides a visual view of how individuals perform and to highlight inconsistent performance gaps.

25 Index to Avg Help


1.1 Purpose: This report shows comparative performance for individuals on a dimension, for several metrics, indexed against the average performance of all the individuals in the report. The report compares performance on several metrics. Each individual of a dimension and the average of the group. It provides a visual view of how individuals perform and to highlight inconsistent performance gaps: you can quickly visualize which regions are performing above average both for revenue and margin, or which one is better than average in revenue but worse in margin, and so on. 1.2 Usage: From drop down list of choices at top of page, select a dimension and two metrics to analyze from., then select Go. The Dimension object defines the bars of the bar charts, and the metrics are used to calculate the indexing positions on.

Filter the scope of the report as required by using Page Filter Prompts at bottom of page. Clicking on the blue colored numbers within tables grants you access to more detailed navigation reports.. .

26 Index to Avg Trended


This dashboard shows trended comparative performance for individuals from a dimension, benchmarked against Average of individuals in this

dimension. The report compares each individual to average for two distinct metrics, over time. It displays relative performance to a monthly
average base index (100 every month), as well as complementary presentation that factors in evolution of average value over time.

26 Index to Avg Trended Help


1.1 Purpose: This report shows trended comparative performance for individuals from a dimension, benchmarked against Average of individuals in this dimension. The report compares each individual to average for two distinct

metrics, over time. It displays relative performance to a


monthly average base index (100 every month), as well as complementary presentation that factors in evolution of average value over time.

1.2 Usage: From drop down list of choices at top of page, select two metrics to analyze from Select Metric 1 and Select Metric 2 prompts and select Go.

From the top of the page, select Dimension 1 and Dimension 2 and select Go . Filter the scope of the report as required by using Page Filter Prompts at bottom of page.

26 Index to Avg : Alternate Views


Detailed Table view Shows detailed view of monthly values by dimension individuals

27 Profile
This dashboard profiles the selected population by the selected attributes. It provides a breakdown of user selected metric 1 and metric 2 for

the individuals within selected dimensions. The line bar chart provides a visual comparison of metric 2 values versus corresponding metric 1
value for each dimension individual.

27 Profile Help
1.1 Purpose: This report profiles the selected population by the selected attributes. It provides a breakdown of user selected metric 1 and metric 2 for the individuals within selected dimensions. The line bar chart provides a visual comparison of metric 2

values versus corresponding metric 1 value for each


dimension individual. This analysis is very useful in comparing and understanding how the metric values performed for different business attributes. For example, how are revenue and quantity shipped spread across different brands, geographic locations, customer categories and markets? 1.2 Usage: From drop down list of choices at top of page, select two metrics to analyze from Select Metric 1 and Select Metric 2 prompts and select Go. Within the report, select Slice by dimensions from the drop down provided.

If necessary, enter a division factor (For example, 1000) in Divide Metric 1 by and/or Divide
Metric 2 by prompts at top of the page and select Go. Filter the scope of the report as required by using Page Filter Prompts at bottom of page.

28 Details
This report shows detailed attributes for a selected population.

2.C Answers features & Segment example


Overview of Answers features used with Sample Sales

Selected Sample Sales Answers Features

1/2

Page help content :

click on help
Multi Navigation : click on hyperlinked figures and select which detail to navigate to. Repeat process and further navigate deeper into details Web Variables leveraging : Change values in top pages pink boxes, and hyperlinks to see contextual functional help on dashboard you are looking at

see how reports queries change


accordingly

Dynamic selection of Metric and dimensions in the dashboards. Pivot table level calculations to extend aggregations levels on top of answers columns calculations A single report offers multiple functional analysis combinations.

Segmentation metadata setup examples (from detailed reports (navigation targets), leverage table dynamic sorting and direct segment/list creation link.

Selected Sample Sales Answers Features

2/2

Conditional Chart series formatting


based on value of series, to allow better visual rendering in charts. Leverage of Filter Groups structure to allow advanced filtering in reports, as well as leverage of presaved prompted filters

Cascading Presentation variables dependencies, leveraging strings

datatypes

Answers level Aggregations : visit definition of answers based metrics with SQL based Aggregations formulas on top of existing RPD objects

Union clause based answers report and charts, that leverage capability of bringing together results of several distinct queries

Range drilling : navigate from a

range of values into a detail report


for this range of values.

Sample Customer Segment 1


Customer Waterfall Segment Sample : counts out all customers in specific criteria showcasing Segmentation capability, profile dashboards navigation setup enabled, list export format configured.

3.A Repository Overview


Overview of features and sample techniques included in the Sample repository

Physical Source Overview


Facts Set of 11 xml independent files, (total size 5 Megs) 3 Fact tables :
Main Fact : FactsRev, Order detail, 5000 records and two years data history Additional Facts : Forecast : 229 records, Inventory : 7200 records

6 Dimension tables
Orders : 5000 records Customers : 240 records (with snowflake attribute for Segments) Employees : 15 records Products : 15 records Market : 15 records Time : Day details and month details, <1000 records. Time dimension dynamically sourced from two tables depending on context of analysis.

Metadata : Physical Layer Overview


Facts Source 1 Dimensions Two distinct unrelated physical sources logically joined in business model
(no cross database joins in physical layer)

19 aliased tables including specific constructs :


Canonical Time (A11, A13) Rolling time construct (A22, A24, A20, A21, A31)

Source 2

Fully Aliased sources best practice

Rolling Months Aliased Sources

Metadata : Logical Layer Overview


Facts
Single logical layer federating the two distinct physical sources Several logical constructs to showcase aggregation and calculations capabilities :
Time series, Rolling months, Canonical time construct and Time Facts table Logical derived fact tables

Numbered objects best practice architecture including metadata descriptions fully populated

Facts

Logical Derived Facts

Metrics Aggregations
Numerous logical RPD aggregations examples
Time series, Rolling Months, Variations, Run rates, Percent of Periods, Grain Fragmentation .

F1, Base Aggregations : Basic aggregations (Sum, Avg, Counts). F2, Time Series Measures Aggregations on Time Dimension. Variants of Period Ago and Period-To-Date metrics like MonthAgo, QuarterAgo, YearAgo, Month-To-Date, Quarter-To-Date, Year-To-Date etc. F3, Rolling Time Series Measures Rolling Aggregations on Time. Contains Rolling 3 Months and Rolling 6 Months measures with different aggregations like Sum, Daily Avg, Monthly Avg etc. This LTS is physically mapped to the Revenue and Inventory fact tables, but the fact tables here join to the time dimension table using a complex join F4, Time Variations calculations : variance with period ago values like Variance to Month Ago, Variance to Year Ago, Quarter-To-Date to Quarter Ago's Quarter-To-Date variance etc.. All metrics in this table are logical calculations based on physically mapped metrics from Base Measures logical tables F5, Level Based Measures: aggregations that always return value at a particular level within a dimension. E.g.:- Revenue at Month level, Billed Quarterly at Year level, Booked Amt for all Products etc. F6, Other Calculations : specific logical aggregations such as Runrates and Seasonality metrics, grain fragmentation measures.

Metadata : Presentation Layer Overview

Presentation Layer organizing all objects into Root Folders and Subfolders Fully fledged version of presentation layer with all logical objects, and example of reduced set presentation layer with selected renamed objects User objects permission restriction : these two users (demo / demo2) have distinct accesses to presentation folders

Metrics Aggregation List

Other Selected Model Best Practices

Multiple Physical Sourcing (without cross db joins) Dimensional Snow flaking Degenerated Facts Attributes Dimension based aggregations Canonical Time Construct (Multi LTS Facts tables) and Time Facts folder Objects Numbering and Derived objects fact tables Multiple Hierarchy paths and Hierarchy drill chaining Segmentation Metadata Physical Layer Aliasing, Pres Layer Layout User level Data Constrain Projects Metadata dictionary

3.B How To Demo


Selected Repository Features
How to show and visit some of the RPD features illustrated in Sample Sales

Selected How to Demo scenarios: Selected Repository Features

Multiple Physical Sourcing

Dimensional Snow Flaking

Hierarchy drill Chaining

Dimensions Based Aggregations

Metadata Dictionary

User Level Data Constraint

Segmentation Metadata

Runrate Logical Aggregations

How to Demo : Multiple Physical Sourcing

1/2

1. Define answers query as shown to the right : Or, alternatively, paste following SQL into SQL Issued under Advanced tab and press

SELECT "D0 Time"."T02 Per Name Month", "F1 Revenue"."101 Revenue (Sum All)", "14 Other Measures"."6-01 Revenue Fcst (Sum All)" FROM "Sample Sales" WHERE "D0 Time"."T02 Per Name Month" BETWEEN '2008 / 01' AND '2008 / 12'

2. Notice that results are showing aligned data for revenue and forecast information

How to Demo : Multiple Physical Sourcing

2/2

3. Show that, in RPD, both revenue and forecast data are sourced from distinct physical tables that have no relationships to each other

4. Explain that, in Logical Model, both logical facts table are defined with same grain (content level tab), and logical dimension tables are each sourced from both physical models (Forecast Time and Actual Time)

How to Demo : Dimensional Snow flaking


1. Define answers query as shown to the right : Or, alternatively, paste following SQL into SQL Issued under Advanced tab and press
SELECT "D1 Customer"."C5 Segment, "F1 Revenue"."1-01 Revenue (Sum All)" FROM "Sample Sales"

2. Results are showing revenue broken down by customer segments

3. Show customer segment physical table joined to Customer dimension as snowflake structure

4. Open Logical Table Source for Customer logical dimension and show that is tied to both Customer and Segment physical sources

How to Demo : Hierarchy drill chaining


1. Define answers query as shown to the right : Or, alternatively, paste following SQL into SQL Issued under Advanced tab and press
SELECT "D4 Product"."P01 Product", "F1 Revenue"."1-01 Revenue (Sum All)" FROM "Sample Sales"

1/3

2. Results show revenue broken down by Products

3. Click on any product and see jump from Product level in product hierarchy, to Region Level in Market dimensional hierarchy

4. Then, drilling down the whole Market hierarchy to jump to Manager level in Employee dimension

5. Then from employee hierarchy, jumping to Customer, then orders details

How to Demo : Hierarchy drill chaining

2/3

6. Open RPD and visit properties of Detailed level in Product Hierarchy. See setup in tab Preferred Drill Path, where Total level for Market hierarchy is defined as next drill.

