View Counts Are the Cheapest Credibility Signal in Telegram OSINT
If you only have one minute to judge a Telegram channel's credibility, spend it on the view-to-subscriber ratio. It is free, public, and almost impossible to fake at scale. Where to see it Public channels exp
If you only have one minute to judge a Telegram channel's credibility, spend it on the view-to-subscriber ratio. It is free, public, and almost impossible to fake at scale.
Where to see it
Public channels expose a web preview at https://t.me/s/<channel>. Channels above roughly 1000 subscribers render a per-post view counter. Below that threshold, no counter is exposed - which is itself useful: small channels are unfalsifiable, so treat their claims as unverified.
Reading the ratio
From scraping a few hundred public channels, a few patterns repeat:
- Healthy: 15-35% of subscribers view a post within the first day. News/analysis channels sit near the low end, meme channels near the high end.
- Inflated: 12k subscribers, 200-400 views per post. Classic purchased-audience footprint. The ratio is stable across every post - real audiences have variance, bot audiences do not.
- Washing: view counts that spike on reposted content from 2-3 partner channels. The repost graph reveals the ring.
The one-minute triage
- Open
t.me/s/<channel>, note subscriber count. - Check 6-8 recent posts' view counts. Compute min/max variance.
- Near-zero variance + ratio under 5% = paid audience, exclude from collection.
This check costs less than a minute and removes most of the noise before any text analysis begins. It is rule 4 of the 11-rule de-junking checklist in my Telegram & Web OSINT Bundle; a free sample of the checklist is here.
Method details and scraping notes on this blog; pipeline runs free on GitHub Actions.
Originally published by Dev.to Security. Aggregated on AIWithGhost for educational purposes — full credit and traffic to the original publisher.