Value or Return on Investment
More and more companies are starting to use Lifetime Value as a means of predicting the success or failure of new marketing strategies before serious money is committed.
In the past, return on investment was what everyone used. ROI is neat and easy to understand. You invest so much in a promotion, and you measure the response. You calculate the net profit from the sales to the respondents, and divide by the amount invested. The result: return on investment. For example, you might invest $40,000 in mailing a 100,000 piece promotion. If you have a 2% response rate, and sell 2,000 items at $100 each, with a net profit of $50 each, you will have a gross profit of $100,000. Subtracting the $40,000 invested in the mailing from the $100,000 gross profit, your return is $60,000. Your return on investment is 1.5 which is very respectable, and understandable.
ROI is clearly the way to measure the immediate result of any direct marketing effort. But with database marketing, the marketer is able to consider more factors and project his profits over a series of years, rather than just looking at the return from a single campaign. The reason why Lifetime Value has proved to be a more sophisticated measurement is that the LTV calculations include such factors as the retention rate, the referral rate and the (long term) spending rate, rather than just the response to an immediate promotion. We are judging our marketing efforts by our ability to retain and increase the loyalty and spending rate of our customers, rather than just considering the result of a single investment. Lets see how this is done.
Here are a set of lifetime value charts adapted from the work of Andy Kardos, President of BCP Direct in Toronto. Andy has worked with a number of different clients. This chart is based on experience he had with several retailers. In this simulation, the retailer was able to keep track of her customers through a proprietary card, so she had a pretty good database built up. She had more than a million customers. She knew the spending habits of the customers who used the card (about 40% of all sales - or about 400,000 customers). She decided to experiment with a long term loyalty program. To do this, she set up a test group of 5,000 customers. The program involved personalized mailings, special offers, an incentive to refer new customers and other special benefits. The program cost about $12 per person per year. To determine whether her program really paid off, she set aside a control group of 5,000 customers who did not receive any of the loyalty program communications or benefits. She compared the results over a three year period.
Chart A shows the lifetime value of the customers in the control group (those not in the loyalty program). Their lifetime value, in the third year, was $162.91. Their initial retention rate was 73% in the first year and 79% in the second year, Their spending rate was $138 and $149 and $161 in the first three years.
In Chart B we look at the lifetime value of the 5,000 customers in the loyalty program. The loyalty program has had a measurable effect on the customers enrolled in it. In the first place, their retention rate increased to 79% and 84%. This added 434 customers who would otherwise have disappeared in the third year. These retained people spent $78,988 alone in the third year. In addition, the referral program really paid off. For a net referral incentive of $15 per customer, in the first year, 6% of existing customers encouraged their friends or relatives to become customers of the retailer. This added 300 new customers in Year 2. Furthermore, the loyal 4,250 customers who were still shopping in Year 3 increased their referral rate to 8%, bringing in a net 340 new customers. These 640 new customers (the referrals over two years) made all the difference in the loyalty program. They brought in $116,480 in sales in Year 3.
The loyalty program also affected the spending rate. It increased to $162, $175 and $182 a big jump over the spending of the control group. Why did these people spend more? Because they were being treated a special people. They were recognized, communicated with. They felt a new special relationship with the retailer. The results could be seen in the numbers on the chart.
To accomplish this gain in spending and retention, the retailer had to spend $12 per customer per year. This was spent on communications, a special 800 number for this favored group with special treatment when they called, member nights and a host of other benefits which amounted to 6.6% of total sales revenue from this group in the third year. Was it worth it? Can the loyalty program be justified? This is where Chart C comes in. In this chart we compare the lifetime value of the control and the test group. In the first year, the LTV of the test group is actually lower than the control group. Why? Because the extra $12 per customer per year spent on loyalty maintenance exceeded the gains from increased retention and spending.
Looking at the program over three years, Andys client was working to build long term loyalty. She used this program to develop a core group of loyalists who spent far more than the customers in the control group. The net gain, $24.28 per customer per year in the third year is really outstanding when you consider that this figure represents profits after all of the expenses of the loyalty program have been included. An early version of Andys results were published in the (Canadian) Direct Marketing News, which, incidentally, is one of the best sources of information on direct and database marketing around.
The payoff of this type of analysis is that it shows what would result from widespread adoption of the loyalty program. If you look at the first year alone, you might say, "Dont consider a loyalty program. It is a loser. LTV does not increase." The return on investment is zero. In the second year, the ROI is much more satisfying. Increased customer loyalty produced a 79% return on investment in the loyalty program. But in the third year, the results are outstanding. We have a 102% return on investment. Increased customer loyalty if you can generate it can be one of the most powerful ways to boost your bottom line better than sales, discounts, coupons, cash back, or mass advertising.
To understand the full implications of the loyalty program, we have to consider the last three lines of Chart C. In line C4, Andy has made the assumption that the retailer could build on the experience gained in the test, and enroll 200,000 in her loyalty program. The 200,000 represents half of the 400,000 regular card users. Each of these 200,000 customers will cost $12 per year for loyalty maintenance. Based on these numbers, the retailer would not gain anything in the way of increased profits from the loyalty program in the first year. Without lifetime value of analysis, such a program would be rejected by most retail chain managements. But looking ahead to the third year, the program would bring in more than four and a half million dollars in net profit after paying $2.4 million for the loyalty program in that year.
What does this $4.8 million dollars really mean? It means that spending to win the hearts and wallets of our customers can have a measurable payoff.. Short term ROI analysis conceals the long term gain. The test group responded. The beauty of this type of analysis is that we have been able to learn the value of the loyalty program without making a costly mistake. We dont have to take out full page ads and tell everyone we are starting a loyalty program, only to fall on our face and withdraw it a couple of years and a couple of million dollars later. By setting up control and test groups, we can judge the effectiveness of our strategies before serious money is committed.
Of course, there is the drawback that it took Andy more than two years to get this information. Management is usually impatient. They want to see results tomorrow, or at least this quarter. What management groups are prepared to wait two years for test results? Not too many. So how can we use LTV as a strategy evaluation tool?
My advice is this. As database marketers, we have to show results. We have to have monthly and quarterly promotions. We have to use RFM analysis to find our most responsive customers. We can measure the effectiveness of these things by ROI analysis. We should keep doing this. But we also need longer term programs. We have to try newsletters, member nights, gold cards, frequent buyer programs, improved customer service etc. The impact of these programs is longer term. The payoff of loyalty building programs can only be measured by analysis such as lifetime value which takes into account retention rates, spending rates, and referral rates. The analysis has to include the cost of the loyalty programs along with the benefits. LTV permits us to pursue the day to day marketing programs but also to set aside test and control groups and measure the impact of our long term loyalty building activities.
There is another aspect to use of LTV to drive strategy. Up to now, we have been considering a loyalty program for all customers. We made no effort to segment our customer base in creating the test and control groups. But once we have run the test, we should do some serious segmentation. Some of the customers responded much better than others. The increased spending, retention and referrals did not come from all customers. The real results probably came from less than half of the customers stimulated by the loyalty building efforts. What can we learn about customers that will help us to segment them into profitable and less profitable quintiles so that we can devote even more than $12 per year to the best customers, much less than $12 per year to the worst customers, and about $12 per year to the average customers? This type of segmentation can really increase the profits over those shown in these simple charts. Customer segmentation methods represent the next step up, once a company has mastered lifetime value analysis. It takes LTV to an entirely new level.