7. Similarly, visit properties of Detailed level in Market Hierarchy. See setup in tab Preferred Drill Path : Total level for Employee hierarchy is defined as next drill.

How to Demo : Hierarchy drill chaining

3/3

8. Full detail of how all dimensions are tied in Sample Sales applications content.

How to Demo : User level Data Constrain

1/2

1. Login as demo/demo, open Sample Sales Reduced subject area in answer. Notice you have access to multiple detail folders, including Customer and Fact Others folders. 2. Define answers query as shown to the right : Or, alternatively, paste following SQL into SQL Issued under Advanced tab and press
SELECT "Other Dimensions".Region, "Other Dimensions".District, "Other Dimensions".Area, "Other Dimensions".Market, "Facts Revenue".Revenue FROM "Sample Sales Reduced"

3. See results showing data for several of Market records

4. Save the report (or copy above SQL in clipboard), logoff and login back as demo2/demo2 user. Open Sample Sales Reduced subject area. Notice that you have access to less folders than when logged in as demo/demo

How to Demo : User level Data Constrain


1. Define answers query as shown to the right : Or, alternatively, paste following SQL into SQL Issued under Advanced tab and press
SELECT "Other Dimensions".Region, "Other Dimensions".District, "Other Dimensions".Area, "Other Dimensions".Market, "Facts Revenue".Revenue FROM "Sample Sales Reduced"

2/2

6. See results showing much more restricted visibility into Market records than with user demo/demo.

7. Open RPD, visit Security Manager (>Manage>Security) and click on demo2 user. Then click on button and visit tab Filters. There see definition of visibility restriction clauses. These clauses can be composed of dynamically set session variables and will apply to all queries for this user.

How to Demo : Segmentation Metadata


1. Click on More Products hyperlink at top of screen and select Marketing from drop down menu. Then click on Create a Segment and select Customers Target Level

1/2

2. Use answers like object on left pane to build criterias as shown on left : -Customers with high revenue (>1000), - who had a sharp mth to mth decrease in revenue (>50% drop), - who do not own product 1,2,or 3, 3. Set Counts to be All Counts, then hit button Update Counts to see results of your counting for distinct customers. 4. Click on the total # of customers at the bottom (hyperlink) to show direct navigation / filtering of dashboards with this list

How to Demo : Segmentation Metadata

2/2

5. Open RPD, visit Marketing Manager (>Manage>Marketing). See Target Levels objects defined. Click on Customers Target Level to visit Segmentation catalog defined, and Qualified list item (i.e. what system is counting) mapped to Customer Key object from Customer dimension.

How to Demo : Metadata Dictionary


1. Open RPD, navigate to >Tools>Utilities and select option Generate Metadata Dictionary 2. Copy resulting folder structure in a new folder : metadata_dictionary\Samplesales under existing path : \OracleBI\oc4j_bi\j2ee\home\applic ations\analytics\analytics\

3. Update Instanceconfig.xml (located in \OracleBIData\web\config\) and extend with following tags before the end of the file :
<SubjectAreaMetadata> <DictionaryURLPrefix>/analytics/metadata_dictionary/</DictionaryURLPrefix> </SubjectAreaMetadata>

4. Restart OBI Web server and OC4J, the Metadata dictionary button should now show up in answers with descriptions of all metdata RPD lineage

How to Demo : Metadata Dictionary

5. Repository object descriptions and full mapping paths available on click from answers interface

How to Demo : Runrate logical aggregations


1. Create folllowing answers query : (paste following SQL into SQL Issued under Advanced tab and press

1/2

SELECT "F1 Revenue"."1-01 Revenue (Sum All)" , "D02 Time Facts"."T62 # of Days", "D02 Time Facts"."T63 # of Weeks", "D02 Time Facts"."T64 # of Months", "14 Other Measures"."1-81 Revenue (Wkly RunRate)", "14 Other Measures"."1-82 Revenue (Mthly RunRate)", "14 Other Measures"."1-83 Revenue (Qtrly RunRate)", "14 Other Measures"."1-84 Revenue (Yrly RunRate)" FROM "Sample Sales" WHERE "D0 Time"."T00 Calendar Date" BETWEEN date '2008-01-01' AND date '2008-02-19'

2. Filter query to a custom time range spanning over a few weeks, as shown

How to Demo : Runrate logical aggregations

2/2

3. Understand the single line results :

Sum of Revenue for the period selected

Average revenue amount per week over the period selected : = (revenue) / (number of weeks) Average revenue amount per month over the period selected : = (revenue) / (number of months) Average revenue amount per quarter over the period selected : = (revenue) / (number of Quarters) Average revenue amount per year over the period selected : = (revenue) / (number of Years)

Calculation of number of days, weeks, months in the period selected. These calculations are decimals, not tied to period selection be over a full calendar month. Note that these objects are shown in the query for explanation, they are not needed in queries using runrates aggregations.

How to Demo : Dimensions Based aggregations


1. Create folllowing answers query : (paste following SQL into SQL Issued under Advanced tab and press

1/2

SELECT "D0 Time"."T01 Per Name Week, "F1 Revenue"."1-01 Revenue (Sum All)", "12 Variations"."1-43 Revenue (Fst oT)", "D02 Time Facts"."T60 First Time Day Dt" FROM "Sample Sales" WHERE "D0 Time"."T01 Per Name Week" IN ('2007 Week 10', '2007 Week 11')

2. Visit Results : notice value of column 1-43 and T60 columns, to understand what this value is, drill on a specific week for instance For a given time object present in the query, 1-43 aggregation returns the FIRST value of the metricin this time object. If the object is a Week, it returns the value for the first day in the week only.

How to Demo : Dimensions Based aggregations


3. To show the setup for this, open RPD, and visit properties of object 1-43, tab Aggregation 4. The Aggregation there is setup as dependant on what dimension is invoked. When time dimension is in the query, the aggregation against it will be First, while aggregation with any other dimensions will be sum.

2/2

3.C Repository Aggregations Detail


Functional details of RPD aggregations
Detail of all logical aggregations included in Sample repository

Metrics Aggregation List

Metrics Aggregation Details : Base & Agos


01 03 Sum All Count Distinct Returns sum of data over any selected dimensions Counts the number of distinct occurrences of a value

04

Month Ago (Mago) 2 Months Ago (2Mago) 3 Months Ago (3Mago)

Returns data for same period as of previous month vs. month of date selected in the query. Example, Mago of revenue metric as of May 19th returns revenue value for Apr 19th of same year. Returns data for same period as of two months before date selected in the query. Example : 2Mago of revenue metric as of May 19th returns revenue value for Mar 19th of same year. Returns data for same period as of three months before date selected in the query. Example : 3Mago of revenue metric as of May 19th returns revenue value for Feb 19th of same year.

05

06

Note : aggregations 04,05 and 06 will return at least month level detail. Ie, if no time object in the report, or if only objects higher than month, then the query will force return one row per month. Incase there are time objects in the reports at month, week or day level, then it will break down the results to this level of detail

07

Quarter Ago (Qago)

Returns data for same period as of previous quarter vs. quarter of date selected in the query. Example, Qago of revenue metric as of May 19th returns revenue value for Feb 19th of same year.
Returns data for same period as of previous Year vs. year of date selected in the query. Example, Yago of revenue metric as of May 19th of year N returns revenue value for May 19th of year N-1 Returns data for same day as of previous week vs. week of date selected in the query. Example, Wago of revenue metric as of May 19th returns revenue value for May 12th

08

Year Ago (Yago) Week Ago (Wago)

09

Note : aggregations 07,08 and 08 will return at least respectively Qtr,Year or Week level month level detail. Ie, if no time object in the report, or if time objects higher than these (respectively) month, then the query will force return one row per each level respectively.

Metrics Aggregation Details : To Dates


10 11 Month To Date Sum Month To Date Daily Avg Month Ago Month To Date Month Ago Year To Date Sum Quarter To Date Sum Quarter To Date Monthly Avg Quarter To Date Daily Avg Quarter To Date Weekly Avg Quarter Ago Quarter To Date Sum Year To Date Sum Year To Date Monthly Avg Year Ago Year To Date Sum Sums all the values from the first day of the month to the selected day in the query Sum of all the values from the first day of the month to the selected day in the query divided by number of days from the start of the month to the current day Month Ago Month-To-Date Value for the same day in past month. For the day in the previous month as the day selected in the query, it returns the sum of all the values from the first day of the month to that period. Month Ago Year-To-Date value for the same day in past month. For the same day in the previous month as the day selected in the query, it returns the sum of all the values from the first day of the current year to that day in the previous month. Sums all the values from the first day of the quarter to the day selected in the query Sum of all the values from the first day of the quarter to the day selected in the query divided by number of months (decimal) from the start of the year to the current period Sum of all the values from the first day of the quarter to the current period divided by number of days from the start of the quarter to the current period Sum of all the values from the first day of the quarter to the current period divided by number of weeks from the start of the quarter to the current period Quarter Ago Quarter-To-Date Value for the same period. For the same period in the previous quarter, it returns the sum of all the values from the first day of the quarter to that period. Sums all the values from the first day of the year to the current period Sums of all the values from the first day of the year to the current period divided by number of months from the start of the year to the current period Year Ago Year-To-Date Value for the same period. For the same period in the previous year, it returns the sum of all the values from the first day of the year to that period.

