पाठशाला Pathshala · हिसाब Hisāb, Unit economics · Lesson 12 · Build
Payback period and the cash trap of fast growth
An eighteen-month payback is a good customer and an expensive one. Grow fast at that payback and the cash hole deepens every month. How to see the trap and set the payback ceiling your runway can carry.
Pathshala, The Founder Library · 11 October 2026 · 6 min read
A company can have customers who are worth three times what they cost, a product they love and revenue doubling every year, and still run out of money. The arithmetic is simple and unforgiving. Each new customer costs cash today and repays it over months. A growing company is always paying for its newest customers before its older ones have finished repaying, and the longer the payback and the faster the growth, the bigger the hole between the two.
This lesson defines payback precisely, shows why growth deepens the cash trough rather than filling it, works the case of an 18-month payback at 100 per cent growth, and gives a method for setting the payback ceiling that the cash in your bank can actually carry.
Payback, defined
CAC payback is the number of months of gross profit from a new customer needed to recover what it cost to acquire them: CAC divided by monthly revenue per customer times gross margin. It is computed on gross profit, not revenue, because revenue that goes straight back out as cost of delivery repays nothing; the [CAC, LTV and payback](/library/cac-ltv-and-payback-the-three-numbers) lesson works through why. Use paid CAC, the spend on a channel divided by the customers it won, so that organic customers do not flatter the number.
The benchmarks are well established. David Skok’s Startup Killer put the rule plainly in 2009: aim to recover CAC in under twelve months, otherwise the business will require too much capital to grow. Bessemer’s Scaling to $100 Million refines it by customer: under twelve months for companies selling to small businesses, under eighteen for mid-market and under twenty-four for enterprise, with the average company between $1 million and $10 million of ARR at fifteen months. Those targets are for companies with access to venture capital. A company funding growth from its own cash needs shorter.
Why growth deepens the trough
Picture the company as a sequence of monthly cohorts. Each is bought for cash on day one and repays a slice of its cost every month after. With no growth, each month’s spend on a new cohort is eventually matched by the repayments of earlier cohorts, and the cumulative cash line falls, bottoms and climbs. With growth, each new cohort is larger than the last, so this month’s outflow is bigger than the repayments from cohorts bought when the company was smaller. The faster the growth, the longer that remains true.
Skok’s SaaS Metrics 2.0 shows the same effect in a full model: subscription businesses face significant losses in their early years, the trough gets deeper as bookings growth increases, and profitability is anaemic once the time to recover CAC extends beyond twelve months. The losses are not a sign that the customers are bad. They are the cost of buying good customers faster than they repay, and somebody has to fund them.
Eighteen months and 100 per cent growth, worked
A Bengaluru company sells software to mid-sized firms. Each customer returns ₹10,000 a month in gross profit, monthly churn is 2 per cent and the company wins twenty customers in its first month of serious selling. At an 18-month payback CAC is ₹1.8 lakh. It plans to double the number of new customers it wins every year.
Run the cohorts forward. At 6-month payback, with the same growth, the acquisition engine consumes ₹39.7 lakh at its deepest, in month seven, and has repaid all of it by month fourteen. At 12-month payback the trough reaches ₹3.5 crore and does not bottom until month twenty-nine. At 18-month payback the line is still falling at the end of month thirty-six, by which point acquisition alone has consumed ₹17.4 crore, before a rupee of salary or rent. Nothing about the customers changed between the three cases except how long each takes to repay; the cash required rose more than forty-fold.
Churn makes it worse than the headline payback suggests. At 2 per cent a month a cohort that would repay its CAC in eighteen months if everyone stayed takes about twenty-two months as a whole, because some customers leave before they finish repaying. At 5 per cent many never do.
The figure starts at the Bengaluru company. Bring payback down to twelve and then six and watch the green line climb out of the hole. Then set growth to zero and see that an 18-month payback is survivable for a company that is not growing; it is the combination that is dangerous. Finally move the cash slider to what you actually have and read the payback ceiling.
