What is customer lifetime value?

At the most basic level, it is a theoretical value of a customer calculated as the net present value of the average predicted profit expected from sales to (or cash flow attributed to) any customer. From a practical perspective, the concept of customer lifetime value is appealing because it not only provides a tangible value to be associated with customer acquisition, but it also directs the key stakeholders in the company to take a long-term view associated with managing and maintaining the customer relationship. The customer lifetime value concept can also provide guidance into strategies for dealing with different kinds of customersmore on that in a bit. There are many models for calculating customer lifetime value. Although these models can incorporate many different variables, most focus on key concepts such as these:
y

Acquisition cost: This incorporates the costs associated with convincing a prospective customer to purchase your product or service. Customer lifetime: This is the duration of the companys relationship with the customer. Retention rate: The annual retention rate is the percentage of customers that remain engaged with your company. Retention cost: These include any costs incurred related to actively maintaining the customer relationship, such as rebates or elimination of service fees. Revenue per customer: The annual average revenue per customer is the cumulative amount of revenue divided by the total number of (engaged) customers. Servicing cost: These costs include ones associated with supporting the customer and providing service.

Gross profit: This is the difference between what is brought in as revenue and the cost of creating and providing the product and/or service prior to deducting the operating expenses (including overhead, direct costs, taxes, etc.). Discount rate: In this context, this is the rate used to calculate the current value of future cash flows.

A customers lifetime value is then a function of the net present value of the revenue per customer over the customers lifetime minus the acquisition, retention, and servicing costs. Of course, there could be other variables as well. For example, when a customer severs his or her relationship, there might be a cost to the business associated with the tasks of attrition. On the other hand, the customer agreement may stipulate that severing the relationship prior to the end of an agreed-to period may incur additional charges (check your mobile phone contract!), which then adds in a variable associated with severance revenue! How can customer lifetime value be used? One simple way looks at the value proposition for customer acquisition. Plainly stated, if the cost of acquisition is less than the customer lifetime value, there would have to be some motive other than profitability that drives customer acquisition! Another use involves more comprehensive customer segmentation. Average customer lifetime value across the entire customer base may be a single data point, but analyzing customer lifetime value within discrete customer segments may influence marketing and sales. For example, a companys overall average customer lifetime value may be $25, but if the customer lifetime value of males between the ages of 18 and 34 is $40, it may make sense to orient the marketing budget toward individuals meeting that specific demographic profile because the potential return on investment is much greater. One more use is looking at evolving optimizations in the

organization to increase customer lifetime value. This can be done by focusing on optimizing the dependent variables, such as lowering the associated costs, increasing customer lifetime, or increasing the profitability. Yet one big challenge for calculating and managing customer lifetime value is accumulating the right data needed for each of the dependent variables. Lets look at one of these variables: duration of the customer lifetime. Calculating customer lifetime requires historical data detailing all customer transactions across all areas of the business. You cannot just depend on the dates of the sales transactions, especially when the sales cycle requires numerous steps each time the customer is engaged. At the same time, there may be other customer touch points that indicate engagement (such as calls to the call center), while monitoring of service usage could signal a reduction in use, signifying an imminent disengagement. In each case, coming up with a definition for customer engagement and determining what data is necessary to confirm that a customer continues to be engaged requires both defined policies and directed data management effort. The same can be said about any of the other variables: What are the actual costs of acquisition? Does that include the costs of manufacturing the product, marketing, and general administrative costs allocated to each customer? What are the ongoing servicing costs? Do those include specific service and maintenance activities for each customer, or do we also allocate part of the infrastructure charges (such as placing a new cell tower) to service costs? Customer lifetime value is a very powerful concept that can help drive specific actions, both strategic and tactical. And you can see that while the concepts are straightforward, the devil is in the details. There can be a lot of value in assembling a program to calculate and monitor customer lifetime value, but

t ll t t lif .

