Telegram Mini Apps are quietly becoming an app store. Should you launch there first?

Telegram crossed a billion monthly users in 2025, and about half of them open it every single day. Somewhere along the way it also grew an app platform. Mini apps — web apps that run inside the messenger itself — now reach roughly half of Telegram's audience. That's a population of app users comparable to a mid-sized app store, sitting inside a chat client most Western product teams still file under 'messaging'.

The interesting part isn't the technology. A mini app is a web page with a platform SDK; we've had those since forever. The interesting part is what happens to your acquisition funnel when the install step disappears. A user taps a link in a channel and is inside your product two seconds later. No store page. No download. No 'create an account' form, because Telegram hands you the user's identity on entry.

So the question founders keep asking us isn't 'is Telegram big' — it obviously is. It's whether a mini app should come before the iOS and Android builds. For some products the answer is a clear yes. For others it's a quarter of wasted effort. The difference is worth being precise about.

What a mini app actually is

Technically: a web app, opened inside Telegram from a bot, a direct link, an attachment menu, or a post in a channel. Telegram draws the frame, supplies the user's identity and theme, and exposes platform features — payments, haptics, biometrics on supported devices, a back button that behaves. Your team writes the same JavaScript it would write for any web product.

The early ecosystem was dominated by tap-to-earn crypto games, and that reputation lingers. But the use cases that have been growing since look a lot more like ordinary commerce: storefronts that take orders without leaving the chat, booking and ticketing flows, loyalty programs, customer support consoles, even SaaS dashboards that live where the team already talks. Business channels with over 10,000 subscribers grew 39% in 2025 — companies are treating Telegram as an owned channel, and a mini app is what turns that channel into a product surface.

One detail from Telegram's own developer docs that changes the math depending on your business: payments split by what you sell. Digital goods and services inside a mini app must be sold in Telegram Stars, the platform's in-app currency — that's how Telegram keeps Apple and Google happy, since the stores demand their cut of digital purchases made on their devices. Physical goods and offline services are exempt: you plug in a normal payment provider and keep card-processing economics. If you're selling t-shirts, the platform tax conversation mostly doesn't apply to you. If you're selling subscriptions to software, it does.

The economics that make founders look twice

Every mobile product team knows the funnel: ad impression, store page, download, install, open, onboarding, signup. Each arrow loses people. The mini app funnel is: tap, you're in. That's the whole pitch, and it's why the acquisition numbers people report from Telegram look implausible next to app store benchmarks.

The figures floating around the ecosystem deserve their caveats. Agencies that build mini apps claim customer acquisition costs 90–95% below app store equivalents, and quote MVP builds at $1,500–$25,000 in two to six weeks, versus $120,000–$300,000 and the better part of a year for paired native apps. Those numbers come from companies selling the work. Take the precision with salt. But the direction is hard to argue with, because the mechanics explain it: one web codebase instead of two native ones, no review queue, and an install step that doesn't exist.

Monetization inside the platform has matured faster than I expected. Catizen, one of the bigger gaming mini apps, has published numbers showing over 30% of its active users paying, with revenue per user around $12. Subscription patterns are adapting to the audience too — weekly plans now generate over half of mini app subscription revenue, up from around 43% two years ago. Users who won't commit to a month will commit to seven days, and the platform makes re-billing frictionless enough that weekly works.

There's also a quieter trend of small software companies launching entire SaaS products as mini apps first — the argument being that Telegram gives you identity, payments, and a viral distribution loop (people share links in chats; that's the whole growth model) for free, and you graduate to a standalone web product once you've proven someone wants the thing. Whether that's the optimal ecosystem for AI SaaS, as some commentators have started claiming, depends a lot on who your users are. Which brings us to the catch.

The catch, in four parts

Geography. Telegram is the default messenger across Eastern Europe, Central Asia, much of Southeast Asia, the Middle East, and a good chunk of Latin America. In the US it's a niche app. A billion users means nothing if your buyers aren't among them — and for a US-centric B2B product, they mostly aren't. This single factor settles the question for more companies than any other.

Discovery. There's no real app store inside Telegram — no search results page to rank on, no featured lists worth much. Distribution happens through channels, ads, and people forwarding links. That's great if you have an audience or can buy one cheaply; it's nothing if you were counting on organic store traffic. You bring your own demand.

Platform risk. We wrote recently about Apple rejecting vibe-coded apps, and the lesson was about gatekeepers generally: when you build on someone's platform, their rules are your rules. Telegram is one company, with governance considerably more opaque than Apple's, a complicated regulatory history, and the ability to change payment terms or API access overnight. Going Telegram-first doesn't remove platform risk. It moves it somewhere with fewer lawyers watching.

The Stars tax. If you sell digital goods, the app store economics you were escaping partially reappear: Stars bought on an iPhone go through Apple's billing, and Apple takes its share on the way in. The advantage over a native app is real but smaller than the headline suggests. Physical-goods sellers skip this entirely, which makes commerce the cleanest fit the platform has.

How to actually decide

Telegram-first makes sense when a few of these are true:

• Your audience already lives there — you run a channel, your community organizes in groups, or your market is one of the regions where Telegram is the default.
• The product is transactional or conversational by nature: ordering, booking, ticketing, support, anything where the session is short and the trigger is a message.
• You sell physical goods or offline services, so the payments story is clean and the platform tax conversation doesn't apply.
• You want to validate demand before committing native-app money. A mini app is the cheapest way the mobile world currently offers to put a real product in real users' hands.

And skip it when your buyers aren't on Telegram, when the product needs deep device access (background location, health data, serious offline behavior), or when an App Store listing is part of how your brand earns trust. Those are real constraints, and no acquisition-cost spreadsheet beats them.

The part I'd underline either way: a mini app is a web app. Built sensibly, the same codebase serves your own domain, which means the Telegram bet is reversible. Keep your product on a URL you own, treat the mini app as a distribution skin over it, and the platform-risk question loses most of its teeth. That's the version of Telegram-first we'd actually recommend — not moving into the building, just opening a very cheap storefront in it.