In June, Estonia exported EUR 188 million of goods to Latvia and EUR 122 million to Germany.
Latvia has fewer than two million people. Germany has more than eighty million.
Statistics Estonia publishes the breakdown with every monthly trade release, and the June 2026 ranking goes Finland at 16% of total exports, Latvia 11%, Lithuania 9%, Sweden 8%, Germany 7%. The four largest destinations for Estonian goods are all countries with fewer than six million people. The biggest economy in Europe is fifth.
So if the shortlist on your wall was ordered by population or GDP, it was ordered by a variable the trade data doesn’t support.
Here’s what I’d use instead. Pick the market where your existing references still carry weight and where one person can cover the buying process, and start from who has already bought from you without being sold to. Size is a tiebreaker. It isn’t a criterion.
One caveat before the method. Those are goods, because goods are what gets published monthly. If you sell software that exact table doesn’t exist for you, and I’m not going to pretend it does. But the reason the pattern shows up isn’t freight, and the part that isn’t freight transfers.
Start with the orders you did not chase
The strongest signal available to you is demand that arrived without being paid for.
Open your invoice list and pull every customer outside Estonia from the last twenty four months. Then do the same with your enquiry inbox. Put each one in one of two columns: we went looking for this, or this turned up [the second column is the one you have never audited].
It’s market research you already paid for and never read.
Count the lost ones and the ones you turned down, not only the ones you won. A country where three deals arrived and you lost all three has demand plus a problem, and problems are cheaper to fix than demand is to create. A country where nothing ever arrived has neither.
What counts as a signal?
Three unprompted enquiries in two years, from unrelated buyers. One large customer isn’t a market. It’s one customer, and building a plan on it is how companies end up with a Dutch subsidiary and a single logo.
The five things that predict what entry actually costs
Everything above tells you where demand might be. These tell you what it’ll cost to go and get it, which is the half that gets skipped.
The language of the buying process. Not whether people there speak English. Whether the contract, the procurement form and the internal business case justifying your invoice get written in English. Different questions, and the second one sets your translation and hiring bill. The demand side is checkable in about an hour per market, and I went through how in my notes on international SEO.
Tender or relationship. If municipalities, hospitals and state agencies buy your category, your cost to enter is paperwork and framework agreements rather than relationships, and the published thresholds decide whether the notice is visible from Tallinn at all. That has its own post, on winning EU public sector tenders. The selection consequence is the part worth saying here: a tender market is slow to enter and hard to be pushed out of, and a relationship market is the reverse. Pick the shape you can fund for two years.
Whether your references transfer. The blunt one. An Estonian logo list means something real in Riga and Vilnius, something smaller in Helsinki, and close to nothing in Munich. I wrote about the reference problem for Finland and Sweden because it is what actually blocks those deals.
How you get paid, and what the admin costs. Estonian buyers expect a bank link rather than a card field, and a shop that only takes Stripe quietly loses conversions here. Every market has one of those expectations and yours is not it. On invoicing, several member states now run mandatory domestic e-invoicing on different timetables, which is not your obligation as a foreign seller but is very much your buyer’s accounts payable process. The German version is in my notes on selling into Germany. Cheapest criterion here to check, and the one most often discovered in month four.
Whether one person can cover it in a day. More on this below, because it’s the criterion people laugh at and then lose a year to.
Score the three you are actually considering
Two points each, ten available.
| Criterion | 2 points | 0 points |
|---|---|---|
| Buying language | Contracts and business cases get written in English | Everything commercial happens in the local language |
| Buying mode matches you | Bought the way you already sell, tender or relationship | The opposite mode to the one you are built for |
| References transfer | Your current customers are recognisable names there | Nobody has heard of any of them |
| Payment and invoicing | Your checkout and invoices already fit local habits | Both need rebuilding before the first sale |
| Day-trip reachable | Out and back in a working day, no hotel | Overnight minimum, each visit costs two days |
One rule about reading the total. The references row isn’t substitutable. Every other cost on that list is paid once. That one gets paid again on every deal for the first two years, so a market scoring zero there isn’t rescued by scoring well everywhere else, and a market scoring two there is worth entering even if it looks small.
The day-trip test, which sounds trivial and is not
Open the current schedule out of Tallinn and ask a narrow question. Not whether a flight exists. Whether there’s a flight out before nine and back after six, on the same weekday, reliably.
If the answer is yes, that market gets eight or ten visits a year from one person and has no hotel line in its budget. If the answer is no, it gets three, and each of them costs two working days plus a night.
Here’s why that decides deals and not just expenses. A B2B sale in a new market closes on the second or third meeting, and the number of meetings you can afford is set by the timetable rather than by the plan. In a market you can day-trip, you can lose a meeting and go back next month. In one you can’t, every trip has to justify its own cost, so you push for a decision before the buyer is ready. That isn’t a travel budget problem. It’s why first-expansion deals die.
Why the smaller market keeps winning
Go back to that June table. Latvia outsells Germany by half again, and it isn’t really about distance in kilometres.
It’s that Latvia scores on several rows at once. Your references are recognisable, the buyers bank and invoice the way you already handle, and a salesperson can be there and back before dinner. Germany, for an Estonian company on its first expansion, scores low on three specific rows: the buying language, the references, and the accounts payable process your invoice has to survive. Being the largest economy in Europe reduces none of those. It raises the hardest one, because competition for a buyer’s attention scales with the market.
None of which means never go to Germany. It means Germany is a market you enter from a position you don’t have yet, and the job of the next market is to build that position. Choose the second market that makes the third one cheap.
Before lunch
Pull every non-Estonian invoice and enquiry from the last two years, sort by country, and mark which ones you went looking for. Score your top three countries on the five rows above. An hour of work, and in my experience it reorders the list more often than it confirms it.
If you want to think through the shortlist before anybody commits to a translation quote, drop me a line on email, WhatsApp, phone or LinkedIn and we can have a quick chat, and my notes on how I work cover the rest. 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.