12

13

14

15

16

17

18
19 20

21

10. Month to Date Sum (MTD)


Definition :
Sums all the values from the first day of the month to the selected day in the query. Exp : MTD Sum of revenue for May 19th of Year N returns sum of revenue for May 1st to May 19th included
Year N - 1
Q1 Q2 Q3 Q4 Q1 Q2

Year N
Q3 Q4

Jan Feb Mar

Apr

May

Jun

July Aug Sep Oct Nov Dec Jan Feb Mar

Apr

May Sum

Jun

July Aug Sep Oct Nov Dec

Use When :
Use this object when, for a given date, you need to see the sum of the metric since the beginning of the month that this date belongs to. This aggregation is useful when performance measurement cycles are at month level to compare metrics from one period to another one. I.e., how were we doing last month or last year at the same day in month, compared to this period.

Logical Formula :
TODATE(Base Metric, "Sample Sales"."H0 Time"."Month)

Limitations :
This aggregation will return at least month level detail. I.e., if no time object in the report, or if only objects higher than month, then the query will force return one row per month. Incase there are time objects in the reports at month, week or day level, then it will break down the results to this level of detail.

11. Month To Date Daily Avg


Definition :
Sum of all the values from the first day of the month to the selected day in the query divided by number of days from the start of the month to the current day Exp : MTD Avg of revenue for May 19th of Year N returns daily Avg of revenue from May 1st to May 19th included
Year N - 1
Q1 Q2 Q3 Q4 Q1 Q2

Year N
Q3 Q4

Jan Feb Mar

Apr

May

Jun

July Aug Sep Oct Nov Dec Jan Feb Mar

Apr

May Avg

Jun

July Aug Sep Oct Nov Dec

Use When :
Use this object to view how average daily value for the whole month is stabilizing as month completes. At any given date in the month, the daily average of all past days in month will be returned. This aggregation is useful when performance measurement cycles are at month level to compare metrics from one period to another one. Ie, how were we doing last month or last year at the same day in month, compared to this period.

Logical Formula :
10 MTD Sum" /"FT1 Time Facts"."T72 MTD Distinct Days T72 is a time fact object that calculates the number of distinct days since the beginning of the month for any date.

Limitations :
This aggregation will return at least month level detail. Ie, if no time object in the report, or if only objects higher than month, then the query will force return one row per month. Incase there are time objects in the reports at month, week or day level, then it will break down results to this level of detail

12. Month Ago Month To Date (Mago MTD)


Definition :
For the same period in the previous month, it returns the sum of all the values from the first day of the month to that period to the same day that is selected in the query, but for last month

Exp : Mago MTD of revenue for May 19th of Year N returns Sum of revenue for Apr 1st to Apr 19th
Year N - 1
Q1 Q2 Q3 Q4 Q1 Q2

Year N
Q3 Q4

Jan Feb Mar

Apr

May

Jun

July Aug Sep Oct Nov Dec Jan Feb Mar

Apr Sum

May

Jun

July Aug Sep Oct Nov Dec

Use When :
Use this aggregation to compare line to line what previous month value was with what current month value is, at any given day in the month. This time series allows to have both values in the same query table, as distinct columns, and hence allows to derive logical calculations between these two aggregated values (either in answers or in RPD)

Logical Formula :
TODATE(Month Ago Base Metric, "Sample Sales"."H0 Time"."Month)

Limitations :
This aggregation will return at least month level detail. Ie, if no time object in the report, or if only objects higher than month, then the query will force return one row per month. Incase there are time objects in the reports at month, week or day level, then it will break down results to this level of detail

13. Month Ago Year To Date Sum (Mago YTD)


Definition
Month Ago Year-To-Date value for the same period. For the same period in the previous month, it returns the sum of all the values from the first day of the year to that period. Exp : Mago YTD of revenue for May 19th of Year N returns Sum of revenue for Apr 1st to Apr 19 th
Year N - 1
Q1 Q2 Q3 Q4 Q1 Q2

Year N
Q3 Q4

Jan Feb Mar

Apr

May

Jun

July Aug Sep Oct Nov Dec Jan Feb Mar Sum

Apr

May

Jun

July Aug Sep Oct Nov Dec

Use When :
Comparing current month YTD to Mago YTD

Logical Formula
CASE WHEN Calendar Month is January 1 THEN AGO(Base Metric", "Sample Sales"."H0 Time"."Month", 1) ELSE 19. Base Metric YTD - Base Metric

Limitations :
Level Base = Month on time dimension.
This aggregation will Always return data detail at month level detail, no matter what time object are included in the report.

14. Quarter to Date Sum (QTD)


Definition :
Sums all the values from the first day of the quarter to the day selected in the query Exp : QTD Sum of revenue for May 19th of Year N returns sum of revenue from Apr 1st to May 19th included
Year N - 1
Q1 Q2 Q3 Q4 Q1 Q2

Year N
Q3 Q4

Jan Feb Mar

Apr

May

Jun

July Aug Sep Oct Nov Dec Jan Feb Mar

Apr Sum

May

Jun

July Aug Sep Oct Nov Dec

Use When :
Use this aggregation when, for a given date, you need to see the sum of the metric since the beginning of the quarter that this date belongs to. This aggregation is useful to track performance when reporting cycles are at quarter level, to compare from current quarter to another past quarter one, or to compare entities during current quarter. I.e., how were we doing last quarter at the same day in quarter, compared to this quarter.

Logical Formula :
TODATE(Base Metric, "Sample Sales"."H0 Time".Quarter)

Limitations :
This aggregation will return at least Quarter level detail. Ie, if no time object in the report, or if only objects higher than month, then the query will force return one row per quarter. Incase there are time objects in the reports at quarter, month, week or day level, then it will break down the results to this level of detail.

15. Quarter To Date Monthly Avg


Definition :
Sum of all the values from the first day of the quarter to the day selected in the query divided by number of months (decimal) from the start of the year to the current period Exp : QTD Monthly Avg of revenue for May 19th of Year N returns Monthly Avg of revenue from Apr 1st to May 19th included
Year N - 1
Q1 Q2 Q3 Q4 Q1 Q2

Year N
Q3 Q4

Jan Feb Mar

Apr

May

Jun

July Aug Sep Oct Nov Dec Jan Feb Mar

Apr Avg

May

Jun

July Aug Sep Oct Nov Dec

Use When :
Use this object to view how average Monthly value for the whole quarter is stabilizing as quarter completes. At any given date in the quarter, the monthly average of all past days in quarter will be returned. This aggregation is useful when performance measurement cycles are at quarter level to compare metrics from one period to another one. I.e., how were we doing last quarter or last year at the same day in quarter, compared to this period.

Logical Formula :
14 QTD Sum" /"FT1 Time Facts"."T74 QTD Distinct Months T74 is a time fact object that calculates the number of distinct months since the beginning of the Quarter for any date.

Limitations :
This aggregation will return at least Quarter level detail. Ie, if no time object in the report, or if only objects higher than month, then the query will force return one row per quarter. Incase there are time objects in the reports at quarter, month, week or day level, then it will break down results to this level of detail.

16. Quarter To Date Daily Avg 17. Quarter To Date Weekly Avg
16. Quarter To Date Daily Avg : Sum of all the values from the first day of the quarter to the selected day
divided by number of days from the start of the quarter to the selected Day Exp : QTD daily Avg of revenue for May 19th of Year N returns daily Avg of revenue from Apr 1st to May 19th included
Year N - 1
Q1 Q2 Q3 Q4 Q1 Q2

Year N
Q3 Q4

Jan Feb Mar

Apr

May

Jun

July Aug Sep Oct Nov Dec Jan Feb Mar

Apr Daily Avg

May

Jun

July Aug Sep Oct Nov Dec

Similar to : 15. Quarter To Date Monthly Avg, but with day denominator instead of month. The use
case and formulas structures similar to the ones for 15. Quarter To Date Monthly Avg.

17. Quarter To Date Weekly Avg : Sum of all the values from the first day of the quarter to the selected day
divided by number of weeks from the start of the quarter to the selected day Exp : QTD weekly Avg of revenue for May 19th of Year N returns weekly Avg of revenue from Apr 1st to May 19th included
Year N - 1
Q1 Q2 Q3 Q4 Q1 Q2

Year N
Q3 Q4

Jan Feb Mar

Apr

May

Jun

July Aug Sep Oct Nov Dec Jan Feb Mar

Apr Weekly Avg

May

Jun

July Aug Sep Oct Nov Dec

Similar to : 15. Quarter To Date Monthly Avg, but with week denominator instead of month. The
use case and formulas structures similar to the ones for 15. Quarter To Date Monthly Avg

18. Quarter Ago Quarter To Date Sum (Qago QTD)


Definition :
For the same period in the previous quarter, this aggregation returns the sum of all the values from the first day of the quarter to the same day in previous quarter as the selected day in the query. Exp : Qago QTD of revenue for May 19th of Year N returns Sum of revenue from Jan 1st to Feb 19 th
Year N - 1
Q1 Q2 Q3 Q4 Q1 Q2

Year N
Q3 Q4

Jan Feb Mar

Apr

May

Jun

July Aug Sep Oct Nov Dec Jan Feb Mar Sum

Apr

May

Jun

July Aug Sep Oct Nov Dec

Use When :
Use this aggregation to compare line to line what previous quarter value was with what current quarter value is, at any given day in the quarter. This time series allows to have both values in the same query table, as distinct columns, and hence allows to derive logical calculations between these two aggregated values (either in answers or in RPD)

Logical Formula :
TODATE(Quarter Ago Base Metric, "Sample Sales"."H0 Time"."quarter)

