The hardest problem in any marketplace is the beginning. Buyers will not come without sellers, sellers will not stay without buyers, and every founder who has lived through it knows the specific misery of a platform that looks busy in the dashboard and feels dead to anyone using it.
Telegram helps here more than most channels, for one reason: the people you need are already gathered in groups, talking about the exact problem your marketplace solves. A freight group where shippers complain about capacity is a list of demand. A group of independent suppliers comparing notes is a list of supply. You can read both before you contact anyone.
What follows is how to work that advantage without ending up with two thousand accounts that never trade.
The practical playbook
Pick one narrow slice of your market, work out which side is harder to get, and recruit that side first. Sellers usually. Then bring buyers in with a promise you can actually keep, because there is now real supply behind it. The number to watch is completed deals in that one slice, not total signups: a marketplace with 5,000 registered users and no transactions is not a marketplace, it is a mailing list.
What works here
01How to pick a slice narrow enough to actually fill.
02Why buyers and sellers need completely different questions.
03The order to recruit them in, and why alternating fails.
04The one number that tells you if it is working.
Start absurdly narrow
The instinct is to launch broad so the market is bigger. It is exactly backwards.
A marketplace for "logistics" needs an enormous number of participants before any given buyer finds a matching seller. A marketplace for refrigerated transport between two specific countries needs maybe fifteen suppliers to feel full. Same effort, completely different outcome.
So define your slice until you can say it in one sentence that includes a place, a category, and a size of deal. If you cannot, your scraped lists will be broad, your messages will be generic, and nothing will transact. Narrowness is not a limitation here, it is the thing that makes the early stage survivable.
A good test: can you name the ten sellers who would make your marketplace useful tomorrow? If the answer is "it depends", the slice is still too wide.
Ask each side a different question
Recruiting buyers and sellers with the same message is the most common mistake, and it shows immediately.
A seller wants to know whether there is real demand and whether they will be paid. A buyer wants to know whether anyone can actually deliver and what it costs. Sending both the same "join our platform" message answers neither, and it signals that you have not thought about their side of the trade.
- For both sides, the same basic checks still apply: is this a real business, is contacting them appropriate, and have they asked not to be contacted before.
| When you find a seller, check | When you find a buyer, check |
|---|---|
| Do they have spare capacity right now? | Do they have a live need, not a someday need? |
| What can they realistically deliver, and where? | Can this person actually authorise a purchase? |
| How fast do they respond to enquiries? | How are they solving it today, and what does it cost them? |
| Are they credible enough to put in front of a buyer? | What would make them try somewhere new? |
Recruit one side at a time
Alternating between buyers and sellers feels balanced. In practice it hides which side is actually stuck.
Get enough credible sellers to serve your first handful of buyers. Then go find those buyers, with a specific promise you can keep because the supply is genuinely there. When a buyer wants something nobody can deliver, that gap is not a failure. It is your next recruiting brief, and it is far more precise than any market research.
This is the well-documented chicken-and-egg problem of two-sided markets: economists Rochet and Tirole showed that getting both sides on board is the platform's core problem, and that *who* participates matters more than how many. For you, the practical version is simply this: fix one side at a time, and measure each slice separately.
Averages lie here. Plenty of sellers in one category does not make up for none in another. A buyer only experiences their own slice, and to them a half-empty marketplace is an empty one.
Handle the first matches by hand
Early on, you are not running a platform. You are personally introducing people, and that is the correct way to do it.
When a seller says yes, confirm what they can really deliver. When a buyer says yes, confirm what they actually need and when. Then introduce them yourself and stay in the conversation until something happens or clearly does not.
Every failed match teaches you something a dashboard never would: whether it broke on price, on trust, on timing, or because you were missing a supplier attribute you did not know mattered. Keep those reasons written down next to the contact. After twenty of them, you will know exactly what your marketplace is missing.
Count deals, not signups
Registration numbers are the easiest metric to grow and the easiest to fool yourself with.
- Completed transactions in your one slice. The only number that proves the model works.
- Time from joining to first deal. If it is measured in months, participants leave before it happens.
- Requests nobody could fill. Your recruiting list for next month, handed to you for free.
- Sellers who have been idle for weeks. They are quietly leaving, whatever the signup chart says.
- People who come back for a second deal. The clearest sign you have built something real.
How we checked this guide
This applies established marketplace thinking to Telegram outreach. It does not cover the parts that are specific to your market: fraud checks, payment handling, regulated transactions, and the contractual questions that come with introducing two businesses to each other all need their own design.
Apply it to your market
Telegram is unusually good at the early marketplace stage because the people you need are already grouped by the problem you solve, and you can talk to them directly. What it cannot do is rescue a slice that is too broad to fill. Make every campaign answer one question: which side am I short of, and in which corner of the market?
Keep sourcing, conversations and follow-up in one place
TeleBoost gives marketplace teams group scraping, filtered lists, campaigns, a unified inbox, tickets and analytics, with separate workspaces if you run more than one market.
Keep learning