Unit Economics

Part B: Understand the Money

Better questions create better lives. In business, one of the better questions is what a customer really costs you.

Say you run an electrical contracting company, and someone tells you to spend more on advertising because every new customer is worth it. Before you say yes, you want to know whether that is true, because by the time a wrong answer shows up in the bank account the money is already spent.

That question is really about unit economics. It is a grand phrase for a plain idea: what happens, in money, when you sell to one customer?

Four numbers for one customer

The first is revenue per sale, what the customer pays you. The second is gross profit per sale, which is that revenue minus the direct costs of delivering it, the materials, labor and fuel that exist only because of that job. Revenue is what comes in. Gross profit is what you actually get to use.

The third is what it costs you to win that customer: the advertising, the tools and the time you or someone else spent, divided by the number of new customers you got. We cover how to measure this, including cost per lead, cost per customer and the fuller number called CAC, in the marketing budget lessons in Track 3, so for now a rough figure from your own records will do. And the fourth is customer lifetime value, which is the total gross profit a customer brings you over the whole time they buy from you.

Notice that the third number counts customers, not leads. A lead raises a hand and a customer pays, and the gap between the two is your close rate. If you track leads but not customers, you are guessing at the one that matters, which is why the CRM lesson later in this course is about keeping a record of who became a customer.

A worked example with made-up numbers

Say an electrical contractor gets a new customer’s first job at $400, and direct costs are $240. Gross profit is $160. Say the owner works out from the records that winning one new customer costs about $120 in advertising and time.

• Gross profit on the first job: $400 − $240 = $160, so the first job leaves $160 − $120 = $40 after paying for the customer.

Now the problem. Say the cost of winning a customer rises to $240, because the leads get a little colder and fewer of them close. On a $160 first job, that customer now costs you $80 more than you earned.

That is not necessarily a disaster, and this is where lifetime value comes in. If the customer comes back, the first job is only the start.

Lifetime value, and the caution that goes with it

Say that, in your own records, customers of this kind have averaged three jobs. Then lifetime gross profit is 3 × $160 = $480, against a cost of $120 to win them. That looks comfortable. Against $240 it still looks decent, but it is a much closer call.

Here is the caution, and I mean it. Lifetime value has to come from what you have actually seen: real customers, real repeat jobs, real dates in your records. It is very easy to imagine a customer who stays for years and refers their neighbors. It is much harder to prove one. If you have not been in business long enough to know how often people come back, use only the first sale for now and treat any repeat business as a bonus. A number based on hope will justify almost any ad spend, and that is how owners talk themselves into losses.

If repeat customers are the core of your model, the Sell to the Same Customer Twice lesson goes further into why it matters.

Before you move on

Take the last ten customers you won and work out revenue and gross profit for each, then make your best estimate of what it cost you to win them. Note how many came back for a second job, because that is the only lifetime value you can trust. If you cannot say which customers were new and which were repeats, that is your first finding.

Two to three hours if your records are reasonably clean, longer if you have to reconstruct them from invoices and text messages. When you reach the marketing budget lessons, the free CPL vs CAC Workbook on the Resources page takes the cost-to-win number further.