How's Your Customer/Cost Benefit Ratio?
Like this article? We recommend
Like this article? We recommend
Like this article? We recommend
Do you know how much each customer you have is worth?
Do you know how much each customer you have costs you?
No? Well, let me be honest: Most of us never will know about any individual customer's cost/benefit ratio, but we certainly can know overall what our customers are worth and what they cost us to get in the first place.
The dot-coms taught us a lot about customer acquisition costs. During the height of the online business frenzy of 1999–2000, online companies were paying $40 or more to gain a new customer (in advertising costs, free shipping, and price discounts) who bought a total of $15 worth of goods with a 30% profit margin. That meant these customers cost the business a net loss of $35.50 each ($15 sale-10.50 (70% cost of goods X 15 = 10.50)-$40 promotional cost = -$35.50). Working with hundreds of thousands of new customers, these numbers do not take long to consume a LOT of venture capital.
The plan was to make up for these initial losses through "lifetime value" of customers. It could have worked, too, but for the reality of a competitive marketplace. Those who were attracted by the discounts tended to be "distracted" by the discounts offered by competitors who seem always to appear in an unending stream. The situation was great for customers, but not so great for the businesses caught up in the frenzy.
Lifetime customer value is a great concept—when used properly. However, we really do have to be careful that we do not engage in wishful thinking and hoping, but instead, we need to build our promotional strategies on real data and cautious projections.
So, what e X actly do I mean by "lifetime customer value?" Customer value to a business refers to the profits you gain from the customer's purchases, so lifetime customer value is the total profits you gain from all the purchases made by a customer during the "lifetime" of their relationship with your business. For e X ample, suppose you operate a restaurant that has a customer, Sam, who eats with you three times a week at lunch, and he does that nine months out of a year. That means Sam brings you $813.37 per year in sales (3 meals per week X 4.33 weeks in a month X 9 months X $7 per meal = $813.37). If your average net profits run 35% before fi X ed costs, Sam is worth $284.68 per year to you in gross profit ($813.37 X .35 = $284.68). If we suppose that Sam continues this pattern of purchases over a period of five years, his lifetime value would be $1,423.40 ($284.68 X 5 years = $1423.40).
Of course, at the opposite end of this spectrum is the one-time diner, Joe, who buys one lunch at $7 and then never returns. Your lifetime value for Joe is $2.45 ($7 X .35 = $2.45). Obviously, if you enjoy lots of Sams among your customers, you can easily afford to invest lots of money in promotions that attract these customers. On the other hand, if you have mostly Joes, practically any promotion you buy will result in paying more to acquire a new customer than that customer is worth to you. At the risk of stating the obvious, you really need to know what kind of customers you have before you can decide how much is too much to pay for promotion efforts designed to attract new customers.
The problem is, how to figure out your relative proportion of Sams and Joes? The implications for how you go about promotion are obviously quite different in each case. In the case of Joe, you must find a promotional method that brings in a steady stream of such one-purchase customers, and the cost per sale you spend on that promotional method must be less than $2.45 per sale or else you will lose money every time you make a sale.
On the other hand, in the case of Sam, you can afford to spend quite a lot more on getting the initial sale from Sam, even taking a significant loss on the initial sale because the cumulative value of Sam-type customers will be quite large over the life of your relationship with that customer. I will suggest that you ought not to get carried away and spend too much on promotion, even in trying to land the Sams of your market. A good rule of thumb in most cases would be to make sure that your cumulative profit is positive by the third sale or by the end of si X months, whichever comes first in your business.
Now, back to our problem: How to determine the proportion of Sams and Joes in your business? The very best way is to track each transaction by customer, analyze your database to determine what proportion of your customers bought once and what proportion are repeat customers, and then compute directly the value of each repeat customer. But alas, most small businesses, and many large businesses too, for that matter, have anonymous transaction systems. Restaurants, grocery stores, convenience stores, laundro-mats, craft stores, bedding stores, and many others share the same problem: They do not know who is buying what from their businesses over time.
By contrast, many service companies—heating and air conditioning services, dentists and doctors and chiropractors, law firms and CPA firms, lawn-mowing services, barber shops and hair salons, carpet cleaning services, and house painters—have all the data they need to determine who bought what from that business over time. Most do not use these data very well, but the data are usually there.
As we consider some of the complications in performing this kind of analysis, we could easily become confused and find that our efforts are wasted, so let me suggest some reasonable parameters that can help you to simplify the process I am about to suggest. First, I would suggest that if your customers return only after two years or more, you will starve to death waiting for repeat business, and so you must treat customers as one-purchase-Joes. That means you must make every sale a profitable event, after considering the promotional costs that produced the sale. If that describes your business, you are done here. Make sure that every transaction is a profitable one. That probably rules out "buy-one-get-one-free" and 75%-off "customer appreciation sales events."
If, however, your customers return several times within a year, you really do need to get a handle on their "lifetime" value. For this analysis, though, I suggest that you consider their "annual" value, that is, how much do you make from them over the course of a year? In such businesses, although a customer might well return many times over many years, for promotional purposes, you probably will not be well served by investing more in promotional costs than you can e X pect to recover within a single year. As I suggested earlier, a si X -month payback is probably close to a practical limit.
