The member count in a group's header and the number of rows that end up in your list are different numbers, sometimes by a lot. Better to know that before plans are built on the larger one.
What shrinks the export
The platform's own cap
Telegram does not hand over the full membership of large groups. Exactly how much it does hand over is not published and changes, so any precise figure you read is somebody's observation on their own accounts rather than a rule — what is actually known.
Hidden members
People who restricted their visibility do not appear in the list at all.
Unusable rows
Of what does export, a share has no username (why that matters), and a share are bots or deleted accounts. Both go out during cleaning — cleaning a database.
Closed messaging
Finally, some of the remainder cannot be written to at all — why part of your list is unreachable.
How to plan
Not from the group size but from a trial run. Collect one group typical of your niche, clean it, and apply the surviving share to the rest. That is the only honest estimate available.
If the volume is short
- More groups rather than a bigger one. Ten medium chats yield more usable rows than one huge one.
- Commenters instead of members. Fewer, warmer — collecting commenters.
- Active members instead of all of them. Same principle — finding active users.
And the conclusion that saves the most: a small accurate list outperforms a large random one — segmenting an audience.
How Neurogram does this
Everything above is manual work that runs into the number of hours in a day. Neurogram does it on your own accounts and without you: it collects channels and audiences, writes comments and replies with a language model, and keeps limits and quiet hours so the accounts survive. See pricing or look at your case.