i t ill

i i tt ti

li f t

t l

PRE I ING AN USING US IMPR VE PR I ABI I Y

MER I E IME VA UE

1 WHY C LCUL TE CUSTOMER LIFETIME V LUE? 1.1 Th V l

i

m
i

L y ly
ti f t i ti i l f t i i Ri t lif ti l t i it t t t t li i i it tf lt i f t t iff i l ll i i t t t i t t t t

l i t R l ti l t li t f . i i lt , t i t l i i .I t t t ll t iff t t i f iff . i iff i l t f t i i t t S i t iff .

illi i M ti RM t t i j ti f i ll t i f ii i t f i t , i i l. iff t t t i ti , t t t f t i t t ,i t t i . t t ti ti it tf

fit ilit , it i i lif ti l f ti t t i

f l RM il iti t i t fi t l . l l ti t l t i l t i. ., li l t t i t i l t t t i iti . f

1.2 Th B l dD l m
A ti f

b w

i ii

,R

RM i t t i ft t i t t i t t t i i t i A i . N t it i ; a 5% increase in retention per G annum will double your customer base in 14 years, assuming all else is equal.

, RM i ti t t t ti lif ti l t l t t i it

ilit f t t t j t .U l l i f t i t, it i iffi lt t i tif t i .I it l t t t t t ti t i iti .

ti

fit

t i l

Too often, C strategies are justified on the basis of the simplistic generalisation that: It costs 5 times more to recruit a new customer than it does to retain an existing one ot only is this factually incorrect for most organisations, it is largely irrelevant. The important question to ask is: How much it is worth spending to recruit each new customer compared to retaining or develop existing ones, in order to optimise profitability? Customer lifetime value is the measurement that provides the answer to this question.
1. .1 e Value of Customer Retention

Customer retention happens in most companies with virtually no special effort or expenditure on retention marketing or business processes. In planning a retention marketing strategy, the questions to be addressed are:y

What increase can I make in customer retention through active intervention? y What is that retention going to be worth to the bottom line? , y Which customers should we be targeting for increased retention? , What do I have to spend on marketing or business y processes to get it? Lifetime value predictions for customers provide the basis for answering these questions.
1. . e Value of Customer Development

Similarly, customer development can also be evaluated in terms of changes to predicted revenue, when varying factors

such as cross-sell, up-sell, referral rate, frequency of purchase, etc.

1. .3 e Value of Prospect Acquisition an Conversion

Lifetime value can also be calculated for prospects, using predicted prospect conversion rates and subsequent customer retention rates to map out the likely revenue from each prospect. By varying prospect acquisition rates, cost-per-sale values, etc., the changes in lifetime value of prospects can be evaluated.
1. .4 ptimising uture Revenues

Comparing the effect on lifetime value by changing the criteria which determine prospect acquisition, customer retention and customer development, enables proper comparisons to be made between the profitability of different marketing strategies. Calculating lifetime value for customers and prospects makes it possible to plan and implement differentiated marketing activity for different customers based on their value (and potential value) to the organisation. Different tactics can then be used to maximise return on investment.
2 A DO E EAN YCUSTOMER I ETIME VA UE

Customer lifetime value is a summary of the net present value of future contributions to profits and overhead that a company expects to make from a customer, from the present time until the customer ends the relationship. et present value can be thought of as simply adjusting future cash flows into todays monetary value. Lifetime value is determined by predicting the various future Events throughout a customers lifecycle with the organisation, and allocating the relevant revenues and costs to those Events.

ENE ITS OF PREDICTING AND USING CUSTOMER

LIFETIME VALUE Measurement Action 3.1 Ne

y

ithout action = A

aste of money

ithout measurement = A am le

Customer Acquisition

Define the optimum balance between acquisition and retention marketing activities, based on comparative payback to the business. y Set value thresholds for acquiring prospect data, such as list rentals, data lease, list swaps, affinity partner data exchange, etc. ecruit more prospects which look like high value y customers. y Set customer acquisition budgets to optimise profitability over a defined payback period. y De-select prospects from campaigns where their lifetime value falls below profitable payback levels. y Allocate prospects to different sales channels based on their lifetime value. y oute enquirers to different call handling resources in call centres based on the value. y Evaluate the cost-benefit of using third party data (geodemographics, lifestyle data, business data overlays, etc.) in terms of that datas ability to enhance the differentiation of prospect values. se What If scenarios to identify the payback of y increasing prospect conversion rates. y Put accurate values on customer referrals set budget thresholds for profitable recruitment through this channel.
3.2 Customer Retention
y

