From 13k installs a day to 76 a day a week later: a Vietnamese studio's emoji keyboard is paying for its tests

In October 22, a keyboard called "Emoji Keyboard: Themes & Fonts" on Google Play gained about 13k installs in a day. A week later, on October 28 and 29, it was down to 76 a day. On October 30 it was back to 7k a day. By November 6 it had 86k cumulative installs. A curve that switches on and off like this doesn't come from users spreading the word. Someone is turning budget on and off. There's nothing in the product worth copying. What's worth studying is the small-budget testing process, and why it still hasn't broken through.

Who's behind it

The store page developer is "Shark Studio - Awesome Apps". The "About the developer" section at the bottom lists a Vietnamese name, NGUYEN VAN MANH, with an address in Hung Nguyen district, Nghe An province, Vietnam (Google Play store page, 11-06). The official site is an unchanged Firebase default page, the support contact is a Gmail address, and the homepage tagline reads "Awesome music effects for you". The studio started out making music editing apps. Its flagship, "Music Editor: Mp3 Cutter", has over 1M installs.

The same account also publishes a noise reducer, a reverse-singing challenge app, a clap-to-find-your-phone app, an IPTV player, a paper-folding tutorial, and an earlier keyboard called "Neon Love Theme: Font Keyboard". The third-party sites androidrank and AppGazers count 16 and 22 apps respectively. This is a typical Vietnamese utility-app studio: it makes whatever category is evergreen and lines up a batch of apps under one account.

What it is

A third-party Android keyboard. The store listing highlights four things: fancy fonts and emoji, cute or flashy keyboard themes, DIY customization (your own photo as the background), and key sound effects. The description states outright that the designs are "facemoji-inspired". It copies Facemoji, the category leader, and doesn't try to hide it.

It has ads and in-app purchases, priced at $4.99–$29.99 per item. It launched in October 8, and by November 2 had reached version 1.1.2. APK mirror sites show versions 1.0.2, 1.0.6 and 1.1.0, roughly one release every two or three weeks.

Of the top three reviews on the store page, two were written before the traffic burst (both in October 16):

"Too many ads!!!!!!!" (Alia Zahira, 11 people found this helpful)

"Why can't I have a decent keyboard, it's so boring, change the theme, I hate this" (Varsha Teekaram, 6 people found this helpful)

The other one is from October 28: "I love my keyboard!! So cute! Arigatou Gozaimasu!" Judging by the tone, many users seem to be kids and young fans (inferred).

Where the volume comes from

Start with the install curve. October figures are estimates from the third-party AppstoreSpy; figures after October 28 are public Google Play numbers. Google Play refreshes only every few days, so the table below uses period averages:

Now look at where it charted. From Oct 29 to Nov 4, it appeared on the Google Play personalized Top Free chart in 18 to 24 countries at once, including the US, UK, Germany, France, Brazil, Saudi Arabia, Thailand, Taiwan and South Korea. On Nov 3 its best rank was #16, with a median rank of #110. By Nov 6 it was down to 8 countries, with Thailand its best at #77. It never charted on any overall chart, and never on Top Grossing.

Taken together, the picture is clear:

• On and off. Organic growth doesn't go from 13k a day to 76 a day a week later, and then to 7 a day the next. That's budget being switched on, off, and on again (inferred, high confidence).
• Worldwide up, worldwide down, together. Developed markets and emerging markets charted and dropped on the same days. That looks like a worldwide Google app campaign, not a boost from content in any one country (inferred).
• Ratings follow the rhythm. On burst days it gained 15–24 ratings a day. When paused, it gained 2–5 a day. Nearly all the new ratings were 5 stars, and the average moved from 10-29 with 3.8 ratings to 11-06 with 4.5. There may have been an in-app rating prompt (unverified).
• No ad creatives found. The Meta Ad Library shows no record of this package name or developer. That doesn't mean it wasn't advertised, since Google app ads are hard to see from outside.