Limitations :
This aggregation will return at least quarter level detail. Ie, if no time object in the report, or if only objects higher than quarter, then the query will force return one row per quarter. Incase there are time objects in the reports at quarter, month, week or day level, then it will break down results to this level of detail

19. Year to date to Date Sum (YTD)


Definition :
Sums all the values from the first day of the year to the selected day in the query. Exp : YTD Sum of revenue for May 19th of Year N returns sum of revenue from Jan 1st of year N to May 19th included
Year N - 1
Q1 Q2 Q3 Q4 Q1 Q2

Year N
Q3 Q4

Jan Feb Mar

Apr

May

Jun

July Aug Sep Oct Nov Dec Jan Feb Mar Sum

Apr

May

Jun

July Aug Sep Oct Nov Dec

Use When :
Use this object when, for a given date, you need to see the sum of the metric since the beginning of the year that this date belongs to. This aggregation is useful to track how cumulative value of the metric is doing for the whole year. The cumulative view attenuates the impacts of weekly/monthly adjustments, and allows sharper business comparison to targets or budgets

Logical Formula :
TODATE(Base Metric, "Sample Sales"."H0 Time".Year)

20. Year To Date Monthly Avg 21. Year Ago Year To Date Sum
20. Year To Date Monthly Avg : Sums of all the values from the first day of the year to the current period
divided by number of months from the start of the year to the current period Exp : YTD Monthly Avg of revenue for May 19th of Year N returns Monthly Avg of revenue from Jan 1st to May 19th included of year N
Year N - 1
Q1 Q2 Q3 Q4 Q1 Q2

Year N
Q3 Q4

Jan Feb Mar

Apr

May

Jun

July Aug Sep Oct Nov Dec Jan Feb Mar

Apr

May

Jun

July Aug Sep Oct Nov Dec

Monthly Avg

Similar to : 15. Quarter to Date Monthly Avg, but with scope of year for numerator instead of
Quarter. The use case and formulas structures are similar.

21. Year Ago Year To Date Sum (Yago YTD) : For the same period in the previous year, it returns the
sum of all the values from the first day of the year to that period. Exp : Yago YTD Sum of revenue for May 19th of Year N returns Sum of revenue from Jan 1st to May 19th included of year N-1
Year N - 1
Q1 Q2 Q3 Q4 Q1 Q2

Year N
Q3 Q4

Jan Feb Mar Sum

Apr

May

Jun

July Aug Sep Oct Nov Dec Jan Feb Mar

Apr

May

Jun

July Aug Sep Oct Nov Dec

Similar to : 18. Quarter Ago Quarter To Date Sum (Qago QTD), but with scope of year for
numerator instead of Quarter.

Metrics Aggregation Details : Rolling Time

22

Rolling 3 Months Sum

Rolling 3 months sum as of selected month : sum of current month plus previous two months value. For January, it gives the value of January plus December and November of the previous year Daily average over Rolling 3 months period : Rolling 3 months sum as of selected month divided by number of days in Rolling 3 Months period. Monthly average over Rolling 3 months period : Rolling 3 months sum as of selected month divided by 3 Rolling 6 months sum as of selected month : sum of current month plus previous two months value. For January, it gives the value of January plus December and November of the previous year Daily average over Rolling 6 months period : Rolling 6 months sum as of selected month divided by number of days in Rolling 3 Months period. Monthly average over Rolling 6 months period : Rolling 6 months sum as of selected month divided by 6

23

Rolling 3 Months Daily Avg Rolling 3 Months Monthly Avg Rolling 6 Months Sum

24

25

26

Rolling 6 Months Daily vg


Rolling 6 Months Monthly Avg

27

22. Rolling 3 Months Sum 25. Rolling 6 Months Sum


Definition :
Rolling X months sum as of selected month : sum of current month plus previous X-1 months value. For January, it gives the value of January plus December and November of the previous year Exp : Rolling 6 Months Sum of revenue for Apr of year N returns sum of revenue from Dec of year N-1 to Apr of year N
Year N - 1
Q1 Q2 Q3 Q4 Q1 Q2

Year N
Q3 Q4

Jan Feb Mar

Apr

May

Jun

July Aug Sep Oct Nov Dec Jan Feb Mar Sum

Apr

May

Jun

July Aug Sep Oct Nov Dec

Use When :
Use this aggregation to track how a metric is structurally trending over time, with limited impact of week to week variations or adjustments. The rolling X months aggregation will give the trailing sum of a metric at any month in time

Logical Formula :
FILTER(All Hist Base Metric USING # of months between selected month and fact month < X) All Hist Base Metric is an logical aggregation that sums up the values for all the month available in db history # of months between selected month and fact month is the calculation of # of months from the selected value in the time dimension, and the month id in the fact table

Limitations :
This aggregation requires that time object month be present in the query, and that no day/week grain object be in the query. Ie, you have to both : use month AND not use day/week with this object.

24. Rolling 3 Months Monthly Avg 27. Rolling 6 Months Monthly Avg
Definition :
Rolling X months monthly average as of selected month : sum of current month plus previous X-1 months value, all divided by X. For January, it gives the value of January plus December and November of the previous year Exp : Rolling 6 Months Monthly Avg of revenue for Apr of year N returns Monthly Avg of revenue from Dec of year N-1 to Apr of year N
Year N - 1
Q1 Q2 Q3 Q4 Q1 Q2

Year N
Q3 Q4

Jan Feb Mar

Apr

May

Jun

July Aug Sep Oct Nov Dec Jan Feb Mar Monthly Avg

Apr

May

Jun

July Aug Sep Oct Nov Dec

Use When :
Use this aggregation to track how a metric is structurally trending over time, limiting impacts of week to week variations or conjectural adjustments. The rolling X months monthly average will give a trailing average aggregation of a metric at any month in time

Logical Formula :
22 or 25 RXM Base Metric / X

Limitations :
This aggregation requires that time object month be present in the query, and that no day/week grain object be in the query. Ie, you have to both : use month AND not use day/week with this object.

23. Rolling 3 Months Daily Avg 26. Rolling 6 Months Daily Avg
Definition :
Rolling X months daily average as of selected month : sum of current month plus previous X-1 months value, all divided by average number of days within X months. Exp : Rolling 6 Months Daily Avg of revenue for Apr of year N returns Daily Avg of revenue from Dec of year N-1 to Apr of year N
Year N - 1
Q1 Q2 Q3 Q4 Q1 Q2

Year N
Q3 Q4

Jan Feb Mar

Apr

May

Jun

July Aug Sep Oct Nov Dec Jan Feb Mar

Apr

May

Jun

July Aug Sep Oct Nov Dec

Sum of metric / Avg number of days

Use When :
Use this aggregation to track how a metric is structurally trending over time, limiting impacts of week to week variations or conjectural adjustments. The rolling X months daily average will give a trailing average aggregation of a metric at any month in time

Logical Formula :
22 or 25 RXM Base Metric / (X x 30.4)

Limitations :
This aggregation requires that time object month be present in the query, and that no day/week grain object be in the query. Ie, you have to both : use month AND not use day/week with this object. The Daily averaging simply results from using fixed denominator of average 30.4 days per month. Therefore it does not take into account differences in the proper number of days for each month. It is possible to enhance the formula using time facts objects to address that limitation

Metrics Aggregation Details : Time Variations


30 31 Value Variation to Month Ago Percent Change to Month Ago Month to Month Ago Compounded Yearly Growth Month To Date to Month Ago Month To Date Percent Change Value Variation to Quarter Ago Percent Change to Quarter Ago Month to Quarter Ago Compounded Yearly Growth Quarter To Date to Quarter Ago Quarter To Date Percent Change Value Variation to Year Ago Value change to same period in previous month Change to same period in previous month, expressed as percent growth of previous month value Variation to last month expressed as a compounded yearly percent growth rate. Example : 2% growth month to Month Ago = 26.82% compounded yearly growth rate. Change to same Month To Date period in previous month, expressed as percent growth of previous month to date value Value change to same period in previous Quarter Change to same period in previous quarter, expressed as percent growth of previous quarter value Variation to same period last Quarter expressed as a compounded yearly percent growth rate. Change to same Quarter To Date period in previous quarter, expressed as percent growth of previous quarter to date value Difference of the value of the metric in the current period with the value of the metric in the same period the previous year Change to same period in previous Year, expressed as percent growth of previous Year value Variation to same Year To Date period last Year expressed as a percent growth rate.

32

33

34

35

36

37

38

39

Percent Change to Year Ago Year To Date to Year Ago Year To Date Percent Change Value Variation to Week Ago

40

41

Difference of the value of the metric in the current period with the value of the metric in the same period the previous Week
Change to same period in previous Week, expressed as percent growth of previous Week value

42

Percent Change to Week Ago

30. Value Variation to Month Ago (Value Var to Mago)


Definition :
Value change to same period in previous month Exp : Value Variation to Month Ago of revenue for May 19th of year N returns the difference between revenue on May 19 th and revenue of Apr 19th of year N
Year N - 1
Q1 Q2 Q3 Q4 Q1 Q2

Year N
Q3 Q4

Jan Feb Mar Sum

Apr

May

Jun

July Aug Sep Oct Nov Dec Jan Feb Mar

Apr Y-X

May

Jun

July Aug Sep Oct Nov Dec

Metric Value on same day in previous month = X

Metric Value on day selected in query = Y

Use When :
Use this aggregation to directly display month to month variation measures in a query

Logical Formula :
Base Metric - 04. Mago Base Metric

Limitations :
This aggregation will return at least Month level detail. Ie, if no time object in the report, or if only objects higher than month, then the query will force return one row per month. Incase there are time objects in the reports at quarter, month, week or day level, then it will break down the results to this level of detail

31. Percent Change to Month Ago (Pct Chg to Mago)


Definition :
Change to same period in previous month, expressed as percent growth of previous month value Exp : Value Variation to Month Ago of revenue for May 19th of year N returns the difference between revenue on May 19 th and revenue of Apr 19th of year N
Year N - 1
Q1 Q2 Q3 Q4 Q1 Q2

Year N
Q3 Q4

Jan Feb Mar Sum

Apr

May

Jun

July Aug Sep Oct Nov Dec Jan Feb Mar

Apr

May

Jun

July Aug Sep Oct Nov Dec

[ ( (Y - X) / X ) - 1 ] *100

Metric Value on same day in previous month = X

Metric Value on day selected in query = Y

Use When :
Use this aggregation to directly display month to month percent variation measures in every row of a query.