Setting the payback ceiling your runway allows
Start with the cash you can commit to acquisition. That is cash in the bank less the fixed cost of running the company for the period you are planning, less a reserve of at least three months of total spend, because the [runway lesson](/library/runway-how-many-months-you-really-have) is right that nobody raises money in the last month. What is left is the floor the trough must not cross.
Then compute the trough for your growth plan at different paybacks and take the longest payback whose trough stays above the floor. For the Bengaluru company with ₹5 crore available for acquisition, the ceiling is twenty-one months with no growth in new customers, fifteen months at 50 per cent growth, twelve at 100 per cent and nine at 200 per cent. With ₹2 crore the ceilings fall to thirteen, eleven, ten and eight. The ceiling is not a benchmark from someone else’s portfolio. It is a property of your own balance sheet and your own growth plan, and it changes every time either does.
Paul Graham’s Default Alive or Default Dead? asks whether a company reaches profitability on the money it already has. Payback above the ceiling is the most common way a company with good unit economics answers no: every customer is profitable in the end, and the company dies before the end arrives.
Payback tells you whether a customer is worth buying. Payback times growth tells you whether you can afford to buy them at that speed.
Four ways out of the trap
Bill annually in advance. A business customer who pays twelve months up front instead of monthly turns a cash payback of many months into one of weeks whenever a year of gross profit exceeds CAC, without changing the accounting payback at all. Offer a discount for annual prepayment smaller than what the cash is worth to you, and make annual the default on the order form.
Raise price or margin. Payback falls in direct proportion to gross profit per customer. A 20 per cent price increase with delivery costs unchanged cuts an 18-month payback to fifteen months or less. Read the [gross margin lesson](/library/gross-margin-why-investors-fixate) for the cost side of the same lever.
Cut the expensive channels. Blended payback hides a range. Compute payback by channel; the field-sales channel at thirty months and the partner channel at eight are not the same business, and the first should be cut or repriced before the second is starved.
Grow more slowly, on purpose. If payback cannot come down before the trough reaches the floor, the honest move is to slow the growth of new customers until it can. Andreessen Horowitz’s 16 Startup Metrics notes that the contribution-margin LTV to CAC ratio also helps a company determine payback and manage its advertising and marketing spend. Slower growth on a shorter payback compounds; faster growth on a long one runs out.
The monthly payback check
On the fifth working day of each month compute payback for the cohort acquired three months ago, by channel, using its actual gross profit to date and its actual CAC. Plot it beside the payback of the cohorts before it; the [cohort lesson](/library/cohort-analysis-for-founders-not-analysts) shows how. Recompute the trough for the next eighteen months of the growth plan, compare it with the cash you can commit to acquisition, and write the payback ceiling at the top of the growth plan.
Then apply one rule. If actual payback is above the ceiling, no channel’s budget rises next month until payback falls or the ceiling rises through new capital. If it is below, the company may grow faster, and the model says by how much. The decision takes ten minutes once the numbers are in place. Skipping it is how good companies become stranded ones.
Nothing here is legal, tax or investment advice. The Bengaluru company is illustrative and the trough counts acquisition cash only; your full cash forecast must add fixed cost, collections and tax.
Sources
- David Skok, Startup Killer: the Cost of Customer Acquisition, forEntrepreneurs, December 2009 — Recover CAC in under twelve months or the business will require too much capital to grow.
- David Skok, SaaS Metrics 2.0: A Guide to Measuring and Improving What Matters, forEntrepreneurs, January 2013 — Early losses deepen as bookings growth rises; profitability anaemic beyond twelve months to recover CAC.
- Mary D’Onofrio, Ethan Ding and Atlas editors, Scaling to $100 Million, Bessemer Venture Partners, September 2021 — CAC payback targets of under 12, 18 and 24 months for SMB, mid-market and enterprise; average of 15 months at $1–10M ARR.
- Paul Graham, Default Alive or Default Dead?, October 2015
- Jeff Jordan, Anu Hariharan, Frank Chen and Preethi Kasireddy, 16 Startup Metrics, Andreessen Horowitz, August 2015 — Contribution-margin LTV to CAC as a measure of CAC payback and a guide to marketing spend.