Other channels were weak. On TikTok there's an account, @emojikeyboard_themes, with 27 followers and 58 videos. The top video got 7,820 views (October 5, "Which keyboard sound do you pick when you type every day?"). Its main angle is "put your idol on your keyboard": "Your bias should be on your keyboard too", "Which CORTIS member do you pick?", riding the K-pop rookie group CORTIS's buzz, with a few hundred views per video. The earliest video is from September 19, before the app launched. That suggests this is the account for the keyboard product line, not a dedicated promo account (inferred). I found no discussion on X, in media, or on Reddit.

After paid traffic stopped, it still got about 3,500 installs a day. Whether that's organic store search or ads that weren't fully paused, I can't tell from outside. The title and package name pack in emoji, keyboard, themes, fonts, DIY and stylish, so store search does send it some traffic.

Who it's competing with

"Theme / font / emoji keyboard" is one of the most crowded categories on Google Play. By total Google Play installs (11-06):

Above it sit Gboard (about 14B) and SwiftKey (about 2B). The table shows two things. First, demand in this category is steady: My Keyboard, launched in August 2026, reached 5.63M in three months, so newcomers can still take off. Second, this one is small even among this year's new releases, and there are several near-identical same-category apps.

Where the money comes from

It relies mainly on ads. People open a keyboard dozens of times a day, and there are slots for interstitial and app-open ads. The "too many ads" review is the price of that. The $4.99–$29.99 in-app purchases are there for themes and ad removal, but it has never charted on any country's Top Grossing chart, so IAP revenue is probably small (inferred). Whether ad revenue covers the cost of paid traffic can't be seen from outside. Since the second burst in November 5, it has been spending again. Whether that's a deliberate budget choice or a sign that payback is failing, I can't tell.

If it were you

The verdict: Don't copy this reskinned keyboard. It hasn't broken out on its own. What's worth learning is its testing process: an evergreen big category, a title packed with keywords, and a small budget run worldwide for a few days. Watch the personalized charts and the organic installs left after pausing, then decide whether to spend more.

Who it's for: small teams already building ad-monetized utility apps, with paid acquisition experience and code they can reskin again and again. For them, it's a cheap way to validate: the "Personalization" category chart has a low bar, and a few thousand installs a day can get you into the top 200 across 20 countries.

Who it doesn't suit: Indie developers with no paid-acquisition experience who want to build a new keyboard from scratch. New entrants in this category almost all grow through paid traffic, and the leaders have tens of millions to hundreds of millions of installs. If you can't work out payback or make creatives, you're just paying tuition to the ad platforms.

How to build it: If you must go after keyboards, don't compete on skin count. Its reviews show two openings: ads that are too heavy, and "too boring". One is to cut ad density and earn from subscriptions, following Fonts Art into the Top Grossing charts. The other is to serve a specific audience, like the "put your idol on your keyboard" idea its TikTok account is testing. Build a dedicated keyboard for a fan community, with fan colors, member stickers and a countdown to a comeback day, rather than adding a photo to a generic template. Be careful with likeness and trademarks, though. K-pop agencies are sensitive about merchandise.

How to get users: Run a Google app campaign with a small worldwide budget for 3–5 days. Check the category charts and the cost per install. Pause for a day or two and see how much organic traffic remains. If it's substantial, raise the budget. TikTok content about idols and "which keyboard sound do you pick" can support this, but this account's data shows that content alone won't bring volume in this category.

When to get in: This business has no window. It's always crowded and won't shut down suddenly. If you do build, the period from the end of the year to just before Lunar New Year is busier, with phone upgrades and fan comebacks clustering, so creatives are easier to produce (inferred).

What would mean you read it wrong: (1) If it restarts paid traffic in December and keeps several thousand organic installs a day after pausing, it has likely proven payback and started scaling up. Then its test results are worth studying closely. (2) If it instead fades out and the developer releases a new keyboard under another name, the template is probably no longer profitable at today's ad prices, and even people who know how to buy traffic can only keep reskinning to test.