Logical Formula :
[ (Base Metric - 04. Mago Base Metric) / 04. Mago Base Metric - 1 ] * 100

Limitations :
This aggregation will return at least Month level detail. Ie, if no time object in the report, or if only objects higher than month, then the query will force return one row per month. Incase there are time objects in the reports at quarter, month, week or day level, then it will break down the results to this level of detail

32. Month to Month Ago Compounded Yearly Growth


Definition :
Variation to last month expressed as a compounded yearly percent growth rate. Example : 2% growth month to Month Ago = 26.82% compounded yearly growth rate. Exp : Month to Month Ago Compounded Yearly Growth on revenue for May 19th of year N returns compounded yearly extrapolation of growth pace between periods of Apr 19th and May 19th of year N.
Year N - 1
Q1 Q2 Q3 Q4 Q1 Q2

Year N
Q3 Q4

Jan Feb Mar Sum

Apr

May

Jun

July Aug Sep Oct Nov Dec Jan Feb Mar

Apr

May

Jun

July Aug Sep Oct Nov Dec

Compounded yearly % growth of (X to Y)

Metric Value on same day in previous month = X

Metric Value on day selected in query = Y

Use When :
Use this aggregation to rapidly convert month to month growth into a compounded yearly percent growth rate. For example, express variation of cash balance over past month into a flat full year percent growth figure.

Logical Formula :
( POWER(Base Metric / 04. Mago Base Metric, 12) - 1) * 100

Limitations :
This aggregation will return at least Month level detail. Ie, if no time object in the report, or if only objects higher than month, then the query will force return one row per month. Incase there are time objects in the reports at quarter, month, week or day level, then it will break down the results to this level of detail

33. Month To Date to Month Ago Month To Date Percent Change (MTD to Mago MTD Pct Chg)
Definition :
Change of month to date current value to last month month to date value expressed as percent change to initial period Exp : MTD to Mago MTD Pct Chg of revenue for May 19th of year N returns the percent growth of revenue between periods from Apr 1st to Apr 19th compared with May 1st to May 19th of year N
Year N - 1
Q1 Q2 Q3 Q4 Q1 Q2

Year N
Q3 Q4

Jan Feb Mar

Apr

May

Jun

July Aug Sep Oct Nov Dec Jan Feb Mar

Apr

May

Jun

July Aug Sep Oct Nov Dec

Mago MTD = X

MTD = Y

Value returned : [ (Y/X) -1 ] x 100

Use When :
Use this aggregation to directly display month to date performance compared to previous month for each record on a query.

Logical Formula :
[ (MTD Base Metric / Mago MTD Base Metric) -1 ] x 100

Limitations :
This aggregation will return at least Month level detail. Ie, if no time object in the report, or if only objects higher than month, then the query will force return one row per month. Incase there are time objects in the reports at quarter, month, week or day level, then it will break down the results to this level of detail

34. Value Variation to Quarter Ago 35. Percent Change to Quarter Ago
34. Value Variation to Quarter Ago (Value Var to Qago) :
Value change to same period in previous Quarter Exp : Value Var to Qago revenue for May 19th of Year N returns difference between value of metric as of May 19 th and value as of Feb 19th of year N
Year N - 1
Q1 Q2 Q3 Q4 Q1 Q2

Year N
Q3 Q4

Jan Feb Mar

Apr

May

Jun

July Aug Sep Oct Nov Dec Jan Feb Mar

Apr

May

Jun

July Aug Sep Oct Nov Dec

Qago Value = X

Current Value = Y

Value returned : Y - X

Similar to : 30. Value Variation to Month Ago, but with scope of Quarter instead of Month. The use case and
formulas structures are similar.

35. Percent Change to Quarter Ago (Pct Chg to Qago):


Change to same period in previous quarter, expressed as percent growth of previous quarter value Exp : Pct Chg to Qago revenue for May 19th of Year N returns percent growth between value of metric as of May 19 th and value as of Feb 19th of year N
Year N - 1
Q1 Q2 Q3 Q4 Q1 Q2

Year N
Q3 Q4

Jan Feb Mar

Apr

May

Jun

July Aug Sep Oct Nov Dec Jan Feb Mar

Apr

May

Jun

July Aug Sep Oct Nov Dec

Qago Value = X

Current Value = Y

Value returned : [ (Y - X)-1 ] /100

Similar to : 31. Percent Change to Month Ago, but with scope of Quarter instead of Month

36. Month to Quarter Ago Compounded Yearly Growth


Definition :
Variation to same period last Quarter expressed as a compounded yearly percent growth rate. Exp : Month to Quarter Ago Compounded Yearly Growth on revenue for May 19th of year N returns compounded yearly extrapolation of growth pace between periods of Feb 19th and May 19th of year N.
Year N - 1
Q1 Q2 Q3 Q4 Q1 Q2

Year N
Q3 Q4

Jan Feb Mar

Apr

May

Jun

July Aug Sep Oct Nov Dec Jan Feb Mar

Apr

May

Jun

July Aug Sep Oct Nov Dec

Compounded yearly % growth calculated betwee Y and X

Metric value Quarter ago = X

Current Value = Y

Use When :
Use this aggregation to rapidly convert Quarter to quarter growth into a compounded yearly percent growth rate. For example, express variation of cash balance over past quarter into a flat full year percent growth figure.

Logical Formula :
( POWER(Base Metric / 07. Qago Base Metric, 4) - 1) * 100

Limitations :
This aggregation will return at least Quarter level detail. Ie, if no time object in the report, or if only objects higher than Quarter, then the query will force return one row per Quarter.

37. Quarter To Date to Quarter Ago Quarter To Date Percent Change (QTD to Qago QTD Pct Chg)
Definition :
Change of quarter to date current value to last quarter quarter to date value expressed as percent change to initial period Exp : QTD to Qago QTD Pct Chg of revenue for May 19th of year N returns the percent growth of revenue between periods from Jan 1st to Feb 19th compared with March 1st to May 19th of year N
Year N - 1
Q1 Q2 Q3 Q4 Q1 Q2

Year N
Q3 Q4

Jan Feb Mar

Apr

May

Jun

July Aug Sep Oct Nov Dec Jan Feb Mar

Apr

May

Jun

July Aug Sep Oct Nov Dec

Metric Year ago Year to date value = X

Current Quarter to Date Value = Y

Value Returned : Compounded Percent growth calculated betwee Y and X

Use When :
Use this aggregation to directly display quarter to date performance compared to previous quarter for each record on a query.

Logical Formula :
[ (14. QTD Base Metric / 18. Qago QTD Base Metric) -1 ] x 100

Limitations :
This aggregation will return at least Quarter level detail. Ie, if no time object in the report, or if only objects higher than Quarter, then the query will force return one row per Quarter.

38. Value Variation to Year Ago 39. Percent Change to Year Ago
38. Value Variation to Year Ago (Value Var to Yago) :
Value change to same period in previous Year Exp : Value Var to Yago revenue for May 19th of Year N returns difference between value of metric as of May 19 th of year N and value as of May 19th of year N-1
Year N - 1
Q1 Q2 Q3 Q4 Q1 Q2

Year N
Q3 Q4

Jan Feb Mar

Apr

May

Jun

July Aug Sep Oct Nov Dec Jan Feb Mar

Apr

May

Jun

July Aug Sep Oct Nov Dec

Similar to : 30. Value Variation to Month Ago, but with scope of Year instead of Month.

Qago Value = X

Current Value = Y

Value returned : Y - X

39. Percent Change to Year Ago (Pct Chg to Yago):


Change to same period in previous Year, expressed as percent growth of previous Year value. Exp : Pct Chg to Yago revenue for May 19th of Year N returns percent growth between value of metric as of May 19 th of year N and value as of May 19th of year N-1
Year N - 1
Q1 Q2 Q3 Q4 Q1 Q2

Year N
Q3 Q4

Jan Feb Mar

Apr

May

Jun

July Aug Sep Oct Nov Dec Jan Feb Mar

Apr

May

Jun

July Aug Sep Oct Nov Dec

Qago Value = X

Current Value = Y

Value returned : [ (Y - X)-1 ] /100

Similar to : 31. Percent Change to Month Ago, but with scope of Year instead of Month

40. Year To Date to Year Ago Year To Date Percent Change (YTD to Yago YTD Pct Chg)
Definition :
Change of year to date current value to last year year to date value expressed as percent change to initial period Exp : YTD to Yago YTD Pct Chg of revenue for May 19th of year N returns the percent growth of revenue between periods from May 1st to May 19th of year N-1 compared with May 1st to May 19th of year N
Year N - 1
Q1 Q2 Q3 Q4 Q1 Q2

Year N
Q3 Q4

Jan Feb Mar

Apr

May

Jun

July Aug Sep Oct Nov Dec Jan Feb Mar

Apr

May

Jun

July Aug Sep Oct Nov Dec

Metric Year ago Year to date value = X

Percent growth calculated betwee Y and X

Current Year to Date Value = Y

Use When :
Use this aggregation to directly display quarter to date performance compared to previous quarter for each record on a query.