Differentiate C

activity according to customer value.

se What If scenarios to identify payback for increases in purchase frequency and/or customer retention from direct sales as well as from indirect causes, such as improvements to customer satisfaction. y Ensure retention of high value customers where they are under competitive threat. y Allocate customers to different sales channels based on their lifetime value. oute customers to different call handling resources in call y centres based on the value. y Create Share of allet models to compare each customers lifetime value with the customers potential value with the organisation. y Profile high value customers look for more people with those characteristics Identify product portfolios of high value customers and protect / promote these products 1. y Track customer lifetime value over time identify trends in customer segments and segment migration. y se customer lifetime value at individual customer level for improving decision making by Customer Services for goodwill payments, reimbursements, upgrades, etc. y Identify prime targets for loyalty scheme membership, or customers suitable for loyalty scheme membership upgrades in advance of their transaction thresholds. y Aggregate individual customer values to company totals to create company lifetime values.
y

3.3 Cross-Sell & Up-Sell

Identify revenues to be gained from specific upgrade and cross-sell actions both for internal products, as well as third party or affinity products. y Identify revenue growth through increasing frequency of purchase. y Compare potential increases against share of allet estimates at customer or customer segment level.
y

3.4 Campaign Management

se customer lifetime value as a decision criterion in campaign selection or de-selection. y se customer lifetime value for automated decision making in lead management and campaign automation tools. y Track the change in values over time of customers and prospects within campaign cells compared to control cells.
y

3.5 CRM Operations / Data ase Marketing / Data ase Management

Provides the business case for implementing a C strategy. y Track the future value of the total customer base over time it is effectively the value of the company! y Provides the means to target and reward Customer anagement staff in C orientated businesses, where customer value, rather than product sales and product revenues, are appropriate measurements of success. y Identify customer groups within the customer base responsible for changes in overall database value isolate the factors causing the changes and take action to either address problems or exploit opportunities. y un What If scenarios to test the effect on customer value where:ew competitor products or services are launched o which are likely to steal customers or Share of Customer. o Withdrawal from existing products or markets. o Enter new markets or launch new products. V
y

Calculating LifeTime Value Lifetime Customer Value (LCV)

(LTV)

or

Jim's Intro: There are two kinds of LifeTime Value measurement - absolute and relative. The first is very difficult to calculate; the second, very easy to calculate and in many ways more powerful than the first. The most difficult part of calculating LTV is deciding what a lifetime is. LifeTime Value is the value of the customer over the LifeCycle (if you don't know what a LifeCycle is, you really should read the article on LifeCycles before reading this one). Lifetime Value doesn't exist without a LifeCycle. We will get into some details on calculating LifeTime Value in a moment, but first, a clarification. The LifeTime Value concept has been horribly abused and misunderstood over the last several years. It is not necessary to figure out an absolute LifeTime Value for a customer or wait "a lifetime" to find out the value to use the concept in managing customer value. If you are new to this LifeTime Value stuff and have not tracked the appropriate parameters, or your company is new and lacks meaningful operating history, you can look for "relative LifeTime Value," link it to customer behavior, and still get leverage from using LTV / LCV in your business model to manage customer value. Here's a very simple example. Say I run the same ad in two different newsletters and get response from both. When I look at these responders, maybe a week later for a content visit or 30 days later for a purchase, I find a high percentage of repeat visitors or buyers from one newsletter, and a low percentage from the

