You are on page 1of 6

The Shaffer 4 C's of Data Visualization

"Clean" Examples
Default in Excel Default in Excel
25 30
20 25
15 20
10 15
5 10
0 Series1
5
Series1
0

Category

B
Category

Category

Category
Categor
A

D
C

E
y
Remove/Lighten Gridlines and lower # of Y-Axis Tick Marks

30 30
20
10
0 20
Series1
Category
Category

10
Series1
B
A

0
Category

B
Category

Category

Category
Categor
A

D
C

E
y

Avoid Rotated Text Rotate Chart Instead


0102030
30
20
Category A Category B Category C Category D
10
0
Series1
Category
Category

Removed Series Name


B
A

30
20
10 Category I
0
Category G Category E Category C
Category A

0 10 20 30
The Shaffer 4 C's of Data Visualization
"Clean" Examples
Order Bars Order Bars
0 10 20 30
Category B Category C Category D Category A
0 10 20 30 Category G
Category B Category J Category C Category E Category D Category H Category A Category I C

Removed tick marks on y-axis

Note - Labels in units of 10 with gridlines. Added minor


gridlines and minor ticks in units of 5.

Ordering Data adds context unless a specific order is required.


Ex. North, South, East, West (save alphabetical order for reference tables)

Remove Border Remove Border


0 10 20 30
Category B
Category G
Category C Category B
Category J
Category D
Category C
Category A Category E
Category D
0 10 20 30 Category H
Category A
Category I
Category F

If reasonable number of bars consider data labels instead of axis labels

Category B 23 Category B 23

Category C 17 Category C 17

Category D 14 Category D 14

Category A 12 Category A 12

Category B 23 Category B 23

Category C 17 Category C 17

Category D 14 Category D 14

Category A 12 Category A 12
The Shaffer 4 C's of Data Visualization
"Clean" Examples

Fine Tune Colors (ex. Dark Gray vs. Black)

Category B 23

Category C 17

Category D 14

Category A 12

Add Context - Ex. add Target Line, Expected Value or Average

Category B 23

Category C 17 Tip - Add target lines and text at last step to


Category D avoid reformatting of chart. Using a dotted line
14
or different color will help contrast the target line
Category A 12 vs. a gridline.
Target 16%

Avoid Stark Borders on Bars Avoid 3D Bars

Category B Category B
23
Category C Category C
17
Category D
Category D
14 Category A
Category A 0 5
12 10
15 25 30

Avoid Narrow Bars Avoid Connected Bars (unless distribution)

Category B 23 Category B 23

Category C 17 Category C 17

Category D 14 Category D 14

Category A 12 Category A 12

Note - distributions are typically graphed as vertical bars, connected and with a border.
The Shaffer 4 C's of Data Visualization
"Clean" Examples
Avoid Repeated Text. Example, if "Category" were truly repeated vs. unique names.
Example with y-axis and one highlighted data point. 3 labels vs. 10 Data points.
26

20

10

0
G B J C E D H A I F
Category

Example with 10 data points and no y-axis.

2
6 2
3
1 1
8 1
7 6 1 1
4 4 1 1
G B J C E D H A2 I1 7F
Category

Add Title (tell the story)


Include Labels ($, # or %), Data Source, Date, Author, etc.

Bananas exceeded target by 7 units in 2011


(# of units sold)

Bananas 23

Apples 17

Grapes 14

Oranges 12
Target 16 units
Source - 2011 Actual Sales from BI Tool
Created by Jeffrey A. Shaffer
2/17/2011
The Shaffer 4 C's of Data Visualization
"Clean" Examples

Sometimes it's necessary to show a % of a total. Example, Bananas as a % of total fruit.


Avoid Pie Charts
Bananas Apples Grapes Oranges

Oranges
18% Bananas
18%
35% 35%
Grapes
21% 21%

26% Apples
26%

Use bar charts for easier comparison and 100% stacked-bar chart for part to whole

Bananas exceeded target by 7 units in 2011


(# of units sold)

23
Bananas

17
Apples

14
Grapes

12
Oranges

Target 16 units

Bananas represent 35% of total unit sales


Bananas 35% Apples 26% Grapes 21% Oranges 18%
Source - 2011 Actual Sales from BI Tool
Created by Jeffrey A. Shaffer
2/17/2011
The Shaffer 4 C's of Data Visualization
"Clean" Examples

Here is an excellent example of these practices in use taken from:


TDWI Best Practices Report entitled "Transforming Finance" by Wayne Eckerson

You might also like