Logical Formula :
[ (19. YTD Base Metric / 21. Yago YTD Base Metric) -1 ] x 100

41. Value Variation to Week Ago 42. Percent Change to Week Ago

41. Value Variation to Week Ago (Value Var to Wago) :


Value change to same period in previous Week Exp : Value Var to Week ago revenue for May 19th of Year N returns difference between value of metric as of May 19 th and value as of May 12th of year N
Year N - 1
Q1 Q2 Q3 Q4 Q1 Q2

Year N
Q3 Q4

Jan Feb Mar

Apr

May

Jun

July Aug Sep Oct Nov Dec Jan Feb Mar

Apr

May

Jun

July Aug Sep Oct Nov Dec

Week ago Value = X Current Value = Y

Value returned : Y - X

Similar to : 30. Value Variation to Month Ago, but with scope of Week instead of Month.

42. Percent Change to Week Ago (Pct Chg to Wago):


Change to same period in previous Week, expressed as percent growth of previous Week value. Exp : Pct Chg to Yago revenue for May 19th of Year N returns percent growth between value of metric as of May 19 th and value as of May 12th of year N
Year N - 1
Q1 Q2 Q3 Q4 Q1 Q2

Year N
Q3 Q4

Jan Feb Mar

Apr

May

Jun

July Aug Sep Oct Nov Dec Jan Feb Mar

Apr

May

Jun

July Aug Sep Oct Nov Dec

Week ago Value = X Current Value = Y

Value returned : [ (Y - X)-1 ] /100

Similar to : 31. Percent Change to Month Ago, but with scope of Week instead of Month

Metrics Aggregation Details : Time Span Variations


43 44 First on Time span Last on Time span First oT returns the value of the first day in the period selected for the query. Last oT returns the value of the last day in the period selected for the query. For any time interval specified in the query row, this aggregation returns the difference between value at beginning of the interval and value at end of the interval. The values at beginning and end date that make up the variation can be displayed by adding aggregation objects Fst oT and Last oT. Info about beginning date and end date for each row are available with time facts objects : First Time Day Dt and Last Time Day Dt. For example, if a query is filtered for a quarter, this will show the Value Variation between first and last day of the quarter. If the query objects are at week level, This aggregation will returns for each row (week level) the value of the last day of each week minus value of the first day in that week. For any time interval specified in the query row, this aggregation returns the growth between value at beginning of the interval and value at end of the interval expressed as a % to initial value. The values at beginning and end date that make up the variation can be displayed by adding aggregation objects First oT and Last oT. Info about beginning date and end date for each row are available with time facts objects : First Time Day Dt and Last Time Day Dt. This aggregation divides the Value Variation aggregation metric by number of months in the time interval considered. Note that number of months is decimal value (not integer), and only dependent on number of days in the time interval. Example : time interval between Jan 13th and Mar 27 = 2.48 months This aggregation converts the Avg Monthly Value Variation aggregation metric into a simple arithmetic monthly average percentage variation. Note, this is value is not a compounded % variation. This aggregation divides the Value Variation aggregation metric by number of years in the time interval considered. Note that number of years is decimal value (not integer), and only dependent on number of days in the time interval. This returns the compounded yearly variation rate for the time interval considered. It projects the growth rate in the interval to a basis of 1 year. Exp 1 : 50% growth over 24 months = 22.47% compounded yearly growth rate. Exp 2 : 2% growth over 1 month = 26.82% compounded yearly growth rate.

45

Value Variation

46

Percent Change

47

Average Monthly Value Variation

48

Average Monthly Percent Change Average Yearly Value Variation

49

50

Compounded Yearly Percent Change

43. First on Time Span 44. Last on Time Span


Definition :
First oT returns the metric value of the first day where there is data in the period selected for the query. Last oT returns the value of the last day where there is data in the period selected for the query. Exp : Fst oT and Lst oT revenue for query filtered as between Apr 01 of year N and May 19 th of year N returns two figures : daily value for Apr 1st and daily value for May 19th
Year N - 1
Q1 Q2 Q3 Q4 Q1 Q2

Year N
Q3 Q4

Jan Feb Mar Sum

Apr

May

Jun

July Aug Sep Oct Nov Dec Jan Feb Mar

Apr

May

Jun

July Aug Sep Oct Nov Dec

[[ (Y - X) / X ] -1 ] x 100

Use When :

Metric Value on First day of time selected = X

Metric Value on Last day of time selected = Y

Use these aggregations when analyzing variations over a time span, and are interested in showing starting and ending values for the time span. Fst oT and Lst oT will adapt to any time grain in the query and will break down results by each time grain.

Hint :
To see what are the actual dates for Fst oT and Lst oT, use time facts objects T60 First Time Day Dt and T61 Last Time Day Dt

Logical Formula :
Base Metric with aggregation rule First (or Last) on Time dimension, Sum on all other dimensions

45. Value Variation


Definition :
For any time interval specified in the query row, this aggregation returns the difference between value at beginning of the interval and value at end of the interval. The values at beginning and end date that make up the variation can be displayed by adding aggregation objects Fst oT and Last oT. Info about beginning date and end date for each row are available with time facts objects : First Time Day Dt and Last Time Day Dt. Exp : Value Variation aggregation on revenue for query filtered as between Apr 01 of year N and May 19 th of year N returns value as of May 19th minus value as of Apr 1st
Year N - 1
Q1 Q2 Q3 Q4 Q1 Q2

Year N
Q3 Q4

Jan Feb Mar Sum

Apr

May

Jun

July Aug Sep Oct Nov Dec Jan Feb Mar

Apr Y-X

May

Jun

July Aug Sep Oct Nov Dec

Use When :

Metric Value that day = X

Metric Value that day = Y

Use that aggregation when analyzing variations over a custom time span, or variations for each row in a query. For any custom time period, or for any time grain rows in a query, Value Variation will show delta between starting and ending values for the time span. It will adapt to any time grain in the query and will break down results by each row.

Hint :
To see what are the actual dates for Fst oT and Lst oT, use time facts objects T60 First Time Day Dt and T61 Last Time Day Dt

Logical Formula :
44 Last on Time - 43 First on Time

46. Percent Change


Definition :
For any time interval specified in the query row, this aggregation returns the growth between value at beginning of the interval and value at end of the interval expressed as a % to initial value. The values at beginning and end date that make up the variation can be displayed by adding aggregation objects Fst oT and Last oT. Info about beginning date and end date for each row are available with time facts objects : First Time Day Dt and Last Time Day Dt. Exp : Percent Change aggregation on revenue for query filtered as between Apr 01 of year N and May 19 th of year N returns the percent growth of revenue between Apr 1st and May 19 th
Year N - 1
Q1 Q2 Q3 Q4 Q1 Q2

Year N
Q3 Q4

Jan Feb Mar Sum

Apr

May

Jun

July Aug Sep Oct Nov Dec Jan Feb Mar

Apr

May

Jun

July Aug Sep Oct Nov Dec

[[ (Y - X) / X ] -1 ] x 100

Use When :

Metric Value that day = X

Metric Value that day = Y

Use Percent Change aggregations when analyzing percent growth over a custom time span, or growth for each row in a query. For any custom time period, or for any time grain rows in a query, Percent Change will show delta between starting and ending values for the time span. It will adapt to any time grain in the query and will break down results by each row.

Hint :
To see what are the actual dates for Fst oT and Lst oT, use time facts objects T60 First Time Day Dt and T61 Last Time Day Dt

Logical Formula :
[ (44 Last on Time 43 First on Time) / 43 First on Time 1 ] * 100

47. Average Monthly Value Variation


Definition :
This aggregation divides the Value Variation aggregation metric by exact number of months included in the custom time interval considered. Note that number of months is decimal value (not integer), and only dependent on number of days in the time interval. Example : time interval between Jan 13th and Mar 27 = 2.48 months Exp : Avg Monthly variation on revenue for query filtered as between Apr 01 of year N and May 19 th of year N extrapolates growth pace of Apr 1st and May 19th over a full single month.
Year N - 1
Q1 Q2 Q3 Q4 Q1 Q2

Year N
Q3 Q4

Jan Feb Mar Sum

Apr

May

Jun

July Aug Sep Oct Nov Dec Jan Feb Mar

Apr

May

Jun

July Aug Sep Oct Nov Dec

(Y - X) / number of months (decimal) between last day and first day

Metric Value on First day of time selected = X

Metric Value on Last day of time selected = Y

Use When :
Use this aggregation to rapidly convert growth over any time period into a monthly change figure. For example, express variation of cash balance over past 10 days or past quarter into a comparable monthly growth figure. This aggregation will extrapolate the values and calculate the average monthly change for each.

Hint :
To see what are the actual dates for Fst oT and Lst oT, use time facts objects T60 First Time Day Dt and T61 Last Time Day Dt. To see the number of months considered in the calculation for each period, use time facts object T64 # of Months

Logical Formula :
45. Value Variation / T64 # of Months

48. Average Monthly Percent Change


Definition :
This aggregation converts the Avg Monthly Value Variation aggregation metric into an average growth expressed as a percent of initial value. Note, this is value is not a compounded % variation Exp : Average Monthly Percent Change of revenue for custom time period Apr 1st to May 19th of Year N returns % value that represents Avg monthly growth of revenue between Apr 1st and May 19th of year N
Year N - 1
Q1 Q2 Q3 Q4 Q1 Q2

Year N
Q3 Q4

Jan Feb Mar Sum

Apr

May

Jun

July Aug Sep Oct Nov Dec Jan Feb Mar

Apr

May

Jun

July Aug Sep Oct Nov Dec

Avg Monhtly % growth of metric value between First Day and Last Day

Use When :

Metric Value on First day of time selected

Metric Value on Last day of time selected

Use this aggregation to rapidly convert growth over any time period into a monthly percentage change figure. For example, express variation of cash balance over past 10 days or past quarter into a comparable monthly percent growth figure. This aggregation will extrapolate the values and calculate the average monthly percent change for each.