other. Repeat behavior indicates higher LifeTime Value, and predicts future repeat behavior, regardless of what the actual monetary LifeTime Value is. I can switch money out of the low repeat newsletter into the high repeat newsletter and get higher ROI without having to measure anything but repeat behavior. By the way, using customer behavior to predict the relative LifeTime Value and loyalty of customers is a 40 year old technique still used by mail order and TV shopping companies today. Large sites with CRM analytics are using this technique, known as RFM, to predict customer value and response to promotions. If you'd like to see more details on using relative LifeTime Value to make ad or product decisions, see the tutorial: Comparing the Potential Value of Customer Groups
Comparing Value of Customer Groups
Jim's Intro: This short tutorial will provide you with all the information you need to implement a simple potential value scoring p rocess. Use these scores to compare the potential value of customers from various ad sources, buying certain products, or visiting specific areas of your site. You will also learn how the concepts of Customer Lifecycles, LifeTime Value, and ROI fit into scoring the potential value of customer groups. If you can compare potential value, you can allocate more spending to higher value customer groups. This article assumes you have some background in customer marketing; you might want to read the short articles on customer profiles and customer models if you don't.

the

Potential

Over the past five decades, a lot of research and testing has been carried out concerning the profiling of customer behavior based on transactional data. The appearance of computers and "data -mining" have allowed even more extensive studies to be carried out. The end result? If you had to pick one variable to predict the likelihood of a customer to repeat an action, Recency, or the number of days that have gone by since a customer completed an action (purchase, log -in, download, etc.) is the most powerful predictor of the customer repeating this action. As each day goes by after the customer completed the action, the customer gets less and less likely to repeat it. Plain and simple. You can run all the fancy data-mining scenarios on "likelihood to buy" or "likelihood to visit" you want to Recency always comes up as the most important v ariable in predicting the likelihood of a customer to repeat an action. Recency is the number one most powerful predictor of future behavior. The more recently a customer has done something, the more likely they are to do it again. Recency can predict the likelihood of purchases, log -ins, game plays, just about any action-oriented customer behavior. Recency is why you receive another catalog from the same company shortly after you make your first purchase from them. They know you are most likely to or der again immediately after your first order. Recency is the most powerful predictor of future behavior. It should not surprise you that Recency is also the most powerful predictor of a customer to respond to a promotion - after all, the more likely a cus tomer is to repeat an action, the more likely they are to respond to a promotion asking for this action (purchase, log-in, download, etc.). If a Recent customer is more likely to repeat an action, and is more responsive to promotions for this action, it follows the more Recent a customer is, the higher their potential value, because Recent customers are the most likely to contribute to profits in the future by responding to your promotions (or simply just coming back by themselves).

Customers who are more Recent have higher potential value than customers who are less Recent, for any given activity. Customers who made a purchase 15 days ago have higher potential value than customers who made a purchase 60 days ago. Customers who logged in last week are much more likely to visit than customers who logged in 30 days ago, and so have higher potential value. Make sense? Great. But how is Recency implemented, how do you actually do anything with this information? Glad you asked. Let's use Recency to compare the potential value of customers coming from two different ads (Ad #1 and Ad # 2) that ran at the same time, for the same duration. The following example uses a spreadsheet, but if you know your way around databases, and can query your customer records, then have at it your way. 1. Identify the groups you want to compare for potential value. In this example, it's the customers who clicked on either of two ads, Ad #1 or Ad #2 (two groups). 2. Decide which activity is most important to you for these grou ps. If you're a publisher, probably log-ins or page views are most important. If you are selling merchandise, you would use purchases. For this example, we will use purchases. An example using visits (or log-ins, if you don't track visits) is below. 3. Import all the purchase records of people who clicked on Ad #1 or Ad #2 into separate spreadsheets. These transactions need to have a date; most interactive activities are date-stamped so this should not be a problem. If an activity you want to profile for potential value has no date stamp, start collecting the dates of activity. 4. Pick a time frame to look at Recency. For page views, it might be 1 week; for purchases, maybe 30 days. The exact length is not very critical, because you are interested in comparing the activity between the Ad #1 and Ad #2 groups you want to know which is "better." As long as you use the same time frame for both groups, you are fine. Pick something reasonable based on what you know about your customers. Anywhere from 30 to 90 days would be reasonable for purchases; let's use 30 days.

