Most growth plans start with the top line. More customers, more orders, more markets. The assumption underneath is that more volume will, sooner or later, turn into more profit. Sometimes it does. Often it does not, and the business only finds out after it has spent a year and a lot of money learning the hard way.
The reason is simple. Growth does not change the economics of a single customer or a single order. It multiplies them. If each unit makes a healthy margin once every relevant cost is counted, growth multiplies a profit. If each unit quietly loses money, growth multiplies a loss, and it does so faster than the business can react.
Unit economics is the discipline of looking at the business one unit at a time, whether that unit is a customer, an order, a contract or a project, and asking a plain question: when we win one more of these, do we end up better or worse off? What it takes is a willingness to look at the numbers honestly before deciding to push.
What unit economics actually means
Start by choosing the unit. For a subscription business it is usually the customer. For a wholesaler or a webshop it may be the order. For a project business it is the project or the client relationship. Pick the unit that best matches how revenue actually arrives and how costs actually behave. If you are unsure, start with the customer, because that is where acquisition costs and retention meet.
Then look at three questions for that unit:
- What does it contribute? Revenue from the unit minus the costs that come and go with it. This is the contribution margin.
- What did it cost to win? Sales, marketing, discounts, onboarding and anything else spent to bring the unit in.
- How long does it last, and does it grow? How many months or years the customer stays, and whether they buy more or less over time.
Put those together and you have a picture of whether a typical unit pays back what it cost, how quickly, and what it is worth over its life.
Count the full cost of each unit
Gross margin from the accounts is a starting point, but it usually flatters the picture. It typically includes the cost of goods or direct labour, and leaves out a long list of costs that also rise when you take on more customers or orders.
A fully loaded contribution margin includes the costs that move with volume, even if they sit in other lines of the accounts:
- Freight, packaging and payment fees
- Returns, credits, warranty work and write-offs
- Customer service and support time
- Account management and the time spent on quotes and follow-up
- Installation, onboarding and training
- Discounts and rebates that are given in practice, not just the list price
Say a company sells equipment with a gross margin of 40 percent on paper. Once freight, installation, a share of the service desk and the typical discount are counted, the contribution per order might be closer to 18 percent. That is still a business, but it is a very different business from the one the leadership team thought it had, and it changes how much can sensibly be spent to win each new order.
The point is not to allocate every last cost. Rent for the head office does not rise with one more customer. The point is to include the costs that genuinely scale, because those are the ones growth will multiply.
Two further costs deserve special attention because they are easy to underestimate.
The first is the cost to acquire a customer. Add up what you spend on sales and marketing over a period, including salaries, commissions, campaigns, events, tools and introductory discounts, and divide it by the number of new customers won in that period. Many businesses have never done this calculation, because the costs sit in several budgets and nobody owns the total.
The second is the cost to serve. Not all customers are equally expensive to keep. Some order often in small quantities, call support every week, need custom documentation and pay late. Others place large, predictable orders and rarely need anything. On average they may look the same. In reality, one group may be carrying the other.
A simple way to see this is to ask the people who do the work. Customer service, logistics and finance usually know exactly which customers take the most effort. Combine their knowledge with a few rough measures, such as number of orders, number of support cases and days to pay, and you can sort customers into a few groups by cost to serve.
Payback period and cohort thinking
Once you know what a customer contributes each month or year, and what it cost to win them, you can calculate the payback period: how long it takes for the contribution to cover the acquisition cost.
Say it costs 30,000 kroner to win a typical customer, and each customer contributes 2,500 kroner a month once every variable cost is counted. The payback period is twelve months. If customers on average stay for four years, that is a sound investment. If half of them leave within the first year, it is not, because many never pay back what they cost.
This is why averages are dangerous. A single average figure for customer lifetime hides the pattern that matters. Cohort thinking solves this. Group customers by when they started, for example by the quarter they signed, and follow each group over time. For each cohort, ask:
- How many are still customers after six, twelve and twenty-four months?
- Are the ones who stay spending more or less than when they started?
- Did the cohort recover what it cost to acquire, and when?
- Do recent cohorts behave better or worse than older ones?
The last question is often the most revealing. If newer cohorts leave faster or spend less, it may mean the business is reaching for customers who are a poorer fit, perhaps because the sales push has widened the net. Total revenue can still be growing while the quality of each new cohort is falling. Without cohorts, that is invisible until it shows up in the profit.
Growth does not fix a weak unit. It multiplies it.
Why growth can make a weak model worse
It is tempting to believe that scale will solve the problem. Volume will bring purchasing power, fixed costs will be spread more thinly, and the margin will improve. There is truth in that for some costs. But there are several ways in which growth actively makes a weak model worse.
- Acquisition costs rise as you reach further. The first customers are usually the easiest to win. They found you, they fit well, they were referred. As you push for more, you reach customers who are harder to convince and less suited to what you offer. Each one costs more to win and tends to stay for a shorter time.
