Translation vs Localisation: Why Crypto Press Releases Flop

Translation changes the language of a press release. Localisation changes the story itself: the headline angle, the proof point, the quote, the local data and the timing, so a reporter in Seoul, Tokyo or Mumbai has a reason to run it for their own readers. A translated release still answers an American editor's question, just in Korean or Japanese. That's why it flops.

I see the same pattern every quarter. A team raises, gets an English release out, and then "does APAC" by sending the same text through a translator and on to four language desks. The text is correct. Nothing is wrong with the grammar. It still sits unopened, because nothing in it was written for the person reading it.

What actually changes between a translated and a localised release?

Most founders picture localisation as better translation. It isn't. Translation works on sentences. Localisation works on the decisions that sit underneath the sentences.

Look at the last row. A reporter who wants to follow up and finds only an English inbox in another time zone will often move on to a story that's easier to finish.

Why do translated releases get ignored by APAC reporters?

Every reporter runs the same silent test on a release: does this matter to my readers today. A US-written release passes that test for a US reader. It talks about US investors, a US news hook and a product that may not even be available in the reader's country. Translation carries all of that across untouched.

There's a second signal too. A translated release tells the desk that the company hasn't committed to the market. No local partner, no local spokesperson, no local number. Reporters in markets that receive a lot of foreign crypto releases learn to spot this fast. I've covered the most common versions of this failure in five localisation mistakes that kill credibility in APAC, so I won't repeat them here.

I posted a short version of this last year and I'd write it the same way today:

web3 projects must focus on localization pro max! Shilika Jain on X, 15 Oct 2025

"Pro max" is the point. Localisation isn't a final step after the release is approved. It's a set of choices made before anyone writes a line.

The translate, adapt or rewrite ladder

Not every announcement deserves a full rewrite in every market. Rewriting everything is slow and expensive, and it can make a simple factual update look like a campaign. I use a three-rung ladder to decide, per market, per announcement.

Rung 1: Translate. Use it when the news is a hard fact that means the same thing everywhere: a mainnet date, a listing that's live, a security fix, a clean data point. The Tangem self-custody survey is a good example from my own work. Its headline stat, that two-thirds of crypto users support self-custody, was picked up across English, Korean, Japanese and Dutch editions, including Bloomingbit in all three of its languages, without a separate pitch per market. A strong number travels on its own.

Rung 2: Adapt. Use it for funding rounds, partnerships and product launches where the facts are global but the reason to care is local. Keep the facts. Rewrite the headline, the lede, the quote and the context paragraph.

Rung 3: Rewrite. Use it when the local reader is the point of the story: market entry, a local partnership, a local exchange relationship, a community event in that city. Start from a blank page in the local language, with the English version as a fact sheet, not a draft.

Three checks place an announcement on the ladder quickly:

1. A reader in this market would do something differently after reading it.
2. There's a local name, number or event in the story.
3. Someone from the company can talk to local press this week.

If none of the three is true, translate it, post it to your local channels and don't pitch it. One true usually means adapt. Two or three mean rewrite.

A worked example: one raise, two framings

The clearest case I can point to is the Bullieverse $4M raise in February 2022. It ran on two tracks with the same facts underneath.

For the Indian business and startup press, the story was an Indian-founded metaverse gaming team raising money. That's how it landed in Economic Times, Inc42, Entrackr, YourStory, VCCircle and Tech in Asia. Those outlets cover Indian founders and Indian capital, so the angle was the company, the founders and the round.

For crypto and tech media the framing changed. It went out as a press release on Cointelegraph, ran on NewsBTC and Investing.com (where the angle was the Chainlink VRF integration) and became a founder interview on Hackernoon. I'll be straight about one thing: the CoinDesk placement in that campaign was sponsored content, not earned. You can read the full breakdown in the Bullieverse case study.

Notice what didn't happen. Nobody took the crypto-native version and translated it for the Indian business press, or the other way round. Same round, same numbers, two stories, because two different audiences were reading.

What should go in a localisation brief?

When I hand an announcement to a native-language writer, I don't send the English release and ask for a translation. I send a brief. It fits on one page.

• The local angle in one sentence. If you can't write it, the announcement belongs on rung 1.
• The facts, numbered. Amounts, dates, partner names, product names. These must match the English exactly.
• Terms that stay in English. Token tickers, product names and protocol names, so the writer doesn't invent local versions.
• The local proof. Users, partners, events or numbers from this market, if there are any.
• Who the quote belongs to. The founder, a local lead or a partner, and whether the writer can draft it for approval.
• What can't be said. Price talk, return language and anything the legal team has flagged for that market.
• The local timing. Embargo time in local time, plus any local event or holiday that affects it.
• The follow-up owner. A name, a language and the hours they're reachable.

Then I ask the writer one question before anything goes out: would your editor run this? If the answer comes with a "but", the "but" is the most useful feedback you'll get.

I wrote this about marketing in general in 2023, and it applies to releases specifically:

what web3 needs currently is mass marketing with high level customisation by diving in deep into the one's product offerings! one strategy doesn't fit all. Shilika Jain on X, 8 Feb 2023

High-level customisation is what the brief buys you. The facts stay fixed. Everything around them is allowed to move.

When is translation good enough, and what does localisation cost?

Localisation isn't free, and pretending otherwise leads to half-done versions in five languages. Budget for these costs before you commit:

• Time. An adapted or rewritten release needs a native writer, a fact check against the English and a sign-off loop. Plan on a week of extra runway per market, more if partners need to approve quotes.
• People. Someone has to answer local follow-ups in the local working day. If nobody can, the localised release creates questions you can't answer.
• Consistency risk. Every extra version is another place a number can drift. One wrong figure in a Korean version is still a wrong figure.

Translation is often fine when the news is a hard fact, when you're only posting to your own local channels rather than pitching press, or when a market's business press works comfortably in English. The crypto press release distribution guide covers where plain distribution fits and where it doesn't.

The honest trade-off: two well-localised markets beat five translated ones. If you can only resource one, pick the market where you have a local name, a local number and a person available, and give it the rewrite. The others can wait for the next announcement.

If you want a second pair of eyes on which rung each market belongs on, that's a core part of my APAC PR work. And when the local reader needs a local face, not just a local quote, the next question is whether you need a local spokesperson.

FAQ

Can AI translation tools replace a native writer for crypto press releases?

They're useful for a first pass on rung 1 announcements, where the news is a hard fact. They don't make the decisions localisation depends on, like which fact leads or whose quote fits. Always have a native speaker who knows crypto check the result before it reaches a reporter.

Should the localised release go out at the same time as the English one?

Not necessarily. It should land on the local morning, which may be hours before or after the English version. What matters is that local reporters don't see the news first on a global outlet and then receive your release as a late echo.

Do I need a different press release for every APAC market?

Only for markets where the announcement sits on rung 2 or 3. A listing update can share one translated text. A market entry or local partnership needs its own version, written for that market.

Who should approve a localised release?

The person who owns the facts (usually the founder or comms lead) checks every number and name against the English. A native-language reviewer checks the angle and tone. Partners named in the release approve their own mentions and quotes.

Sending your next announcement into Korea, Japan or India? Book a 30-minute call and I'll tell you which markets need a rewrite and which can take a translation.