A native speaker costs 120% more in Finnish and 31% less in Ukrainian

Across eighteen languages on Preply’s own pages, read on 9 October 2026, the premium a native speaker commands runs from +120% in Finnish to −31% in Ukrainian. The order tracks how concentrated each language’s native speakers are in expensive countries.

And it moves. We first read the same pages on 24 August 2026. Six weeks later Korean’s premium had gone from +40% to +9%, and Italian and Chinese had changed sign: native teachers went from dearer to cheaper than non-native ones. Any single price snapshot, including our first one, is a reading, not a fact about a language.

Every figure below was read from the platform’s own public page on the date shown. None of them are ours. We are not in these tables. Where we could not verify something first-hand it is named as unverified rather than filled in from hearsay.

Take the numbers and use them. No permission needed, no credit required, no link required. They came from public pages and you can re-read those pages yourself, which is the only reason a price table from a company that sells lessons is worth anything at all.

Download as CSV: 9 October 2026 read · 24 August 2026 read. Every row with its source and the date it was read, released CC0.

Preply publishes a native-versus-non-native comparison on every language’s tutor page. Read across all eighteen languages in our catalogue on one day, it does not point the same way twice, and the order it falls into is not random. The last column is the same comparison six weeks earlier.

The ends of the table held. Norwegian stayed near +38% and Ukrainian near −31% on both reads. The middle did not.

Korean fell from +40% to +9%. It was our first read’s headline, and it is now in the middle of the table. English went the other way, from +14% to +33%, and German and Dutch nearly doubled. Italian and Chinese changed sign. Spanish went from non-native teachers costing more to exact parity.

We cannot tell you why, and Preply does not say. These are averages over whoever is listed that day. A move can be teachers changing their prices, or teachers joining and leaving the pool, or a change in how the average is computed. The page does not publish how many tutors stand behind each figure, so a thin cell can swing on a handful of people.

What it does tell you: if you read a claim that native speakers cost X% more in a language, ask when it was read. Ours was wrong for Korean within six weeks.

If the number really tracks geography rather than nativeness, then it should still move inside a single language, between countries where that language is natively spoken. It does, and Preply publishes the breakdown itself. It shows four countries per language and the four change between reads, so a dash means that country was not shown that day.

Portuguese breaks the pattern. Mozambique topped that table at $32 in August and tied Portugal at $28 in October, above Brazil both times. Mozambique is one of the lowest-income countries in the world. On a cost-of-living story it should be at the bottom.

We do not know why, and guessing would be worse than saying so. The most likely explanation is thin data: a country cell backed by a handful of tutors is an average over almost nothing. That is a caveat for every table on this page, not just this one.

A teacher sees a different axis. Juliana Freire, a Brazilian Portuguese teacher (learnwithju.com), told us: “Students tend to choose between Brazilian and European Portuguese based on their personal goals, such as travelling, living in a particular country, or communicating with family and friends, rather than price alone.” If the choice is the variety first and the price second, a country column mixes two markets, which is one more reason not to read it as a price for nativeness.

These pages are personalised by location. Read from a European address, Preply’s price filter came back in Danish kroner while its comparison table stayed in dollars. Someone reading from another country may be shown different numbers on the same URL, so date and locate any figure you quote.

Advertised is not paid. On Preply, lessons are bought as a subscription: its help centre states learners are charged every 28 days, that plan lessons do not carry over and expire at renewal, and that separately bought extra lessons survive two full cycles. A 365-day year holds just over thirteen 28-day cycles. That arithmetic is ours, not a claim they make.

Two reads are two reads. This is a two-point series, not a trend. Re-read the pages before citing any number as current.

KoloLingo is a marketplace for independent language teachers, so we sell the thing this page prices. Read it with that in mind.

The defence is that we are not in any table here, no figure is ours, every one of them is checkable on someone else’s site, and the numbers do not flatter us. The cheapest tutors in this dataset are cheaper than anything a new platform can conjure, and the incumbents’ selection is far larger than ours. If you want the argument that does cost us something, our fee comparison includes the row where a competitor beats us outright.

Every language in the first table links to its own write-up, with the reasoning rather than just the numbers and the free routes to check before you pay anyone.

Corrections are welcome. If a figure here is wrong, or a platform publishes something we missed, write to hello@kololingo.com and we will change it and say who corrected us.