Telegram Analytics Bot for Group Moderation

A telegram analytics bot reports on what its moderation systems detect: how many violations a group received, which types, how those numbers move over a week or a month, and how risky an individual member looks. The Telegram Bot App is that kind of bot. Its analytics describe safety and enforcement, not growth. You will not find subscriber-count graphs or message-volume vanity metrics here. You will find per-group violation statistics, individual user spam-risk reports, a live feed of moderation actions, and a public API to look up any user's spam rating by Telegram ID.

This guide explains what the analytics cover, where to read them, and where they stop, so you know exactly what kind of telegram analytics bot this is before you add it.

Moderation analytics, not growth analytics

Most tools that call themselves a telegram analytics bot count messages, track active hours, and chart member joins and leaves. The Telegram Bot App does not do that. Every number it shows comes out of moderation: a message had to be scanned, classified, and acted on for it to appear in the data.

So the metrics answer different questions. Instead of "how many people posted today," the dashboard answers "how many of those posts broke a rule, which rule, and how confident was the detection." That makes the data useful for one job: telegram group analytics for safety and policy decisions. It is the right tool if you want to know whether your spam threshold is too loose, which violation type dominates your group, or whether a specific account is worth banning. It is the wrong tool if you want engagement reports for a marketing channel.

Being clear about this matters because the underlying detection runs whether you look at the dashboard or not. The analytics are a window onto work the bot is already doing on every message.

Per-group violation statistics and trends

Open a group in the dashboard and you get an aggregate picture of how moderation has played out in that community.

The overview covers group status, member count, admin count, and configuration checks (profile picture and invite link present). Activity metrics show total messages seen, total violations detected, and total punishment time applied. From those, the dashboard derives a punishment rate per 1,000 messages and an average punishment duration, so a small group and a large one can be compared on the same scale.

The violation breakdown is the centre of telegram moderation analytics. Every detected violation is sorted by type with an exact count: pornographic content, sexual content, toxic language, spam, language violations, profanity, insults, threats, custom badwords, forwarded messages, prohibited media, invite links, and unauthorized bots. This tells you what your group actually struggles with rather than what you assume it does. A group whose breakdown is 80% invite-link spam needs different settings than one dominated by toxicity.

Temporal analysis adds the time dimension. The dashboard shows activity over the last 7 days and the last 30 days, so you can see whether violations are climbing or falling. The trend lines make a spike after a public mention, or a steady decline after you tightened a threshold, easy to read. The full breakdown of these dashboards lives in the Managing Your Bot Through the Dashboard guide.

Individual user spam-risk reports

Group totals tell you about the room. User intelligence reports tell you about a person.

Search any member by name, handle, or Telegram ID and the bot returns a profile built from that user's moderation history. The headline figure is a spam-risk score from 0.0 to 1.0, calculated by the same AI Spam Intelligence model the bot uses to auto-kick accounts at 0.75 and above. The score reflects offense rate, the confidence of past detections, profile characteristics, and behaviour patterns, not a single bad message.

Below the score, the report breaks down confidence statistics (mean, median, 95th percentile), the high-confidence violation rate, and the most common violation types and reasons for that user. Then it lists the full violation history chronologically: each offense with its timestamp, type, confidence level, and the reason that triggered it.

That detail is the point. You can see not only that an account violated rules but how confidently each call was made, which lets you separate a genuine repeat offender from a member who tripped a filter once. The User Intelligence and Group Analytics article walks through reading these reports in full.

The live punishment feed

The dashboard also carries a real-time stream of enforcement. The live punishment feed uses a WebSocket subscription to show moderation actions as they happen, displaying the most recent 20 punishments across the whole system — every group the bot protects, not only yours.

Each entry shows the punished user, the violation type, the detailed reason, the confidence score, and the timestamp. It is the closest thing to watching the bot work. For a new admin deciding whether the detection is too aggressive or too lax, a few minutes of the feed says more than a settings page does. The feed reflects the same enforcement logic documented in the Automated Punishment System guide.

