A translation app's 100k installs in six weeks: Arc Translate's trick isn't translation, it's a Turkish app factory

Arc Translate is an AI translation tool from Codeway, a Turkish app factory, released at the end of September. It hit 100k installs in six weeks and charted in the Top Grossing tool category in Mexico and the US. Google Translate has every feature it offers, and for free. It works because of a clever name, paid acquisition from day one, and a weekly subscription paywall that shows up right after install. The playbook is worth studying. The translation category itself isn't worth entering.

A store page written for Mexicans

Open Google Play in the Mexico store and it isn't called Arc Translate. It's "Traduce en Vivo: Arc Translate" (live translation), with the subtitle "Traductor de inglés a español" (English to Spanish translator) and the line "130 languages, voice, photo, text, AI translation". The description uses Mexican colloquial Spanish: "celular" instead of "móvil", "plática" (chat) instead of "conversación": "Translate your voice mid-conversation, and translate everything your camera sees." In-app prices run MX$39 to MX$1,349 (Google Play Mexico store page, 11-06).

In the US store the name flips to "Live Translator: Arc Translate", with "Live Translator" up front. In 2025 September, Apple's live translation on AirPods, plus Google and Samsung call translation, made "live translate" a trending search term. The iOS version launched 10-05 as "Arc Translate: Live Translator" in the US store and "Traduce en Vivo" in the Mexico store (App Store, 11-06).

One translation app, different names and accent-matched copy by country, targeting one audience across two markets: Spanish speakers going back and forth between Mexico and the US.

Who made it

The developer is listed as Codeway Technologies, an app factory in Istanbul. Its job postings claim 600M+ downloads, 60 apps, and 400 people. The iOS seller is DEEP FLOW SOFTWARE SERVICES - FZCO, another Codeway entity that holds Cleanup, Learna, and Chat AI. The 19 Codeway apps visible on Google Play have roughly 280M cumulative installs. Industry estimates put its monthly revenue at $25M+ (Sparrow Apps blog, 2026-04).

So this isn't one indie dev's hit. It's another product from an assembly-line factory. Around the same time it also launched Scanner (10-05) and Noticed (10-19), testing new products in batches.

How the volume came in: switch on, then steady

Data: Google Play install counts. The early-October figures are third-party estimates.

Two weeks after launch, nothing. Then the taps opened, and it added 2k to 6k a day steadily. Organic breakouts don't look like this. This is what paid campaigns switched on look like (inference).

Codeway's head of marketing has spelled out the company's rule: "The first thing we do when we launch an app is turn on paid acquisition, starting from day 0, hour one." Its channels are Meta, TikTok, Google, Apple Ads and creators. Each month it runs 200 to 500 creatives, spread across about 10 ad accounts (Sparrow Apps blog). In early November, the Facebook page for its cleaning app "Cleanup" had 3,839 ads running on Meta. Over the same period, the new app Scanner already had Meta ads sending users straight to Google Play, starting from 10-25. Its account in the Google Ads Transparency Center was promoting 9 apps at the same time.

We couldn't find Arc Translate's own ad creatives. Many of Codeway's ads land on an intermediate page, so you can't tell which app they point to. So "it was bought with ads" is a strong inference, supported by three things: the install curve opens like a tap, 100k installs produced only one or two hundred ratings (about one in a thousand, which organic word of mouth rarely does), and it charted only in Mexico and the US.

As for organic reach: TikTok searches for "arc translate" return no relevant videos, nobody has mentioned it or Codeway on X in the past two months, and there's no press coverage.

Where the money comes from: a paywall right after install

Codeway's paywall is one template: a short onboarding, then the paywall pops up right away. It can be dismissed, offers a 7-day trial, and sells only weekly subscriptions (Sparrow Apps blog). Reviews in the Mexico store spell out how the template feels:

• "Didn't like it. Said it was free, and then you have to pay weekly." (1 stars, 10-30)
• "The app is fine, but why does it ask for money the moment you install it?" (3 stars, 10-14)
• "plata y más plata" (money, and more money. 1 stars, 10-20)
• "I bought the membership, got charged, and right away translation said the app was busy, try again later. It stole my money and does nothing." (1 stars, 11-02)

US store: "How do I cancel my subscription?" (10-27), "My phone got stuck in another language, I just wanted to snap a photo and see what it said, and it wants a subscription! Something this simple and it wants a subscription!" (10-20), "Used the camera on a Japanese menu, and the Japanese text just didn't translate" (10-19). (Google Play reviews, Mexico and US, October to 11-05)

There are five-star reviews too: "a good way to study and practice English", "it saved me when I was in a pinch". Ratings vary across sources. The store page shows 4.2 stars from 155 ratings in 11-06. Another snapshot from 11-06 shows about 3.4 stars from 89 ratings, with big chunks at 1 and 5 stars and very little in between.