- Cash goes out before it comes back. If payback takes twelve months, every new customer is a twelve-month loan from the business. The faster you grow, the more cash is tied up, and a profitable model on paper can run out of money in practice.
- Cost to serve creeps up. New segments, new markets and new product variants bring complexity. Support, logistics and administration grow faster than expected, and they rarely shrink again.
- Discounts become the norm. Growth targets put pressure on sales, and the easiest lever is price. A campaign discount becomes the standard price, and the margin quietly erodes across the whole customer base.
- Attention leaves the existing customers. When the organisation is focused on winning new business, the customers it already has get less care. Retention falls, and the base that was supposed to fund growth starts to shrink.
None of this means growth is wrong. It means growth should be pushed when the unit is sound, and held back while it is not. Fix the unit first, then scale it.
Building a simple view from the accounts you already have
You do not need a new system or a data project to get started. A spreadsheet and a few weeks of work will usually give you a view that is good enough to act on. Here is a practical sequence:
- Pick the unit and a period. For example, customers over the last twelve months.
- Pull revenue per unit. Most accounting or invoicing systems can export revenue per customer or per order.
- Assign the direct variable costs. Cost of goods, freight, payment fees, returns and credits. Many of these can be linked to individual orders or customers.
- Estimate the indirect variable costs. Support time, account management, onboarding. Use simple drivers, such as number of orders or support cases, rather than precise time tracking.
- Calculate acquisition cost. Total sales and marketing spend divided by new customers won in the period.
- Group the customers. By start date for cohorts, and by segment, size, channel or product for comparison.
- Look for the pattern. Which groups have a strong contribution, a short payback and good retention? Which ones fail on one or more of those?
Expect the first version to be rough. A rough view that is honest is far more useful than a precise view that nobody has built. Agree on the assumptions, write them down, and improve the numbers that turn out to matter most.
A common result is that a fairly small share of customers produces most of the contribution, and a noticeable group contributes little or nothing once the cost to serve is counted. That is not a reason to drop those customers overnight. It is a reason to understand why, and to decide what to do about it.
The levers that move the numbers
Once the picture is clear, the useful question becomes which levers to pull. There are four main ones, and they tend to work best in combination.
Price. Price is usually the most powerful lever and the least used. A small increase goes straight to the contribution margin, and for many businesses it has more effect than a large cost programme. Look at where discounts are given, whether they are justified, and whether the price reflects the value you actually deliver. Also look at pricing structure: minimum order sizes, delivery charges and fees for custom work can recover the cost of expensive behaviour without raising the headline price.
Mix. Not all products, segments and channels are equally profitable. Shifting the mix towards the ones with a strong unit, by focusing sales effort, adjusting the offer or changing where marketing money goes, can improve the overall economics without winning a single extra customer.
Cost to serve. Simplifying how customers are served often brings large savings with little effect on what they value. Self-service ordering, standard documentation, fewer product variants, clearer service levels and better onboarding all reduce the effort per customer. Start with the groups that are most expensive to serve.
Retention. Keeping a customer longer spreads the acquisition cost over more months and gives more time for them to grow. Look at why customers leave, especially in the first year, and fix the most common causes. Often the problem lies in onboarding or in the first few deliveries, not in the product itself.
How leaders should use it in decisions
Unit economics is only useful if it changes decisions. A few habits make that happen.
Make it part of every growth decision. When someone proposes a new market, a new segment, a campaign or a new channel, ask for the expected unit economics alongside the expected revenue. What will it cost to win each customer? What will they contribute? How long until they pay back? The answers will be estimates, but asking the questions changes the conversation.
Set limits, not just targets. Instead of only setting a revenue target, agree on the acceptable range for acquisition cost and payback period. Sales and marketing then have room to grow within limits that protect the business.
Review cohorts regularly. Once a quarter, look at how the latest cohorts compare to earlier ones. This is the early warning that growth is starting to cost more than it brings.
Be careful with incentives. If the sales team is paid only for volume, it will deliver volume, including customers who will never be profitable. Adding a measure of margin or retention to the incentive makes the whole organisation pull in the same direction.
Where to start
You can make real progress in a few weeks. A practical first step:
- Choose your unit, most likely the customer, and pull twelve months of revenue per customer from your systems.
- Add the costs that clearly scale with volume, and estimate the cost to serve using simple drivers.
- Calculate an honest acquisition cost from your total sales and marketing spend.
- Group customers by start quarter and by segment, and look at contribution, payback and retention for each group.
- Pick the one lever, price, mix, cost to serve or retention, that the numbers suggest will make the biggest difference, and act on it.
Then hold a short meeting with the leadership team and ask one question: if we doubled the number of customers like the ones we have today, would we be better or worse off? If the answer is clearly better, push for growth with confidence. If the answer is unclear, you have found the work that should come first.