The public spam-rating API

The analytics are not locked inside the web dashboard. The Telegram Bot App exposes a public API that returns a user's spam rating by Telegram ID. You query an ID, you get back that account's risk assessment — the same score the bot uses internally.

This exists so other tools can reuse the rating. A second bot, a registration form, or an external moderation script can check an incoming user against the spam database before letting them in, instead of waiting for them to misbehave first. The rating draws on cross-referenced external spam databases as well as the account's own history, so it carries signal even for a user who has never posted in your group.

What the analytics do not include

Honest data means naming the gaps. This telegram analytics bot does not provide:

• Message-volume, active-hours, or engagement dashboards. Counts here are violation counts, not activity-for-its-own-sake counts.
• Member growth, retention, or join/leave funnel charts. The bot blocks invite-link spam; it does not add or invite members, and it does not chart membership growth.
• Text or OCR analytics. Image scanning classifies images (porn/racy/strict confidence scores) without extracting any text from them.
• Conversational or AI-chat replies. The bot moderates; it does not chat.
• Sentiment "mood of the room" reporting as a growth metric. Sentiment and toxicity detection feed the violation breakdown, not a community-vibe graph.

If you need those, this is not the bot for them, and pretending otherwise would waste your time.

Where the numbers come from

Every analytic traces back to a detection system, all of which run on the free tier for core moderation:

• AI Spam Intelligence (Bayesian risk 0.0–1.0, auto-kick at 0.75) and Spamfinder (an ML content classifier with a 0–100% threshold) produce the spam counts and feed the per-user risk score.
• NSFW image scanning across photos, GIFs, stickers, and profile pictures produces the pornographic, sexual, and prohibited-media counts.
• Sentiment and toxicity detection (toxicity, profanity, insults, threats) and custom badwords produce those violation categories. Language enforcement across roughly 33–43 languages produces language violations.
• Link handling (blocking t.me/ invite links and flagging malicious URLs via Google Safe Browsing) produces the invite-link and link counts.

The free tier includes 500 image scans and 1,000 sentiment analyses per month plus all core moderation. Paid plans (Gold $4.99/mo, Platinum $9.99/mo, Ultimate $49.99/mo, around 20% off annually) raise those quotas; they do not add a different class of analytics. The full anti-spam stack behind the spam numbers is covered in the Anti-Spam Protection guide.

Frequently asked questions

Is this a growth analytics bot for Telegram?

No. It is a moderation analytics bot. Its telegram group analytics measure violations, enforcement, and per-user spam risk. It does not report message volume, active hours, member growth, or engagement. If you want subscriber and activity dashboards, this bot is not built for that.

What can the individual user reports actually tell me?

Each report gives a spam-risk score (0.0–1.0), confidence statistics for that user's past detections, their most common violation types and reasons, and a timestamped history of every offense with its confidence level. That is enough to judge whether an account is a repeat offender or a one-off false positive before you act.

How real-time is the punishment feed?

It streams over a WebSocket and shows the most recent 20 moderation actions across every group the bot protects, with user, violation type, reason, confidence, and timestamp on each. New punishments appear as they happen.

What is the public spam-rating API for?

It returns any user's spam rating by Telegram ID so external tools can reuse the assessment. For example, you can screen new joiners against the spam database from your own bot or form. It draws on the account's history and cross-referenced external spam databases.

Do I need a paid plan to see the analytics?

No. Core moderation and its analytics run on the free tier, which includes 500 image scans and 1,000 sentiment analyses per month. Paid plans (Gold, Platinum, Ultimate; about 20% off annually) raise the scan quotas rather than gating a separate analytics product.

Does the bot count messages or track who is most active?

No. The only message figure is a total used to scale violation rates (punishments per 1,000 messages). There is no per-member activity ranking, no active-hours chart, and no engagement scoring as a feature.

Get started

If moderation analytics are what you want (violation trends, spam-risk reports, a live enforcement feed, and a queryable spam rating), add the bot to a group first. The dashboard becomes available once it is present and you are an admin. The How to Add the Bot to Your Group guide takes you through it, or start from the Telegram Bot App home page.