There's a moment about ninety days into a new market entry when the board asks how it's going. Someone opens a dashboard. Revenue is either ahead of plan or behind it. A mood settles over the room, and that mood starts making decisions - both moods are wrong.
Early revenue in a new market is the least reliable number a business produces. It is distorted by launch effort, founder attention, honeymoon pricing, and the small pool of buyers who will try anything new. None of those conditions persist, so a number built on them predicts nothing.
What early revenue actually measures
The first quarter of revenue measures your launch, not your market. It tells you whether your announcement landed, whether your first hire is energetic, and whether your existing network extended into the new geography. All useful things to know. None of them predict year two.
The failure pattern is consistent. A strong first quarter convinces the leadership team the model has transferred. Investment accelerates. Then the founder's attention returns home, the early adopters churn at rates nobody modelled, and the business discovers that acquiring a customer in the new market costs three times what it does at home. By then the cost base is built.
The reverse pattern is just as expensive. A quiet first quarter reads as failure, the entry is starved of support, and a market that needed eighteen months of patience is abandoned at month six. The post mortem blames the market. The market was fine.
The two signals that ARE worth reading
If early revenue is noise, what should the first ninety days be for? Two things.
First, channel economics. Not blended acquisition cost, which averages away the truth, but the cost and conversion of each channel measured on its own. In your home market you already know which channels carry the business. The first question abroad is whether those channels exist, what they cost, and how they convert. If your home market runs on referral and the new market has no referral network yet, your unit economics are not transferring, whatever the revenue line says.
Second, repeat behaviour. One purchase is curiosity. The second purchase is the market telling you something. Cohort your earliest customers and watch what they do next. Repeat rates, expansion within accounts, time to second order. These numbers stabilise long before revenue does, and they are much harder to flatter.
None of this will be statistically clean at ninety days. It does not need to be. You are looking for direction, and direction shows up in behaviour long before it shows up in revenue.
Write the decision down before you enter
The discipline that separates good entries from expensive ones is set before launch, and it fits on one page.
Before the first hire, write down what the month twelve review will look like. Put the review date in the calendar now, and put it in the board papers, so it cannot quietly move when the mood is good or the quarter is busy.
The page holds three things. First, the channel numbers you need to see: which channels you expect to carry the business, what acquisition should cost through each one, and what conversion looks like at home as the reference point. Second, the repeat behaviour that would confirm demand: the repeat rate, the time to second order, or the account expansion you would need to believe the market is real. Third, the kill condition, written as plainly as the go condition. If we cannot acquire below a set cost through at least one channel by month twelve, we exit. Name the number.
The targets you set now will be rough. That is fine. Rough targets set in advance beat precise judgments made in the moment, because the ones made in the moment are made by whoever is most confident in the room that day.
Then hold the review on the date, against the page. Not at month three, and not against how the launch felt. If the page says stay, stay, and give the market the time you said it needed. If the page says exit, exit, and do not let a good month talk you out of it.
The entries that go wrong are rarely killed by bad markets. They are killed by decisions made early, on the wrong data, with total confidence. The first ninety days tell you nothing about the market. They tell you whether you are disciplined enough to wait for the numbers that do.
