How metrics are calculated
The same metrics appear in many blocks on datool.ru channel pages: subscribers, views, average reach, ERR, engagement ER and others. This page explains exactly how each one is calculated, what the change is compared with, why the line is dashed for fresh posts and what the data does not contain at all.
Where they appear
- The “Key numbers” tiles on the channel dashboard (Dashboard) and the summaries on the “Post reach”, “Subscriber engagement”, “Ad statistics” and other tabs.
- The table and tiles of the “Similar channels” tab (Similar channels).
- Ratings and collections (Ratings).
Every tile and block has an “i” or “?” icon — a tooltip with a short definition right on the page.
Subscribers
Subscribers — how many subscribers the channel has now. The daily history is the number of subscribers at the end of each day, Moscow time.
The history starts on the day we began watching the channel: MAX does not provide past numbers. If there is less data than the selected period, the chart is captioned “since tracking began”, and the change is counted from the first known day.
Views and reactions
Views — how many views the posts published in the period have now. These are not “views per day”: MAX does not report how many people looked at the channel yesterday. So old posts barely gain views any more, while yesterday's posts are still collecting them.
Reactions — how many reactions (emoji under a post) the posts published in the period have.
Note
Not all old posts have reactions yet: we are reading them gradually, so engagement for distant periods may be understated.
Post reach
Post reach is its views right now. A fresh post is still collecting views, so all reach metrics are calculated only from posts older than 48 hours (two days) — otherwise fresh posts would pull the result down.
- Average post reach — the posts' views divided by their number. The average is sensitive to single “hits”: one viral post noticeably raises it.
- Median post reach (“Post reach” tab) — the main reach figure: half of the posts got less, half got more. The median ignores single hits and better reflects a typical post.
Example. A channel has 5 posts with 1,000, 1,200, 1,100, 900 and 20,000 views. The average reach is 4,840, the median is 1,100. A typical post of this channel gets about 1,100 views; the average is inflated by one viral post.
Why the end of the line is dashed
Chart points that include posts younger than two days are drawn dashed with hollow circles: they show the reach of all posts as of now, and it will still grow. The tooltip of such a point says “Views are still growing…”. Records and “best day” captions are calculated only from points without fresh posts.
ERR — what share of subscribers sees a post
ERR — post reach divided by the number of subscribers, × 100 %. It shows what part of the subscribers sees a post on average.
- On the dashboard ERR = average post reach ÷ subscribers × 100 %. The dashed line on the tile is the ERR of the previous period of the same length.
- On the “Post reach” tab ERR = median post reach ÷ subscribers × 100 % — so it depends less on viral posts.
Subscribers are taken at the end of the period, and on a daily chart — at the end of each day.
How to read it. ERR 30 % — a typical post is seen by roughly every third subscriber. ERR above 100 % happens: non-subscribers see the post too (reposts, links, search). A very low ERR with many subscribers is a reason to check whether the audience is alive (see Quality index).
Engagement ER
ER (engagement) = reactions ÷ views × 100 %. It shows how often readers react to posts: ER 2 % means two reactions per 100 views.
- We add up the reactions and views of all posts and divide the sums rather than averaging per-post ER: a post with 10 views and one reaction has ER 10 %, and it must not outweigh big posts.
- Only posts whose reactions we have already read and whose views we know are counted.
- ER summaries use posts older than two days; the daily trend uses all posts of the period.
Details — Subscriber engagement.
Reach spread
Spread = P90 ÷ P10: how many times a strong post of the channel outperforms a weak one. P90 is the reach exceeded only by the best 10 % of posts, P10 is the reach below which only the weakest 10 % fall.
A spread of 3 — the channel is stable, a strong post gets three times more than a weak one. A spread of 20 or more — reach jumps around: some posts take off, others barely get any views. With a large spread, an advertiser should look at the median rather than the average reach.
Posting frequency and ads
- Posts per day — how many posts are published per day on average in the period.
- Ad share — what part of the period's posts look like ads (erid labelling, advertising links and other signs — How we detect ads). The load scale: up to 10 % — low, up to 25 % — moderate, up to 50 % — high, beyond — very high. Growth of the ad share is shown in red.
- Ads per week (“Ad statistics” tab) — how many ad posts are published per 7 days on average.
- Mentions — how many times other channels mentioned the channel: links and reposts, excluding ad posts. Reposts — how many times other channels reposted this channel's posts; they are also counted as mentions.
Comparison with the previous period
Under a tile's number there is a change badge and a “was …” line. We compare with the same span right before the selected one: “30 days” — with the previous 30 days.
- Sums (posts, views, reactions, mentions, subscriber growth) are compared as daily averages. The current period ends “now”: “7 days” at 10 a.m. is six full days and ten hours, while the previous period is seven full days. Without this, the badges would show a drop every morning while the channel works as usual. The numbers shown and “was …” remain sums — only the size of the change and the arrow are recalculated.
- Averages and shares (ERR, ER, average reach, median, ad share, subscribers at the end of the period) are compared as they are.
- Percentages (ERR, ER, ad share) change in percentage points (p.p.): ER was 2 % and became 3 % — that is +1 p.p.
- The comparison is shown only if we watched the channel for the whole previous span. Otherwise a channel we have followed for half a year would get “+6000 % per year”. The “All time” period has no comparison.
- A grey badge without an arrow, “no change”, means the change rounds to zero.
Fractions are rounded the usual way: 30.59 % becomes “30.6 %”.
Calculation window: no more than 365 days
Blocks calculated from the posts themselves (best posts, formats, posting time, ads, median and spread of reach, ER) look back no further than the last 365 days: big channels have hundreds of thousands of posts. For “Year”, “All time” and long custom dates such blocks are captioned “for the last 12 months”.
Blocks based on daily statistics (subscribers, views and posts by day) show the full history.
Chart steps
| Period | One chart point on the dashboard |
|---|---|
| 7 and 30 days | a day |
| 90 days | a week (from Monday) |
| Year, all time | a month |
| Custom dates | up to 31 days — a day, up to 120 days — a week, longer — a month |
On the “Post reach”, “Similar channels” and some other tabs the step depends on the window length: up to 92 days — by day, up to 183 — by week, longer — by month. A point's tooltip always includes the year: “5 Oct 2026”, “23 – 29 Mar 2026”, “March 2026”. The point with today is paler — the day is not over yet.
What the data does not contain
- Comments — we do not collect them: neither the texts nor the count.
- Forwards of a post to private messages — MAX does not provide them. How often other channels quote the channel is on the “Mentions” tab.
- Views for a specific day — only the current views of each post.
- Subscribers before tracking began — MAX does not provide past numbers.
- Subscribers' gender, age and city — MAX does not provide such data. Blocks based on the subscriber list (subscriptions and unsubscriptions by day, retention, core readers) exist only for channels verified by the owner — Owner verification.