Telegram Channel Stats — Subscribers, Reach & Engagement Rate

One audit row per public Telegram channel: subscriber count, median / mean / max views, engagement rate as a percentage of subscribers, posting frequency, last-post date and the channel's own photo/video/link totals.

Built for the job Telegram itself won't do — deciding whether a channel is worth buying a placement on. A subscriber count alone is the number channel owners quote; median views against subscribers is the number that says whether those subscribers are real and still reading.

No login. No phone number. No API key. No Telegram account of any kind.

---

What one row looks like

Measured, not illustrative — this is a real row from t.me/s/telegram:

{
  "channel": "telegram",
  "channelTitle": "Telegram News",
  "channelDescription": "The official Telegram on Telegram. Much recursion. Very Telegram. Wow.",
  "channelPhoto": "https://cdn1.telesco.pe/file/fzfcAqQE856VYhKMfz-3YI7W85c…",
  "subscribers": 9540000,
  "photosCount": 16,
  "videosCount": 228,
  "linksCount": 378,
  "sampledPosts": 20,
  "medianViews": 1370000,
  "meanViews": 1498850,
  "maxViews": 2860000,
  "engagementRatePct": 14.4,
  "postsPerWeek": 1.3,
  "newestPostAt": "2026-08-26T19:12:34.000Z",
  "oldestSampledPostAt": "2026-05-14T16:08:31.000Z",
  "withMediaInSample": 16,
  "forwardedInSample": 0
}

field	what it is
subscribers	A real number. Telegram renders "9.55M"; this expands it to 9550000
medianViews	Median views across the 20 most recent posts — the headline reach number
meanViews · maxViews	Mean and best post. A mean far above the median means one viral post is carrying the average
engagementRatePct	medianViews ÷ subscribers × 100. The audience-quality number
postsPerWeek	Cadence, derived from the real dates of the sampled posts
newestPostAt	Is this channel still alive? A last post six months ago is the single most useful disqualifier
withMediaInSample · forwardedInSample	Content mix — a channel that is 20/20 forwards is an aggregator, not an audience
photosCount · videosCount · linksCount	The channel's own lifetime totals
recentPosts	Optional: the 20 posts the maths was computed from, so you can check the numbers

Input

input	default	what it does
channels	["telegram","durov"]	Handles or links — telegram, @telegram, https://t.me/telegram all work. Duplicates are removed, so a channel is never charged twice in a run
includeRecentPosts	false	Attaches the 20 sampled posts — id, date, views, length, media flag
proxyConfiguration	Apify Proxy	Datacenter is sufficient

One request per channel

The audit reads t.me/s/<channel> — a single page that carries both the channel header and its 20 most recent posts. Every figure above comes off that one page, with no second request and no paging, which is why a 500-channel audit finishes in minutes and costs $0.45.

Numbers are null when they are unknown, never zero

A channel that doesn't publish its subscriber count reports engagementRatePct: null — not 0. Zero would read as "dead channel" when the truth is "the number wasn't published," and a buyer sorting a list by engagement would bury a good channel on a fabricated figure. Same for medianViews on a channel with no visible posts, and postsPerWeek where a single post can't establish a cadence. Every row carries sampledPosts so you can see what each figure was computed from.

Channels with no public preview are reported, not billed

Telegram answers a normal page for a private channel, a preview its owner disabled, or a handle that isn't a channel at all. Those are listed by name under noPublicPreview in the run summary and are not charged — so a run against a list of 500 half-stale handles bills only for the ones that produced an audit, and the run summary tells you which of your handles produced nothing rather than leaving you to assume the channel is dead.

Pricing

Pay per event: $0.0009 per channel audited, and no run-start fee.

Nobody in this category sells channel stats as a standalone row — the alternatives bundle it as a surcharge on a message scrape, so auditing 500 channels means also paying to download messages you don't want. Against the closest per-channel rates: $0.0009 here versus $0.001 and $0.002 + a $0.005 start fee. And no competitor derives median views, engagement rate or posts-per-week at all — those are the reason to run this.

A 500-channel audit is $0.45. A run that finds nothing costs nothing.

Source

t.me/s/<channel> is Telegram's own public web preview, served to any visitor. The host publishes no robots.txt at all. Public channels are published by their owners to be read publicly; you are responsible for what you do with the data, which for outreach means the marketing and data-protection rules wherever you are sending from.