You cannot acquire your way out of a market of 1.3 million people.
Every founder here says some version of that sentence. Very few follow it to the conclusion, because the conclusion is uncomfortable and it isn’t “we should do retention better”.
The conclusion is that you should start turning customers away.
Three things follow from a finite list of possible buyers. Net revenue retention becomes your growth engine rather than your defence. The highest leverage retention decision you will make happens before anybody signs anything. And if you sell seats to Estonian companies, your expansion ceiling is set by the Estonian labour market rather than by your product.
That last one is the part nobody prices in, so let’s start there.
The number that should reorganise your pricing page
Aleph’s benchmarking work, drawing on full year 2025 data across several hundred B2B software companies, puts median net revenue retention at around 102%. Benchmarkit’s figure is 101%, down from roughly 105% in 2021.
Fine. Medians are nearly useless on their own.
The split that matters: usage-based companies came in at 108%, seat-based at 98%.
Sit with the second number. Below 100% means the existing customer base shrinks in revenue terms every year. A seat-based company with median retention is running down an escalator, and all the new logos are doing is keeping it level.
Now ask what seat expansion actually requires. It requires your customer to hire people.
In Estonia that is the hard part. I wrote recently about who to hire first when the talent pool is this small, and the short version is that the constraint on Estonian headcount growth is supply rather than ambition. So a seat-priced product sold into Estonian companies has its expansion rate governed by how fast those companies can recruit, in a country where recruiting is the bottleneck.
You have handed your growth curve to somebody else’s hiring problem.
Usage-based or hybrid pricing expands with what the customer does rather than how many people they employ. Activity can grow without a single new contract of employment. In a labour-constrained market that is not a pricing preference, it is close to a structural requirement.
Why a bad-fit customer costs more here than it would in Germany
Here’s the argument I’d actually make in a board meeting.
In a large market, a bad-fit customer who churns in month seven is a rounding error. You learned something, you move on, the pipeline refills.
In a market of 1.3 million, that same customer costs you three things at once. A logo permanently removed from a finite list, because they will not buy again and they now have an opinion. A reference who is in the room at the same four conferences you attend every year. And a quarter of your support capacity, spent on somebody who was never going to succeed with the product.
Churn in a small market is not a percentage. It’s a person with a name who is still in the room.
Which is why I’d argue that qualification is retention work, and that it belongs to whoever owns retention rather than to whoever owns the number for this quarter.
The practical version: write down the profile you refuse to sell to. Not the ideal customer profile, everybody has one of those and it changes nothing. The refusal list. Too small to get value, wrong workflow, no internal owner, buying because a board member suggested it. Then actually decline those deals, in a quarter when you need the revenue, which is the only time the policy gets tested.
I have never seen a company regret this and I have watched several wish they had done it two years earlier.
What marketing is for after the sale
The standard answer is lifecycle emails and an onboarding sequence. That’s customer success with a marketing budget attached and it’s not what I mean.
In a market this size the buyers know each other. They were at the same university, they sit on each other’s boards, they go to the same two events in Telliskivi. Your reference list is not social proof decorating a website. Your reference list is your demand generation channel.
So the job after the sale is making customer success legible to the rest of the market. A named case study with real numbers in it. Your customer on a panel rather than you. The integration one customer asked for, announced publicly, because three others wanted it and did not ask.
Eight visible, genuinely successful customers in Estonia is not a testimonials page. In some categories it is most of the market, and it does more than any campaign you could buy.
Where high retention and real danger look identical
Now the caveat, because the metric lies in small samples and this is where I’d push back on the founder presenting it.
With forty customers, a net revenue retention figure of 115% can be two accounts expanding hard while the rest sit flat or shrink. On the chart that looks like a healthy expansion motion. It is actually concentration risk wearing a retention metric.
The test takes five minutes. Recompute net revenue retention with your largest customer excluded. Then with your two largest excluded. If the number falls below 100%, you do not have an expansion motion, you have a dependency, and the strategic question is about that dependency rather than about retention.
The second trap is the benchmark itself. Those reported tier figures, with SMB nearer 97%, mid-market around 108% and enterprise higher again, say that your peer group is your contract size and not your country. An Estonian company selling to small businesses should not be comparing itself to an enterprise software median and concluding it has a retention problem.
The limit, stated plainly
Retention cannot rescue a product nobody needed.
If expansion from two accounts is carrying your net figure, look at gross revenue retention instead, which strips expansion out and tells you what share of revenue simply stays. That is the number that reveals whether customers want the product or whether a couple of champions are keeping the lights on.
Poor gross retention is a product and fit problem. No amount of expansion motion, usage-based pricing or customer success headcount fixes it, and in a small market you will run out of fresh names to try it on before you run out of ideas.
Where I come out
The small market is not the disadvantage founders treat it as. It just moves the leverage earlier, into pricing structure and into which deals you decline, and away from the acquisition machine that a bigger domestic market would let you build.
If you are seat-priced and selling domestically, that is the thing I would look at this month, before the next hire and before the next campaign.
If you’d like to think through your own retention arithmetic or whether your pricing model is capping you, drop me a line on email, WhatsApp, phone or LinkedIn and we can have a quick chat, and my notes on getting marketing and sales pointed the same way cover the handover this depends on. I’m taking on a small number of contracts at the moment, so I’ll tell you straight whether it’s something I could help with.