What you need to know is how many "repeat-purchase-Sams" you have among all your anonymous customers who bought from you in the past 12 months. So here is how you can find out: Ask them. No, you do not have to ask all of your customers. You can ask just a few, and you get a surprisingly good idea of the makeup of your customer base. Statistics really is a wonderful science! The larger the sample you use, the more accurate will be the result, but beyond a hundred or so, the results will not change in any meaningful way. In fact, if you will get just 10 surveys a day for a week, that will get you more than 90% accurate results.
You can ask as many questions as you like, but I suggest that you ask at least the following two questions:
During the past year, appro X imately how many purchases have you or any members of your family made from this store?
- 1
- 2
- 3
- 4
- 5
- 6
- 7
- 8
- 9
- 10+
On average, what did you spend on each purchase you or your family member made from this store?
- $0–5
- $6–10
- $11–20
- $21–30
- $31–40
- $41–50
- $51–60
- $61–70
- $71–80
- $81–90
- $91–100
- $101+
You will need to adjust the response choices to fit your business, of course. If a typical customer might visit your business several times every week (donut shop, convenience store, lunch diner, newspaper stand, and so on), just increase the numbers of each choice. The idea is to capture the full range of customer behaviors. You always need "1" because you will survey your customers at the point of sale, and this could always be their first purchase.
To capture spending patterns, you may also need to change the amounts you list as choices. If you sell convenience store items, your biggest sale ever might be less than $50, with most falling into the range of $10–20. If you sell computer or car repair services, though, your minimum purchase might be $50, with sales ranging up to several thousand dollars. Adjust the dollar amount of the choices you offer, and make sure you give around 10 options from which your customers can choose. This will allow you to get a more detailed understanding of your customers and their buying habits.
Okay. As an illustration, let's assume we have collected data for a couple of weeks. We tried our best to get surveys at all different times of the day and on all days of the week in which we are open. Below are the results we found from 140 surveys we conducted.
Question 1
|
Response: Number of Purchases During Previous 12 Months |
Number of Times this Response was Chosen |
% of Total Responses |
|
1 |
53 |
38 |
|
2 |
35 |
25 |
|
3 |
23 |
16 |
|
4 |
7 |
5 |
|
5 |
4 |
3 |
|
6 |
2 |
1 |
|
7 |
0 |
0 |
|
8 |
3 |
2 |
|
9 |
0 |
0 |
|
10+ |
13 |
9 |
| Total | 140 |
|
Question 2
|
Response Average Amount of Typical Purchase |
Number of Times this Response was Chosen |
% of Total Responses |
|
$0–5 |
23 |
16 |
|
$6–10 |
13 |
9 |
|
$11–20 |
53 |
38 |
|
$21–30 |
30 |
21 |
|
$31–40 |
6 |
4 |
|
$41–50 |
2 |
1 |
|
$51–60 |
0 |
0 |
|
$61–70 |
2 |
1 |
|
$71–80 |
0 |
0 |
|
$81–90 |
7 |
5 |
|
$91–100 |
1 |
1 |
|
$101+ |
3 |
2 |
|
Total |
140 |
|
Combinations: Total Purchases During the Previous 12 Months
|
Purchase Amount Per Customer (Response to Question 1 X Response to Question 2) |
Number of Times this Purchase Level Was Reported |
% of Total Responses |
Weighted Average Computation Amount* X % |
|
$0–5 |
18 |
13 |
45.00 |
|
$6–20 |
28 |
20 |
364.00 |
|
$21–50 |
51 |
36 |
1836.00 |
|
$51–100 |
18 |
13 |
1350.00 |
|
$101–200 |
13 |
9 |
1950.00 |
|
$201–300 |
7 |
5 |
1750.00 |
|
$301+ |
5 |
4 |
1505.00 |
|
Total |
140 |
100 |
8800.00 |
|
Weighted Average Annual Customer Purchases |
|
|
62.86 |
Okay. Now we have an estimate of the typical annual revenue we get from each new customer: $62.86. If our profit margin is 50% before promotional cost, each new customer is worth an average of $31.43, less whatever we pay to get them to become our customer. If we pay more than this amount to gain each new customer, we are probably losing money on every new customer we get—not a smart bargain.
After you have this information about your customer behaviors, you can then move on to develop a promotional plan that recognizes these limits. Every promotional tactic you adopt that is designed to bring in new customers ought to be analyzed carefully to determine what that tactic cost you for each new customer it brought in. To determine which tactic produced what customer is not always easy, but you can simply ask every customer, or a sample of customers over time, a couple of questions during their checkout process to help you figure out what brought them in.
Remember that each ad or promotional tactic should have a specific purpose: to bring in new customers, to increase the average sale, to encourage repeat purchases, to reassure buyers after they have bought, to encourage customer referrals, etc. This article focused on those advertisements and promotional tactics that are designed to bring in new customers. Each kind of promotional tactic should be analyzed separately to determine how well it is working in achieving the goals you have for that particular tactic. If you use the kind of systematic, data-driven analysis presented in this article, you will be much more certain that you are spending your precious promotional cash on tactics that improve your bottom line—and that, after all, is the bottom line.
To contact the author: tbergman@tkb.com.