Hint :
To see what are the actual dates for Fst oT and Lst oT, use time facts objects T60 First Time Day Dt and T61 Last Time Day Dt. To see the number of months considered in the calculation for each period, use time facts object T64 # of Months

Logical Formula :
46. Percent Change / T64 # of Months

49. Average Yearly Value Variation


Definition :
This aggregation divides the Value Variation aggregation metric by exact number of years in the time interval considered. The number of years is decimal value (not integer) and only calculated based on number of days in the time interval. Exp : Avg Yearly variation on revenue for query filtered as between Apr 01 of year N and May 19 th of year N returns the value growth of revenue for the whole year, extrapolating growth pace of Apr 1st and May 19 th over the whole year.
Year N - 1
Q1 Q2 Q3 Q4 Q1 Q2

Year N
Q3 Q4

Jan Feb Mar Sum

Apr

May

Jun

July Aug Sep Oct Nov Dec Jan Feb Mar

Apr

May

Jun

July Aug Sep Oct Nov Dec

(Y - X) / number of years (decimal) between last day and first day

Metric Value on First day of time selected = X

Metric Value on Last day of time selected = Y

Use When :
Use this aggregation to rapidly convert growth over any time period into a yearly value change figure. For example, express variation of cash balance over past 3 months or past 5 quarters into a comparable yearly growth figure. This aggregation will extrapolate the values and calculate the average yearly change for each.

Hint :
To see what are the actual dates for Fst oT and Lst oT, use time facts objects T60 First Time Day Dt and T61 Last Time Day Dt. To see the number of months considered in the calculation for each period, use time facts object T66 # of Years

Logical Formula :
45. Value Variation / T64 # of Months

50. Compounded Yearly Percent Change


Definition :
This returns the compounded yearly variation rate for the custom time interval considered. It projects the growth rate in the interval to a full basis of 1 year. Example 1, a 50% growth rate over 24 months is equivalent to a 22.47% compounded yearly growth rate. Example 2, a 2% growth over 1 month is equivalent 26.82% compounded yearly growth rate. The formula for compounded growth rate leverage the power arithmetic function. Exp : Compounded Yearly Percent Change on revenue for query filtered as between Apr 01 of year N and May 19 th of year N returns the full year percent growth of revenue, using a compounded extrapolation of actual growth pace of Apr 1st and May 19th.
Year N - 1
Q1 Q2 Q3 Q4 Q1 Q2

Year N
Q3 Q4

Jan Feb Mar Sum

Apr

May

Jun

July Aug Sep Oct Nov Dec Jan Feb Mar

Apr

May

Jun

July Aug Sep Oct Nov Dec

(Y - X) / number of years (decimal) between last day and first day

Use When :

Metric Value on First day of time selected = X

Metric Value on Last day of time selected = Y

Use this aggregation to rapidly convert growth over any time period into a compounded yearly percent growth rate. For example, express variation of cash balance over past 3 months or past 5 quarters into a flat full year percent growth figure.

Hint :
To see what are the actual dates for Fst oT and Lst oT, use time facts objects T60 First Time Day Dt and T61 Last Time Day Dt. To see the number of months considered in the calculation for each period, use time facts object T66 # of Years

Logical Formula :
[ POWER ( (1 + "46 Percent Variation" / 100) , (1 / "T66 # of Years) ) 1 ]* 100

Metrics Aggregation Details : Level Based (1/2)


Full Month Sum Month Ago Full Month Sum Sum of the metric value for the full month. This aggregation is always calculated at the month level.. For example, when viewed against Week level, it shows the same total month's value for all the weeks in that month. Full month's value of the metric for the previous month. This aggregation is always calculated at the month level. For example, when viewed against Week level, it shows the same total month's value for all the weeks in that month. Full month's value of the metric for the same month in the previous year. This aggregation is always calculated at the year level. For example, when viewed against week level, it shows the same total month's value for all the weeks in that month. Full month's value of the metric for 12 months prior to the current month. This aggregation is always calculated at the month level. For example, when viewed against week level, it shows the same total month's value for all the weeks in that month. Sum of the metric value for the full quarter. This aggregation is always calculated at the quarter level. When viewed against Month or Week level, it shows the quarter's value for all the weeks and months in that quarter. Sum of the metric value for the full week. This aggregation is always calculated at the week level. When viewed against day level, it shows the week's value for all the days in that week. Sum of the metric value for the full year. This aggregation is always calculated at the year level. When viewed against quarter, month or week level, it shows the year's value for all the quarters, month, and weeks in that year. Full year's value of the metric for the previous year. This aggregation is always calculated at the year level Full quarter's value of the metric for the previous quarter. This aggregation is always calculated at the quarter level

51

52

53

Year Ago Full Month Sum


12 Months Ago Full Month Sum Full Quarter Sum

53

54

55

Full Week Sum

56

Full Year Sum Year Ago Full Year Sum Quarter Ago Full Quarter Sum

57

58

Metrics Aggregation Details : Level Based (2/2)


Sum for All Customers Sum for All Markets Sum for All Products Sum for All Employees Sum of the metric for all the Customers. It always returns the value at Total level irrespective of the level within the Customer dimension that is used in the query Sum of the metric for all the Markets. It always returns the value at Total level irrespective of the level within the Market dimension that is used in the query Sum of the metric for all the Products. It always returns the value at Total level irrespective of the level within the Product dimension that is used in the query Sum of the metric for all the Employees. It always returns the value at Total level irrespective of the level within the Employee dimension that is used in the query Sum of the metric for all the dimensions except Time. It always the value at the Total level when viewed against all the dimensions except time. When viewed against Time, it returns the value at that level. Sum of the metric for all the Orders. It always returns the value at Total level irrespective of the level within the Orders dimension that is used in the query Sum of the metric for by orders. This aggregation always returns the value at the detail order level irrespective of the level within the dimension that is used in the query Sum of the metric for by Customers. This aggregation always returns the value at the detail order level irrespective of the level within the dimension that is used in the query Sum of the metric for by Employees. This aggregation always returns the value at the detail order level irrespective of the level within the dimension that is used in the query

61

62

63

64

65

Sum for All Dims But Time

66

Sum for All Orders

67

Sum by Orders

68

Sum by Customers Sum By Employees

69

Metrics Aggregation Details : Other Aggs (1/2)


This aggregation will fragment physical data to a lower granularity then it is in the database, using the pattern of last year spread. Example : Forecast data is available by month in db, this metric will take the 'by month' information, and break it down at day level using the same spread as how last year's actual daily values were spread over the same month. This aggregation will apply similar break down on the following dimensions : Time, Customer, Market. Ie, forecast data available only by month by market level is fragmented further down to be "by customer, by product, by day" according to how that pattern was the previous year.

76

Fragmented Year Ago Pattern

77

Fragmented Quarter Ago Pattern

This aggregation will fragment physical data to a lower granularity then it is in the database, using the pattern of last quarter spread. Example : Forecast data is available by month in db, this metric will take the 'by month' information, and break it down at day level using the same spread as how last quarter's actual daily values were spread over the same month. This aggregation will apply similar break down rule on the following dimensions : Time, Customer, Market. Ie, forecast data available only by month by market level is fragmented further down to be "by customer, by product, by day" according to how that pattern was the previous quarter.
This aggregation will fragment physical data to a lower granularity then it is in the database, using linear rule spread. Example : Forecast data is available by month in db, this metric will take the 'by month' information, and break it down at day level simply by dividing by number of days available in each month. This aggregation will apply similar break down on the following dimensions : Time, Customer, Market. Ie, forecast data available only by month by market level is linearly fragmented further down to be "by customer, by product, by day". For any time interval specified in the query row, this aggregation returns the sum of the metric value over this interval divided by the number of days in that period. To visualize beginning and end dates, and number of days used in the calculation, use time facts objects : First Time Day Dt Last Time Day Dt and # of Days. For example, if a query is filtered for a quarter, this will show the average daily value for this quarter. For any time interval specified in the query row, this aggregation returns the sum of the metric value over this interval divided by the number of weeks in that period. To visualize beginning and end dates, and number of weeks used in the calculation, use time facts objects : First Time week Dt Last Time week Dt and # of weeks. Note that number of weeks is decimal value (not integer), and only dependent on number of days in the time interval. For example, if a query is filtered for a quarter, this will show the average weekly value for this quarter.

