Somebody told you to work out what a customer is worth. So you did. You pulled a year of orders, found that the average customer spends about $240 with you over twelve months, and then you read the rule everyone repeats: spend up to a third of lifetime value to acquire one. Eighty dollars. You checked your ad account, saw you were buying customers for sixty, and concluded you had room to push.
So you pushed. The dashboard agreed with you the whole way. Return on ad spend held up. Revenue went up. Every number you had been taught to watch said this was working.
Then March arrived and you could not pay for stock.
Nothing in that story is a mistake in the arithmetic. The lifetime value was real. The customers did come back. The problem is that the rule you followed answers a question you were not actually asking. It tells you whether a customer is worth buying. It says nothing whatsoever about when you get the money, and when is the only thing your supplier, your landlord and your card issuer care about.
Your Lifetime Value Is Measured in Months. Your Bills Are Measured in Days.
Here is the shape of the trap, and it is worth slowing down for because almost every piece of advice on customer acquisition sits inside it.
Lifetime value is a forecast. It describes money that will arrive across the next six, twelve or twenty-four months, assuming the customer behaves roughly like the ones before them. It is a perfectly good number for deciding whether a customer is worth having.
Your obligations are not forecasts. The supplier invoice has a date on it. The rent has a date on it. The card that funded last month's ads has a date on it. None of those dates move because your cohort analysis looks promising.
So when the advice says "you can afford to spend $80 to acquire a customer," read it precisely. It means you can afford it eventually. Whether you can afford it this month is a completely different calculation, and nobody hands you that one.
Lifetime Value Is Revenue. Your Supplier Wants Gross Profit.
Before the timing problem, there is a smaller error sitting underneath it that makes everything look considerably better than it is.
Most owners compute lifetime value on revenue. Average order value times how many times someone buys. That $240 figure is almost always $240 of sales, not $240 of money you keep.
Run the two side by side on a store with a 30% gross margin, buying customers at $60:
On revenue, $240 of lifetime value against $60 of acquisition cost reads as a 4:1 ratio. Comfortable. Above the 3:1 rule of thumb everyone quotes.
On gross profit, that same customer produces $72. Against $60 of acquisition cost, you are keeping twelve dollars. Before you have paid for a single thing that is not the product itself.
Same business. Same customers. One number says you have room to double your spend, the other says you are nearly working for free. The gap between them is not a rounding difference, and it is the reason the ratio rule keeps failing people who applied it honestly.
The number that governs what you can pay for a customer is lifetime gross profit, not lifetime revenue. If you have never separated those two, that is the first afternoon's work, and it is the same exercise as working out which of your products actually makes you money rather than which one sells the most units.
The Payback Window Is the Only Number That Answers "When"
Once you are working in gross profit, there is exactly one question left, and it is the one the whole business turns on.
How many days pass between paying for the click and having that money back in your account?
That is the payback window. Not a ratio, not a multiple, not a score out of ten. A number of days.
It is also the number that return on ad spend structurally cannot give you. ROAS is a snapshot of an ad's performance. It tells you money came back at some point relative to money spent. It does not tell you when, it does not tell you how much of it you keep, and it does not tell you how much cash you need standing by in the meantime. You can run a 3x ROAS all quarter and still be genuinely short of money, and if that has happened to you, you are not confused. Those two facts are entirely compatible.
This is also the honest reason the usual advice feels like it works right up until it doesn't. A profitable customer acquired on a long payback window is still a profitable customer. You will get the money. You just will not get it in time to do the next thing you were planning to do with it.
Work Yours Out This Week. It Takes an Afternoon, Not Software.
You do not need a modeling tool for this, and you do not need a year of clean data. You need one export and a spreadsheet.
Step one. Take everyone who bought for the first time in a single month. Pick a month at least ninety days behind you so the story has had time to finish. That group is your cohort.
Step two. Work out what you paid for them. Total marketing spend for that month divided by the number of new customers it produced. Not customers, new customers. If a meaningful part of that budget was aimed at people who had already bought from you, that spend does not belong in this number.
Step three. Add up the gross profit that cohort has produced since, week by week, cumulatively. Every order those specific people have placed, revenue minus cost of goods, minus the variable costs that move with each order. Shipping you absorb. Transaction fees. Pick and pack. The return rate on that category, because a refunded order is not gross profit and pretending otherwise is how this number flatters you.
Step four. Find the week the running total crosses what you paid. That is your payback window. Write it down as a number of days.
Most owners are surprised by it, and the surprise usually runs one direction. The number is longer than they assumed, because they were unconsciously comparing acquisition cost against revenue from the first order rather than gross profit across several.
What the Ranges Mean, With the Caveat Nobody Attaches
There are working ranges for this, and they are useful as orientation. They are not a standard.
The clearest statement of them comes from Leighton at Leading Social, an agency running acquisition for ecommerce brands across Europe and the US. Under 30 days, in his framing, is a strong position. Thirty to sixty is healthy, and he puts most ecommerce brands there. Sixty to ninety is workable but wants real discipline about capital. Past 120 days you are not scaling so much as financing your own growth, which he is careful to say is a legitimate choice rather than a failure, but a choice you should be making deliberately.