The money mostly comes from impulse downloads plus trial-to-weekly-subscription conversion, not from reputation or retention. It ranked #106 in Mexico's Top Grossing tools and #151 to #194 in the US, so it's making money, but not on a large scale (inference).

Who it's competing with

Translation is one of the most crowded tool categories on Google Play. Dozens of apps with translat in their name have been installed tens of millions of times.

Data: Google Play, App Store, 11-06.

Among this year's new translation apps, Arc sits in the lower middle. It charts because the bar for the tools Top Grossing list, past place 100, is low, and because it started taking money earlier and more aggressively than free-with-ads rivals.

What actually sets it apart

There's no product difference. Voice conversation, camera, text, 130 languages, "advanced AI models": Google Translate and dozens of similar apps have all of it. Users are still complaining that it can't translate the screen and handles Japanese menus poorly.

The difference is all in distribution: paid budget and creative output ready on launch day; a paywall template tuned across dozens of apps; Android and iOS listed under several company entities; language-app marketing experience from Learna (54.4M installs), a language-learning app; and store pages written in local accents. Whether Learna sent it traffic is unverified. These are things a copycat can't get at the start.

If it were you

In one line: don't chase this app, learn its playbook. Translation is fully covered by big players for free and crowded by tool shops. Arc squeezed out a little revenue because a factory resourced it, and the product itself has nothing to copy.

Who it suits: tool teams that already have paid-acquisition cash flow, can make creatives, and have paywall templates. For them, "proven need + new hot-term naming + country-specific copy + paid on launch + soft paywall with weekly subscriptions" is a formula that can validate revenue quickly. If it doesn't work, stop.

Who it doesn't suit: indie devs who have never run ads and are counting on organic store traffic. In this category your competitors are Google Translate, free and preinstalled, and Codeway, with hundreds of creatives a month.

How to do it: if you copy anything, copy the cheapest move, which works without paid acquisition: name the same app differently in each country using the words locals search for ("Traduce en Vivo" in Mexico, subtitle "English to Spanish"), and rewrite the description in local colloquial language, not machine translation. Pick a narrow scene where big players' features are still weak, not general translation, such as one particular pair of languages for a specific kind of file, menu, or medication leaflet.

How to get users: Codeway's method is paid campaigns in the first hour, with Meta, TikTok, Google, and Apple Ads all live together. A small team should first build organic traction through store search terms and localized names, measure the trial-to-paid conversion rate, and then test in one country with a small budget.

When to get in: no deadline. This is an always-on category, not a window.

What would show you misread it: (1) over the next month Arc's daily new installs hold or rise and it climbs into the top 50 tools charts, which means translation subscriptions still have money the big players haven't taken; (2) it starts charting in countries beyond Mexico and the US, which means Codeway thinks the app is worth pushing harder; (3) the reverse: if by December it has dropped off both charts and updates have stopped, that means even Codeway didn't make it work, and latecomers should definitely not bother.

Data and sources

• Installs, charts, ratings: Google Play, 2026-10-29 to 11-06. The early-October installs are third-party estimates.
• Store page copy and pricing: Google Play Mexico and US stores, App Store US and Mexico stores, 11-06.
• Reviews: public Google Play reviews from the Mexico and US stores, October to 11-05. Originals are in Spanish and English; these are translations.
• Ads: Meta Ad Library and Google Ads Transparency Center, early November.
• Codeway's playbook and scale: Sparrow Apps blog "How Codeway built a $25M/month app empire" (2026-04-08), Codeway's job postings, and a discussion on Reddit r/iOSAppsMarketing.
• Where it says "estimated" or "inference", those are estimates and inferences.