5. Sort the purchase records for Ad #1 from most Recent to least Recent and find out what percentage of the people who clicked on Ad #1 and made a purchase have made at least one more purchase in the past 30 days. Count back 30 days using the transaction dates, total the number of customers making a purchase, and divide by the total people in the spreadsheet. Perhaps it is 20% . Note: The software that comes with the book will automatically aggregate multiple transactions by customer and sort customers by their most Recent transaction for you). 6. Run the same analysis for people who clicked on Ad #2 and made a purchase. Let's say only 15% of these people have made at least one purchase in the past 30 days. 7. You're done, and you know the answer. A higher percentage of people who clicked on Ad #1 are Recent - active and purchasing - when compared with Ad #2. This means Ad #1 generates customers with higher potential value. You need to take this into accou nt when analyzing the success of the ads. Do you understand how powerful this idea is? If you go through this process for customers grouped by product they bought first, you can determine which products generate new customers with highest potential value. Go through this process for customers grouped by which area of the site they visit most, and you will find which areas generate highest potential value customers. If you go through this process for customers grouped by the demographics or the survey data they provide, you can determine which data points define customers with the highest potential value. This is a simple example of how companies with experience in managing remote shopping customers find ways to maximize sales and minimize expense. The customers, through their actions, tell them which route is the most profitable to

take. The most Recent customers for any particular activity are always the ones most likely to repeat that activity, and so have a higher potential value. You can track multiple activities for the same customer groups. In the first example, you found customers who clicked on Ad #1 and made a purchase are more Recent on purchases, so they have a higher potential value on the activity "purchases." But what about the Recency of people who clicked on the ads for visits? If they keep coming back, they could be of some future value. Let's see how this Recency study might look. 1. Visits / log-ins example: Import all the visits (or log-ins if you don't track visits) into two separate spreadsheets of people who clicked on Ad #1 or Ad #2 (need date stamp). 2. Pick a Recency cut-off. Again, we are interested in a comparison, so the number isn't critical. Let's use 1 week. 3. Sort each spreadsheet from most Recent to least Recent and find out what percentage of the people who clicked on Ad #1 have visited (logged -in) at least once in the past week, as was done above for purchases. You might come up with 10%. 4. Run the same analysis for people who clicked on Ad #2. You might come up with 30% who have visited / logged -in at least once in the past week. 5. You're done, and now you have an interesting situation. It appears the customers who clicked on Ad #1 have a higher potential value on purchases, but people in general who clicked on Ad #2 have a higher potential value on visits. Maybe they're just tire kickers, or maybe they're doing research. We'll take a closer look at finding answers to this situation on the next page of the tutorial. Note that this method is based on the actual facts of customer behavior - not speculation or "best guess" theories. The behavior of the customer is the most accurate yardstick you will find for assessing potential value. Once you complete studies like these, you can begin to organize all your business practices around the potential value of the customers they generate. If

you allocate money away from activities generating low potential value customers, and allocate this money to activities generating higher potential value customers, you will become more profitable over time.

Let's say you're not satisfied with using relative LifeTime value as a proxy for absolute LifeTime Value. You're a glutton for punishment, or your boss wants a hard number. No problem. Here are a few issues we need to put on the table when discussing the calculation of LTV: 1. If you haven't been in business long enough to know the Lifetime of a customer, just put a stake in the ground by looking for defected best customers. Look at customers who have spent or visited the most with you and then of these, look at the ones who haven't made a purchase or visit in some time (6 - 9 months, for example). In all likelihood, the last purchase or visit was the end of the LifeCyclewhen considering best customers who have stopped buying or visiting. When best customers stop, they're usually all done. Then look at first purchase or visit date for these customers, calculate your Lifetime, and use this length of time as the "standard" customer LifeTime, realizing the average lifetime is probably much shorter. 2. Frequently, a customer will defect for a few years and then come back. This is cool, and normal. Their life changed somehow and they left, and now they need you again. Most offline marketers would call a customer who has had zero activity for over 2 years a defected customer. Online, it's more like 6 months for the average customer, unless you are in a classic seasonal business. If the customer starts up again,