Take those as one practitioner's working range across the brands he sees, because that is what they are. They are not a published benchmark and there is no governing body that blessed them. A range built from other people's businesses cannot know what your terms are, what you sell, or how much cash you have standing behind you.
Which brings us to the part nobody says.
A 75-Day Window Is Comfortable or Fatal Depending on Numbers That Have Nothing to Do With Marketing
Every source that teaches this teaches it against an abstract scale. Under sixty, good. Over a hundred, worrying. As if the number could be graded on its own.
It cannot, and this is the whole point.
Your payback window is not competing against a benchmark. It is competing against the other clocks already running in your business.
Take a store with a 75-day payback window. Put it in two different businesses.
Business A pays its suppliers on 60-day terms. The stock arrives, the invoice falls due at day 60, and the money from the customers that stock attracted does not fully land until day 75. Every month of growth opens that gap wider, because a bigger month means more stock bought on the same terms against money that arrives later. Growth is actively making the problem worse, which is the specific cruelty of this situation: the better the month, the tighter things get.
Business B pays on 90-day terms, or buys from a supplier who ships in two weeks instead of ten. Same 75-day window. Entirely comfortable. The money is home well before anything is due.
Same marketing performance. Same ads, same creative, same acquisition cost. Opposite verdicts, decided entirely by terms negotiated with a supplier who has never seen your ad account.
So the useful version of this exercise is not "is my payback window good." It is: write your payback window next to your supplier terms, your inventory lead time, and your rent, and see which one comes first. If the money gets home before the bills do, you have room. If it does not, you have a gap, and the size of that gap is how much cash you need sitting in the account before you are allowed to increase spend.
That gap is the same money we described as turning profit into inventory, seen from the other end. That piece is about where your cash physically goes. This one is about when it comes back.
Four Levers, in the Order a Small Store Can Actually Pull Them
If the window is longer than your obligations allow, there are four things that move it. They are not equally available to you, and the usual advice lists them in the wrong order for a business your size.
1. Sell more on the first order. The fastest lever, and the most reliably underused. Every additional dollar of gross profit collected on order one is a dollar you do not have to wait for. A genuine bundle, a larger default size, a considered add-on at the cart. Not a discount, which moves the number the wrong way. This is the lever that works the same day you pull it.
2. Fix the margin on what you already sell. Slower, and it means going back to cost of goods, freight, and what each order really costs you to fulfill. Unglamorous, permanent, and it improves the window on every customer you will ever buy.
3. Lower what you pay for a customer. The lever everyone reaches for first, which is why it is third here. Worth working on, genuinely. Also the one you have the least direct control over, because the auction has other people in it and their budgets are not your decision.
4. Get the second order to arrive sooner. Powerful when it is available. Here is the caveat almost nobody attaches: it is not available to everyone, and no amount of email marketing will make it available.
That warning comes from the retention side of the industry rather than the acquisition side, which is what makes it credible. If you sell mattresses or furniture or anything else a person buys once every several years, the repeat purchase is not slow because your flows are badly built. It is slow because of what you sell. Sending more email at a category problem produces more email, not more orders. The honest move there is either to widen what you sell into things that genuinely accompany the main product, or to accept that levers one through three are the ones you have and to run the business accordingly.
Knowing which of those two situations you are in is worth more than any tactic, and it is the same question underneath whether your second order is where the profit actually lives.
Before You Raise Spend, Break Your Own Numbers on Purpose
One last habit, borrowed from people who do this for much larger brands and scaled down to fit.
When you decide to increase spend, you are placing a bet on a projection. The sensible version of that bet is not to trust the projection. It is to stress it first, deliberately, and see whether you would still be all right if it came in worse than you hoped.
Take your payback window and rerun it with two pessimistic assumptions at once. Acquisition costs 20% more than it does today, because you are buying more volume and volume is usually more expensive. Repeat purchases come in 15% weaker, because the customers you buy at higher spend are typically a little less enthusiastic than the ones who found you first.
Now look at the window. If it still lands inside your supplier terms, raise the budget with some confidence. If it does not, you have found the ceiling, and the ceiling is real regardless of what the return on ad spend is telling you this week.
This is also the most common way small stores get hurt by a genuinely good month. The early cohort looks excellent, spend goes up to match it, and the number quietly degrades at the exact moment there is more money riding on it. Stressing the inputs beforehand costs nothing and is the difference between scaling and gambling.
Worth saying plainly: none of this means spend less. A short payback window is permission to spend aggressively, and plenty of stores are being far too cautious with budgets their cash position could easily support. The point is to size the spend to the window rather than to the dashboard. That is also the difference between the ads being profitable and your bank account knowing about it.
The Number to Have Before Your Next Conversation About Budget
If you take one thing from this, make it a single line written somewhere you will see it.
It takes us N days to get our advertising money back, and our suppliers want paying in M days.
Two numbers. Everything else follows from the relationship between them. If N is smaller than M, you have room and you should probably be using more of it. If N is larger than M, the difference is the cash you need in reserve before you are entitled to grow, and no ROAS figure will change that.
That one line will tell you more about what your store can safely do next quarter than any dashboard you own. It is also, not by coincidence, the first thing worth knowing before anyone starts spending money on your behalf.
