Telegram channel statistics explained

Medians for every category of the English segment: half the channels are above, half below. Recalculated overnight from our own daily snapshots; the date of the snapshot sits above the table.

Engagement is median views divided by subscribers. Posts are the last 30 days. Score is Place Score, 0 to 100.

Median post views divided by subscribers. A thousand subscribers and two hundred views is 20 %.

For a reader this says whether the channel is alive or asleep: subscriptions can be bought and abandoned, views are harder to fake. For an owner it says how many people actually see the work.

Every category has its own norm, and it is in the table above, column «Engagement». Size matters too: under a thousand subscribers the benchmark is 60 %, for a channel above a million it is 10 %.

Engagement above 150 % is treated as impossible: that many views do not happen without inflation, and such a channel scores zero on this row with a note.

What counts is not the gain itself but its shape — from daily measurements over 30 days. Fewer than seven measurements and the row stays empty.

A steady chart earns the full weight. Two things lower it: a day that stands out from the rest by more than three standard deviations, and a month where the channel lost subscribers on more than half the days.

For a reader this answers «is this channel inflated»: a bought thousand arrives within an hour and shows on the chart as one step. For an owner it shows whether people stay after a surge.

Two numbers from posts: reactions per view and forwards per view. The benchmark is roughly 3 reactions per 100 views and 1 forward per 100. Reactions carry 60 % of this row, forwards 40 %.

A forward is worth more than a reaction: it brings a new person, while a reaction stays inside. A news channel is often read in silence — high engagement, low reactions — and that is a genre, not a defect.

The row does not appear immediately: reactions are counted from collected posts, and a newly listed channel has none yet.

The benchmark is about 20 posts per 30 days. Silence and flood lower the row equally: past 60 posts a month each one is worth less, because subscribers stop opening them.

The catalogue itself posts more often than the benchmark: the median for each category is in the table above, and for news categories it runs two or three times past twenty. That is not a contradiction — twenty posts earn the full weight, they are not a ceiling; a news feed and a personal blog live at different rhythms, and a channel is worth comparing with its own category rather than with the benchmark of the formula.

Two parts. Age: a year of existence earns the full part, a younger channel gets a share of it. Profile: description, avatar and link in place make up the second part.

For a reader, age says the channel was not built yesterday for a one-off. For an owner it says that part of this row arrives on its own, with time, and part depends on whether the profile is filled in.

A review is left by a person with a confirmed Telegram account, and its weight depends on how confirmed that account is. The other five rows are measured by machine — this one comes from outside.

With no reviews the row does not become a zero: it drops out of the calculation together with its weight, and the score is computed from the rows that were measured. A channel without reviews is not punished — there is simply less known about it.