78

Fragmented Flat Pattern

80

Daily RunRate

81

Weekly RunRate

Metrics Aggregation Details : Other Aggs (2/2)


82 Monthly RunRate For any time interval specified in the query row, this aggregation returns the sum of the metric value over this interval divided by the number of months in that period. To visualize beginning and end dates, and number of months used in the calculation, use time facts objects : First Time month Dt Last Time month Dt and # of months. Note that number of months is decimal value (not integer), and only dependent on number of days in the time interval. Example : time interval between Jan 13th and Mar 27 = 2.48 months. For example, if a query is filtered for a quarter, this will show the average monthly value for this quarter. For any time interval specified in the query row, this aggregation returns the sum of the metric value over this interval divided by the number of quarters in that period. To visualize beginning and end dates, and number of quarters used in the calculation, use time facts objects : First Time quarter Dt Last Time quarter Dt and # of quarters. Note that number of quarters is decimal value (not integer), and only dependent on number of days in the time interval. For example, if a query is filtered for a quarter, this will show the average quarterly value for this quarter. For any time interval specified in the query row, this aggregation returns the sum of the metric value over this interval divided by the number of years in that period. To visualize beginning and end dates, and number of years used in the calculation, use time facts objects : First Time year Dt Last Time year Dt and # of years. Note that number of years is decimal value (not integer), and only dependent on number of days in the time interval. For example, if a query is filtered for a year, this will show the average yearly value for this year. This aggregation retrieves the percentage that a time value represents to the total of the current month. For any grain of the query that is below the month level detail, this will return how much the value is for the total month it belongs to. If the query grain is higher than month, this aggregation will return results at month level (a row for each month, with 100% as the value). Note that this metric will consider the amount of data in the db, and extrapolate the result in case data is not available for full month in the db (in order to show consistent percentage values where data is available for full month) This aggregation retrieves the percentage that a time value represents to the total of the current quarter. For any grain of the query that is below the quarter level detail, this will return how much the value is for the total quarter it belongs to. If the query grain is higher than quarter, this aggregation will return results at quarter level (a row for each quarter, with 100% as the value). Note that this metric will consider the amount of data in the db, and extrapolate the result in case data is not available for full quarter in the db (in order to show consistent percentage values where data is available for full quarter) This aggregation retrieves the percentage that a time value represents to the total of the current year. For any grain of the query that is below the year level detail, this will return how much the value is for the total year it belongs to. If the query grain is higher than year, this aggregation will return results at year level (a row for each year, with 100% as the value). Note that this metric will consider the amount of data in the db, and extrapolate the result in case data is not available for full year in the db (in order to show consistent percentage values where data is available for full year).

83

Quarterly RunRate

84

Yearly RunRate

85

Seasonal Percent of Total Month


Seasonal Percent of Total Quarter

86

87

Seasonal Percent of Total Year

4. Cloning Samples content on


other environments with :

CAF V1.0
Content Accelerator Framework
Cloner Utility

CAF V1 Utility
What it is

CAF V1 is a user friendly utility that allows to clone any existing OBI EE

calculations, report and dashboards setup into any target OBI EE environment.
After one dashboard is completed with its appropriate reports, calculations, variables, RPD objects, views layout and formatting, why not apply its structure as a template to other functional cases ?

CAF V1.0 utility makes this task very straightforward and simple.

Samples Sales content can easily be used as source content for CAF V1 in order to duplicate logical constructs or webcat layouts to any target environment.

CAF V1.0 Value offer


OBI Functional Value Library (Samples or other)
Customer Functional Domain Owner Customer business
Marketing Manager
Gain & Refine Customer Insight Design Customer Segments Develop Retention & Growth Campaigns

Customer Agent

Provide Personalized Customer Service

Real-Time Offer Allocation

Generate & Execute Lead

Measure Performance

Customer

Request Service

Respond to Offer

Adopts selected content from functional BI Value library for his domain

Increased processes efficiency Increased OBI EE ROI

Library Configuration Customer IT

Customer Environment

Deploys selected content setup, Simple & Easy Fast Secure optimized

CAF V1.0 Cloning


Practical Overview

How to replicate content from any existing environment (Sample Sales in this example)

into any different target environment (Paint in this example)

with only a few clicks, a few seconds, and zero editing work

CAF V1.0 Cloning Overview


4 intuitive steps to complete cloning of any content

Step 1 : Launch Cloning utility for selected reports (or dashboards) Step 2 : Declare source and target environments (Pick Subject Area from choices) Step 3 : Provide Mappings for Required Objects (declarative mapping) Step 4 : Provide name / path for cloned reports Step 5 (Optional) : rename RPD objects created

CAF V1.0 Cloning Overview


Step 1 : Launch Cloning for selected reports (or dashboards)

CAF V1.0 Cloning Overview


Step 2 : Declare source and target environments

CAF V1.0 Cloning Overview


Step 2 : Select Target Subject Area

CAF V1.0 Cloning Overview


Step 3 : Declare Mappings for Required Objects

Cloner module identifies minimum object required to build the selected reports Cloner module will create necessary logical objects in target RPD if/where needed based on minimum mappings provided Cloner module will create all webcat reports in target environment using mapping information and any new rpd object created

CAF V1.0 Cloning Overview


Step 4 : Provide name / path for cloned reports

V1 : Use Case
1. Operational user :
Chooses report from list of templates Expresses functional context for the target report Benefits : Learn by seeing what OBI EE can functionally deliver Deliver better expression of needs Reach deeper BI requirement faster

2.

IT Developer / IT :
Run CAF on selected reports Select base columns to map Run Synchronizer to adapt object names Benefits : Fast and automated development process Reduced risk in design development Better leverage of OBI EE capabilities

Content Accelerator Framework


Enablement Technology for all OBI EE stakeholders
Rapid Time to Value, Speed up Implementation & adoption : fast deploy, fast learn

Targets
Any OBIEE Environment

Templates Library
RPD Web -cat

Increase ROI : broader & deeper functional leverage of OBI EE Adopt Best Practices :
- Functional : all what OBI EE can really accomplish -Technical : fully utilize OBI EE architecture & features

No Specific OBIEE Skills required

CAF
Cloner
Synchro nizer

Reduce Dev Risk : Automated high Quality designs Drives better & Deeper usage of our platform

Stakeholders
Customers Consultants Partners PMs Field Engineering QA Support Help POC Quality

Help drive platform enhancements


Secure dev process : reduce time and risk (less QA)

Content Accelerator Framework


Some Use Cases examples

1.

Customer sees example of dashboards / reports that would apply to his functional area, and wants to implement on his own environment quickly, Customer is interested in leveraging some logical calculations / RPD features in in his environment, but not sure how to build it, Customer creates reports or dashboards template formats that he needs to use across several functional areas Customer renames existing objects in RPD, and needs to upgrade webcat with these

2.

3.

4.

5.

CAF V1 & Value Layers of OBI EE Content


CAF V1.0 Coverage
Dashboards Structure & Page layouts

Dashboards

Field & Variable Prompts Reports Views Layout Columns Syntax & Formulas

Reports

Filtering Query Structure Webcat Foldering Structure Presentation layer objects Logical Layer : Logical calculations

RPD

Logical Layer : Root Logical Model Physical Layer RPD Variables, Security, Marketing

Database, OLAPs, ETL,

Aggregations, Physical Model ETL logics

CAF V1 Demos : overall scenario context


Overall on site POC scenario context
Source Library environment : Customer environment (for training purposes) Requirement : showcase some of OBI EE capabilities by cloning many Sample Sales reports, dashboards and logical metrics in customer environment. Timeline : 15 min Online or Offline Release : 10.1.3.4 CAF V1 Utility installed Online

Source : Sample Sales

Target : Paint (vanilla) Webcat

Webcat

RPD

Target RPD

CAF V1

Offline

Offline

CAF V1 Demos : Cloner Module


How it works
Source Webcat
1. Parses XML content of selected reports 8. Creates resulting report XML files in Target Webcat

Target Webcat

Online or Offline

Online or Offline

Cloner
2. Parses Source RPD with XUDML 5. Parses XUDML Target RPD

Target RPD

Source RPD

3. Identifies Logical calculations in selected reports

6. Recognizes existing Logical calculations


7. Modifies Target RPD

Offline Offline
4. Identifies minimal mapping info needed from user

CAF V1 Utility
What it does

1. 2.

Clones Reports, Dashboards, RPD logical formulas Allows Saving of mappings info

3.
4. 5. 6. 7. 8.

Allows formula overriding


Introduces Optional / required mapping Handles level mapping Traceable Logs & Backups Sources from any existing content Synchs up RPD + Webcat names

CAF V1 Where to get it on OTN ?


http://www.oracle.com/technology/products/bi/enterprise-edition.html

CAF V1 = free utility,

Not a Licensed product, not maintained

INSTALL
Download 8 megs install batch file from OTN, with simple install directions Requires 10.1.3.4 & JDK 1.6 Installs by pasting 7 files in any existing 10.1.3.4 environment Comprehensive functional user Guide

5. Switching/Updating Alternative Datasets


How to switch to industry specific datasets, or update dataset dimensions to custom values

Switching to Alternative prebuilt datasets

Product and Market dimension values are stored in two distinct Xml files : Product.xml and Market.xml Alternative sets of product and market data are provided with the 1.1 install of sample sales application.
For product dimension, a choice of 6 industry flavors is provided (Communications, Energy, Food, Media, Medical, Banking) with 6 additional distinct files. For Market dimension, a file for EMEA one for North America and one for APAC are provided as well.

To switch from one data set to another one :


Stop BI Server Rename original product.xml file to product_Generic.xml Rename file of your choice from Product_Ind_XXX_xml to product.xml Start BI server.

Same procedure applies to switching to alternate Market data

Switching to Alternative prebuilt datasets

Using generic - Market.xml - Product.xml files

Using files : - Market_Emea.xml - Product_Ind_ Comm.xml

Updating Dataset to custom values

1/2

You can also simply update data from XML datasources using MS Access software for example :

1. Import XML tables that need to be changed in a blank answers db


Access menu : >File>Get External Data>Import Use filter for Files of type XML and browse to the location where you have the source XML files Import the file (click ok to window suggesting table structure and data import)

2. Update content of the table within Access Db.

Updating Dataset to custom values


3. Once data is updated, then Export table back under XML format :
select table to save and click menu >File>Export. Use save as type XML and browse to the location where you have the source XML files Save by overriding the original file name (you may want to keep a backup copy of original file)

2/2

4. Correct tablename Tag of each changed files :


Using windows explorer, navigate to the directory where you have saved updated XMLs Open updated xml files with notepad Scroll to line # 2 and replace the text from : <dataroot xmlns:od="urn:schemas-microsoft-com:officedata" .> to : <Table Name="Product"> or <Table Name=Market"> or other depending on the file you are changing (use the name of the file) Scroll down to last line of the file, and replace string from </dataroot> to </Table>

5. Stop and Start OBI EE server. You should now see your updated data in reports