6 ratings put it at No. 9 on Croatia's Top Grossing: what a "local Cal AI" looks like in the Balkans

Kalorino is an AI calorie-tracking app built for the former Yugoslav region. Its selling points are recognizing local dishes like ćevapi, burek and sarma, and speaking the local language. It peaked at No. 9 on Croatia's App Store Health & Fitness Top Grossing, yet across the whole region it has only 6 ratings. This piece covers how it earned these few dollars, and whether this "pick a small language, and hit it with one local dish" template is worth taking to other countries.

How many calories is a plate of ćevapi

In 2026 Sep 29, the TikTok account @kalorino.app posted a video captioned "Ćevapi u somunu. Koliko bi ti procijenio/la kalorija? 👀" (ćevapi, the Balkan grilled sausages, stuffed in bread: how many calories would you guess?), tagged #sarajevo.

The video got 110k views and 572 likes. The same account's six or seven other posts that week covered burek, breakfast fried eggs, Pilates, and eating chocolate when stressed, each getting between two or three hundred and over six hundred views (TikTok, 11-06). 110k views produced only 0.5% likes, and the account still has just 72 followers, so this video was most likely boosted with paid money (inference).

ćevapi is the core of its whole pitch. The second paragraph of the App Store description reads: "Ask a big app how many calories ćevapi has and it says '4 190 calories.' A real serving is 5 or 10 pieces with bread, close to 600 to 900 calories. Ask about burek and it gives you British cereal." Locals get it immediately, which makes it easy to spread as creative.

What it is

Snap a photo of your plate, say "two fried eggs and a serving of pita," type it in, or scan a barcode, and within seconds it returns calories, protein, carbs and fat, logged against your daily goal. Portions can be entered as "pieces, servings" or in grams. The interface comes in seven languages: Bosnian, Croatian, Serbian, Montenegrin, Macedonian, Slovenian and English.

No free tier. Web pricing is €7.99/month or €34.99/year, annual plans get a 3-day trial, and on the web you enter a card up front and are charged only after the trial ends (kalorino.app pricing page). The iOS app launched in 2026 Sep 14; the Android version came earlier, on Google Play in Aug 26, now 5,000+ installs (Play listing, 11-06).

The developer, Astirna LLC, is registered in Delaware, USA. Its website says it does performance marketing, and it also has a few small products like wedding photos, tinnitus, and a Chrome new-tab page. It isn't an indie developer who wanted to build a health app; it's a studio that runs ads and launches products on the side.

How the 6 ratings got onto the bestseller chart

(App Store Croatia storefront, daily rank; No. 49–90 on the same chart in the Slovenia storefront)

In a single day it dropped from 11 to 66, then to 142, and two days later was back at 39. Rankings jumping like this show that a few subscriptions a day is enough to move up or down this chart. On the same day's chart, Cal AI was No. 5 with 1,588 ratings, MacroFactor No. 8 with 48 ratings, and Welmi No. 9 with 36 ratings (App Store Croatia storefront, 11-06). Croatia has under 4M people, so the bar for the Health & Fitness Top Grossing is low.

So "charting in the top grossing" doesn't mean it has broken out; it only means it really did sell a few €35 annual subscriptions a day. We can't see chart data for its home markets, Serbia and Bosnia, and its money is most likely coming through the web, so store rankings only show part of it.

Where users come from

• Web questionnaire + paid acquisition. Confirmed. The homepage button "Izradi svoj plan" (build your plan) leads to a questionnaire; after answering, you subscribe on the web via Stripe, and only after the subscription succeeds do you get download links for iPhone and Android. The website code has a Google Ads conversion tag, recording utm parameters and click IDs from Facebook, Google, TikTok and Bing; this attribution was wired up as early as 18 Sep, four days after the iOS launch. Web payments bypass Apple's and Google's 15–30% commission and are compliant in the EU.
• TikTok. On 28 and 29 September it posted 7 videos in two days, like batch-testing creatives; the ćevapi one was one of the results.
• Instagram / Facebook. It has accounts (IG 506 followers), using the same ćevapi creatives; whether it's running Meta ads was not verified.
• Store search. The app name is literally "AI brojač kalorija" (AI calorie counter), the words locals would actually search, with little competition in this language (inference).

No media coverage; it can't be found on Reddit, X or Product Hunt, and the founder hasn't disclosed any revenue.

Who it's competing with

The big apps are all on Croatia's charts, but they've only translated the interface: MyFitnessPal has 6,468 ratings and ranks No. 2 on Top Grossing, Cal AI No. 5, and Yazio and Lifesum sit at No. 8–20, with food databases built for Western Europe.

Several products already target the same point, "local dishes + AI", all appearing 2026:

Other small languages have people using the same template too, such as Lithuania's Maistiqo, launched in July, with 13 ratings and a best chart position of No. 25 in Lithuania's Health & Fitness.

Kalorino ranks higher than these few, and it doesn't look like the product has anything others can't match. The edge seems to be sharper paid acquisition and web funnel execution: seven languages rolled out together, web card-up-front trials, and attribution wired up in the first week.

If it were you

One-line verdict: Kalorino hasn't broken out yet. What's worth studying is the template behind it: find a small-language market where a big app has only translated the interface, use a local dish people eat every day as the hook, and collect money with a web questionnaire plus subscription. The template is cheap and replicable, but the ceiling of any single market is very low, and in the former-Yugoslav region three or four players are already competing.

Who it's for: small teams already able to run Meta, TikTok and Google ads, with a "questionnaire → web subscription → app download" funnel; ideally someone who understands the food and speech habits of a particular small-language market.

Who it's not for: indie developers who can't buy traffic and count on organic store traffic; anyone hoping to build big revenue from a single country. Such markets can chart with a few sales a day, which also shows how few sales a day that really is.

How to do it:

• Don't crowd into the former Yugoslav region. Pick a language where a big app has only translated the interface and the local dishes are distinctive: the three Baltic states, Hungary, Romania, Bulgaria, Greece, Turkey and Czechia can each be checked first, to see how high the Health & Fitness Top Grossing bar is there and whether anyone is already doing this (inference).
• Build the dish library and portion sizes first, not AI. Photo recognition is something anyone can plug in. What locals care about is "how many pieces in a sarma" and "how big is one pita," and that part has to be curated by hand and set up so people can log by count.
• Copy its pricing: no free tier, annual plan with a short trial, web checkout to skip the store cut.

How to get users: One dish per video. Comparison hooks like "the big app says this is 190 kcal, it's actually 700" work best. Test a dozen-plus creatives on TikTok with a small budget, scale whichever one breaks out, then port it to Instagram and Facebook. In the store, name the app with the local-language "AI calorie counter" keyword.

When to move: each small-language market has its own window, and whoever arrives first takes the local search terms and the dish in the creatives. 2026 already has several countries with similar apps, so the later you start, the fewer open languages you can pick from. The former Yugoslav markets are already late.

What would mean this read is wrong: if Kalorino's rating climbs to the hundreds over the next two or three months, it holds steadily in the top 10 of the Croatian Top Grossing, or it charts in more countries, then localization really does lift conversion and this template is worth more than this write-up suggests. If its rank drops out of the top 200 and TikTok goes quiet, it was just one studio's test that didn't pan out.