they would be a new customer, for marketing and modeling purposes. They will more likely behave like a new customer than a current customer. The behavior will ramp and fall off all over again, just like it did in their previous LifeCycle with your business. That doesnt mean you cant use the same customer number, or combine the old behavior record with the new behavior record in the customer service shop. In fact, knowing how long on average a customer defects before they come back can be a useful promotional tool. But there has been a significant break in behavior, and this customer is more likely to behave as a new customer than a customer who has been with you the whole time. Thats just the way it works. Theyre likely to be interested in different products, for example. You decide if its a new lifetime or not based on your business. In most cases, from a marketing perspective, and for the purposes of LifeTime Value, they should be treated as a new customer. Otherwise, all your customers will have infinite lifetimes, and you lose the relevance of the metric. 3. Another challenge to calculating LifeTime Value: usually much of the data you need to complete the simple calculation are not available, or can't be agreed upon by all the players, especially if you are in a big company. If you don't know what the average unit returned costs you in terms of overhead, you can't do the calculation. If you don't know what the average number of customer service calls per unit shipped is and what the calls cost, you can't do the calculation.

This is a particularly difficult problem for offline retailers, who don't have a database that captures nearly enough relevant data. Here's one way approach it if the operational data you need is unclear. Try to focus on the average unit sold, and break up all the revenue and cost components that comprise the unit. Once you get to a profit / unit, just multiply by units sold to a customer over the "lifetime," minus overhead and promotional costs, and you get LTV. Average price, cost of goods sold, gross margin...should be easy to find. To get customer service costs, look at how many units you move annually, and divide by annual customer service cost. Do the same thing for returns, and so on, until you know the costs / unit sold of all the elements going into a sale. Don't forget credit processing, after sale support, etc. For example: Net Profit per Unit Analysis: Average Sale Price Cost of Goods Sold Gross Margin Credit Clearing Revenue Ship & Handle Cost of Ship & Handle Call Center $40.00 100% (36.00) (90%) 4.00 (.80) 6.00 10% (2%) 15%

(4.00) (10%) ( .80) (2%)

(1 call every 5 sales) Returns and Processing (5% of Sales) Fraud / Merchandise Loss (1% of Sales) Promotional Costs / Discounts / Ads (2.00) (5%)

( .40)

(1%)

( .80)

(2%)

Net Profit per Unit

$1.20

3%

LTV Calculation and Customer Acquisition Cost Calculations: Say the average customer buys for 2 years, then stops for at least 1 year. Therefore, we define the LifeTime of a customer as 2 years. Over 2 years, the average customer makes 16 purchases. 16 x $1.20 Profit per Unit = $19.20 LTV of the average customer The average customer recruits 3 other customers. The maximum acquisition cost of a new customer should be 4 x $19.20 = $76.80 to breakeven. By the way, I'm not a fan of including pass-a-long or referral customer value in an individual customer LifeTime Value. If you do, what are the pass-a-long

customers worth? It's double counting. Use it to look at acquisition costs as in the example above, but don't include it in LTV calculations. The sum of all your customer LifeTime Values should equal your future profits; if you include the value of pass-a-long customers in LifeTime Value, you will over estimate profits. Don't be surprised if you find some customer groups have negative LTV's its very common. This is the part of LTV analysis usually forgotten, because it literally means you would be more profitable if you had fewer customers. And explaining that to your boss (if you have one) is often a challenge, even on a positive day. Good luck! After measuring customer value, the next step is to manage customer value - to make money by creating very high ROI customer marketing campaigns and site designs. The Drilling Down book describes how to easilycreate future value and likelihood to respond scores for each customer, and provides detailed instructions on how to use these scores to continuously improve the profitability of your customers.