8 min read
Telegram Bot or Mobile App: What a Small Business Should Choose
“We need an app” is how every third automation conversation starts. Almost always the real task turns out to be taking orders and booking clients — and for that an app is neither the best tool nor remotely the cheapest. Here's the honest comparison, including the cases where an app really is the right call.
The confusion exists because the word “app” sounds like a synonym for “serious”. The reasoning goes: big companies have apps, so we should too. But big companies don't build apps to take orders — they build them for customers who open them daily, and they spend budgets comparable to a small business's annual turnover doing it.
The key difference between a bot and an app isn't even the build cost. It's that an app has to be installed. Between your ad and the first order sits a step where most people drop out: find it in the store, wait for the download, grant permissions, register. A bot opens with one tap inside a messenger that's already on the phone.
What breaks when you bet on an app
- Installation. Someone saw your ad and is ready to order now — and you're asking them to install software first. A share of them drops out right there, and it's the most expensive loss in the whole funnel.
- Space on the phone. An app used once a month is the first thing deleted when storage runs out.
- Store review. Every update goes through Apple's and Google's review — from a few hours to several days — and a rejection isn't yours to control.
- Two platforms. iOS and Android are effectively two builds and two ongoing maintenance streams, unless you accept cross-platform compromises.
- Standing fees. An Apple developer account costs $99 a year, Google Play $25 once. Not large sums, but they're permanent — even if nobody uses the app.
- Paying for installs. Building the app isn't the end; you then pay to get it installed. A bot has no such line in the budget at all.
What each costs and how long it takes
Telegram bot
from $300 · 1–2 weeksTakes orders, books appointments, accepts payments. Opens with one tap, nothing to install, works identically on iOS and Android — because it isn't software, it's a conversation in a messenger.
More about this serviceWebsite or web shop
from $500 · 1–3 weeksWhat you need when people should find you in Google and see the full catalogue. A page opens straight from the link in your ad — again, with no install step in between.
More about this serviceMobile app
thousands of dollars · monthsThis isn't a service I offer, so I'll put it plainly: studios charge thousands per platform and take months, plus ongoing maintenance for every iOS and Android release. For taking orders specifically, that rarely pays for itself.
When an app genuinely is justified
- People use the service daily — trackers, fitness, learning, work tools. Then the icon on the home screen works for you rather than against you.
- It has to work offline: warehouse operations, deliveries in basements and on the road, field measurements.
- It needs deep access to the phone: background location, scanning, recognition, the camera as a core function rather than a way to send a photo.
- You already have a large loyal audience that will definitely install it. That is: business first, app second — not the other way round.
- Push notifications are critical and your audience doesn't use Telegram — then the alternatives really are limited.
The third option people forget
Between a bot and an app sits a middle form — a PWA: a website that can be added to the phone's home screen. It gets an icon, opens full-screen without an address bar, can work offline and send notifications. And it stays a website: no stores, no review queue, an update ships the moment you release it.
This costs hundreds rather than thousands, as an add-on to a site or an MVP. A PWA doesn't replace a native app where deep hardware access is required, but it covers most of the “we want an icon on the home screen” requests — in weeks rather than months.
What to pick for your actual task
Taking orders and booking clients
Telegram botThe shortest path from ad to order: the person taps a link and lands straight in the conversation. I broke down what such a bot costs, component by component, in a separate article.
Read articleBeing found in search
WebsiteGoogle doesn't index bots — only a website can appear in search results. If customers look for your service online, that's where to start, with the bot alongside it rather than instead of it.
More about this serviceValidating a new product idea
MVPWhen you're not taking orders but building a new service, what you need is neither a bot nor an app but a working version of the product — with sign-up, payments and an admin panel, in 2–4 weeks.
More about this serviceHow to decide in a minute
First question to ask yourself: how often will the customer use this? Once a month or less — a bot or a website, and an app will simply eat the budget. Several times a week or more — then it's worth pricing an app or a PWA.
Second: where do people come from? If it's ads, Instagram or word of mouth, every extra step between the ad and the order costs you customers. A bot removes that step entirely, which is exactly why it almost always wins at taking orders.
Frequently asked questions
Customers judge convenience, not technology. A bot that takes an order in three taps and confirms it immediately looks more serious than an app they have to install first. What actually looks unserious is something else — replying to messages the next day.
For them there's still the website form and ordinary contact details — which is why a bot and a site usually work as a pair rather than as substitutes. And if your audience barely uses messengers at all, that's clear from the very first look at the task, and I'll say so directly.
Yes, and that's the sensible order. Within a few weeks a bot shows you real numbers: how many people order, whether they come back, what they ask most often. With that data the decision about an app becomes evidence-based rather than intuitive — and the app itself comes out cheaper, because the flows are already proven on real people.
The review itself is usually a day to a few days, but that's only the last step. Before it comes development for two platforms and preparing the store listings. And you go through review again on every update, whereas a bot updates instantly.
Not sure which one you need?
Describe how customers reach you and what they need to do — I'll tell you straight whether you need a bot, a website or something more. The conversation is free.