Who Is Learning Morse Code in 2026? Notes From 176 Countries

When we started MorseKit, the question we heard most was some version of who is this for? Not unkindly. Morse has been off the shipping lanes since 1999 and off the amateur radio licence exam in most countries since the mid-2000s. A reasonable person would expect the audience to be a few thousand radio operators and a few hundred nostalgic engineers.

We now have five months of data, and that reasonable person would be wrong. This is what it shows, minus the parts a competitor would find more useful than you would. That omission is explained at the end, because it is deliberate and you deserve to know what is being left out.

Where these numbers come from

Everything below is read from the app's aggregate analytics on 6 September 2026, covering the period from launch in April 2026 to that day. Aggregate means exactly that: counts by country, by device language, by hour of the day. There is no per-person data in any of it, and there is nothing in it we could use to identify anyone. Country comes from the device's region, language from its system setting. Both are rough, and both are honest enough for what follows.

One caution. A country with a single session is a country where one person opened the app once. We are going to tell you about those places because they are genuinely delightful, not because they represent a user base. The countries that matter statistically are the ones in the ranked section.

Every continent, Antarctica included

People in 176 countries and territories have opened MorseKit. That covers every inhabited continent, and one session from the uninhabited one. There are only around 195 countries in the world, depending on who is counting, so the more interesting question is which ones are missing, and the honest answer is: mostly places with very few phones.

Some of the far-flung places are worth a moment. Most of these are a single session, and each one is a person who, in that place, decided to learn Morse:

The browser tools on this site, which need no install, have been used from more than eighty countries in the same window. What people reach for most is not the translator but the decoders: a video of a flashing light or an audio recording that someone needs read. People arrive with a message already in hand.

The countries that surprised us

The top three are the ones you would guess: the United States, Canada and the United Kingdom, in that order. English-speaking, large, with long amateur radio traditions. After that the list stops behaving.

There is a pattern in there that we did not expect, and we offer it as a guess rather than a finding. Countries with a recent military service tradition, an active scouting movement or a strong radio club culture show up far above their population. Countries where Morse was never part of civic life sit low even when they are large. Morse does not seem to spread at random. It spreads along old lines.

Forty-nine languages on people's phones

We built the app in 27 languages, which felt ambitious at the time. The phones it is installed on are set to 49. Roughly a third of learners use a device whose language is not English, and the largest non-English groups are Russian, French, Italian, Chinese, Spanish and Japanese, all close together.

Then there is the long tail, which is the part we keep coming back to. Someone learns Morse on a phone set to Māori. Someone on a phone set to Luxembourgish. Icelandic, Uzbek, Georgian, Mongolian and Tamil are all in the list. One person has their phone set to Latin, which we choose to believe is a classics teacher with a sense of humour.

The app speaks 27 languages, and only about half of the 49 on people's phones are among them. The rest see English. That gap is the most concrete thing this data has told us to fix, and there is more on it at the end.

Why people say they came

The first thing the app does is ask why you are here. Not everyone answers, but most do, and the shape of the answers is the single most useful thing we have learned about this audience:

Before launch we assumed amateur radio would be the majority. It is a strong minority, but the majority of people learning Morse in 2026 have no practical reason to. They are doing it because it is interesting, and because it turns out to be one of the few skills you can learn in five minutes a day and feel getting better every week. We wrote about that feeling, and the argument against it, separately.

What practice actually looks like

A few things from the aggregate that describe the habit rather than the people:

• Roughly two in three lessons that get started get finished. For a five-minute lesson that is a good ratio, and the missing third is, in good part, people who ran out of hearts rather than patience.
• Sunday is the busiest day, by a small but consistent margin. Every other day is close to flat. Morse is a weekend thing more than a commute thing.
• Evenings win. Practice peaks between seven and nine at night, local time, and bottoms out around seven in the morning. More people practise at midnight than at ten in the morning, which says something about who these people are.
• Added together, the time people have spent practising comes to the better part of a year. Not a year of calendar time: a year of continuous, minute-by-minute Morse, in five months. That is the number that made us sit down.

The mistakes that stop people are the same everywhere, incidentally. We listed the six that come up most, and nothing in the country data changes that list.

What we are not telling you, and why

You will have noticed that this article has no headline number. No total learners, no downloads, no lessons completed, no revenue, no split between platforms. That is not modesty and it is not because the numbers are small.

It is because those figures are worth more to the three or four other teams building Morse apps than they are to you. A reader wants to know whether an app is alive, whether it is used by people like them, and whether the makers are paying attention. Countries, languages, reasons and habits answer all of that. A download count answers a different question, and it is not one a learner is asking.

So the rule for this post was simple: ratios, ranks and counts of countries and languages are in. Anything a competitor could put on a slide is out. If you are one of those teams and you are reading this anyway: hello, the Romanians are yours to figure out too.

What this changes for us

Three things, concretely.

Languages before features. More than twenty languages appear on people's phones that the app does not speak. Some are small. Romanian is not, and Romania is a top-five country. That is the obvious first candidate.

The curious majority is the audience. We had been designing for the radio operator in our heads. Seven in ten users are not that person. They want the daily lesson to be short, the feedback to be kind, and the app to be pleasant at midnight on a Sunday. That is a different product brief from the one we started with, and a better one.

Radio stays. The one in five who are here for the air are not the majority, but they are the reason the code still exists to be learned, and they are the ones who tell us when something is wrong. The prosigns, the Q codes and the callsign practice are staying exactly where they are.

If you are one of the 176 countries, thank you. If you are Antarctica, please get in touch. We have questions.