How to Report Someone on Telegram and What Follows

The button takes two taps and then nothing visibly happens. No confirmation that matters, no case number, no reply. That silence is why people press it twice, or press it and then wonder whether they did the right thing at all.

Here is the short version. The report menu is not stored in your app, the server sends it and can ask you for a written comment. Reporting a message and reporting an account are different calls with different reach. The report goes to human moderators, and the platform publishes how much it removes. And the same button is what puts ordinary accounts into a limited state they then have to argue their way out of.

What follows covers what a report actually sends, why the menu is server driven, where the button sits on each client, the bar that appears above a stranger, message reports against account reports, the ten reasons and which to pick, reporting inside a group, scammers and impersonators, the serious channels for illegal content, reporting an advertisement, what happens afterwards, what reporting does not do, the alternatives, what being reported feels like from the other side, and what teams running many accounts get wrong. Every rule quoted below was pulled from Telegram's own documentation and safety pages on 13 August 2026.

What Reporting Actually Sends

Knowing what leaves your device makes the rest of this far less mysterious.

The message, not your opinion of it

A report carries the identifiers of the messages you selected and the peer they came from. It does not carry a summary you wrote unless the flow asks for one, and it does not carry your reading of the situation. The moderator sees the content and the context around it. What they cannot see is everything that happened before you started selecting.

That is why selecting the right messages matters more than most people expect. One message chosen well says more than ten chosen carelessly, because the person reading it has to decide from what you sent rather than from what you meant. Pick the ones that would convince a stranger with no history in the conversation. That is the actual standard, and applying it takes about ten seconds longer than not applying it.

Your identity is not the point

The report is attributed to your account on the server side, since somebody has to be accountable for false reports. What it is not is a message to the other person, and nothing about it appears in the chat. No badge appears on your profile either, and nothing changes in how the chat looks to you.

People assume the other side gets told who complained. Nothing in the documented flow sends that, and the reported account learns about a restriction rather than about a reporter. The distinction matters most in small groups, where people assume the complaint can be traced back by elimination.

Nothing comes back to you

There is no ticket, no status page and no reply. The call returns a result meaning the report was sent and no further action is required, and that is the entire receipt you get. Screenshot the moment if you need a record, because nothing on the platform will remember it for you.

The Menu Is Not in the App

This is the part almost nothing written about reporting mentions, and it explains why the options keep changing.

The server writes the options

The report call can answer with a menu rather than a confirmation. That answer carries a title and a list of options, and the client is told to show them and call again with whichever one you pick. The options are not compiled into your app. They arrive with the answer, which is why an option can exist today and be gone next week.

So the list you see is decided per case, on the server, at the moment you tap. Two people reporting two different things in the same week can be shown different menus, and a menu can change without any app update at all. Guides that print a fixed list of choices are describing one screenshot rather than the system.

It can ask you to write something

A second possible answer asks for a comment. It carries a flag saying whether the comment is optional, which means the platform can decide that a particular category of report is worth nothing without a sentence from you. That flag is set per case, so the same category can be optional once and required the next time.

When that screen appears, treat it as the only place your judgement enters the process. Write what a stranger would need in order to see the problem, not what you feel about it. Two plain sentences beat a paragraph of adjectives, and the moderator is reading dozens of these.

Only the third answer means done

The flow ends when the call returns the result meaning the report was sent. Until then you are still inside a conversation with the server, which is why abandoning the screen halfway leaves nothing behind. Closing the app at the comment step sends nothing at all, which is the most common way a report quietly fails.

Where the Button Lives on Each Client

The gesture differs per platform, and the help pages give all of them.

Android, iOS, desktop and web

On Android you tap the message and choose Report from the menu. On iOS you press and hold it. On Desktop, Web and the macOS app you right click the message and select Report, then choose a reason. The gesture differs but the flow underneath is identical, so nothing about your choice of device changes the outcome.

Reporting a whole chat

Opening the profile of a user, group or channel gives a report entry that applies to the peer rather than to a single message. It is the right control when the problem is the account itself. Use it when there is no single message worth pointing at, or when the messages are already gone.

The distinction matters because the two paths reach different places in the protocol, which the next section covers. Reporting the peer when you mean one message throws away the evidence you had. That is a one way loss, because you cannot attach the message afterwards.

Reporting from a preview

You do not have to join a channel to report it. Opening a public link shows a preview with the same report entry, which is how most people report content they were sent rather than content they follow. The join button and the report button sit on the same screen, and you can use the second without touching the first. Most people never notice the second, which is why so many reports come from members rather than from passers by.

This is worth knowing before you go looking for evidence. Joining a channel to document it puts your account in its member list, exposes you to its content, and in some cases marks you as an audience for whatever the operator does next. If a preview is enough, stop at the preview. The only reason to go further is evidence that lives behind the join, and that is rarer than it sounds.

The Bar That Appears Above a Stranger

When somebody you do not know writes to you, the app puts a strip at the top of the chat, and that strip is not decoration.

What the strip is made of

The server sends a settings object for that chat with a set of flags, and the client draws whichever ones are true. One says whether you can add the person as a contact, one whether you can block them, and one whether you can report them for spam. The client draws a strip from those flags, so the strip is a picture of server state rather than a fixed feature.

The wording of that last flag is worth reading closely, because the documentation says whether you can still report the user. It is a state that expires rather than a permanent control, and it goes away once the conversation stops looking one sided. Report before you reply if you already know what the message is, because replying closes that door.

The hints beside it

The same object can carry the month the account registered, the country its number came from, and the dates when the name and the photo were last changed. Clients show these above a stranger's first message. They are supplied by the server rather than inferred by the app, which is why they are consistent across devices.

Read them before deciding anything. An account registered this month, with a number from a country unrelated to its story, that changed its name last week, is telling you what it is without any research on your part. None of those three facts is proof on its own, and all three together rarely mean anything innocent.

Why it disappears

Reply once and the bar goes. The platform treats a reply as evidence that the conversation is wanted, which is reasonable and occasionally inconvenient. A single question mark counts, so decide before you answer rather than afterwards.

Reporting a Message Versus Reporting an Account

Three different calls sit behind what looks like one feature, and they are not interchangeable.

The message report

The main call takes a peer, a list of message identifiers, an option and a comment. It is the one that returns the menu described above, and it is the only path that carries specific evidence. Everything useful about reporting flows through this call rather than through the others.

Use it when the problem is something that was said or posted. The moderator gets to read the thing itself rather than a description of it. Select every message that forms the pattern, not only the worst one, because a pattern is what a single message cannot show.

The peer report

A separate call reports the peer with a reason chosen from a fixed list and a comment. There are no message identifiers in it at all, so it is a judgement about an account rather than about content. The reason you pick is doing all the work, which makes this the one call where the list below matters most.

This is the right call for an impersonating profile, a channel whose whole purpose is the problem, or a user whose messages you have already deleted. The comment field carries more weight here than anywhere else, because it is the only thing distinguishing your report from a reflex.

The one tap spam report

The bar's button maps to a third call that takes only the peer. No reason, no comment, no message list. It is the fastest and the least informative of the three, which is exactly why it exists. It is built for the moment a stranger writes and you want them gone before reading twice.

Its documented refusals are worth knowing if you ever see one. It answers that the peer is invalid, that the message identifier is invalid, or that you have not joined the channel you are trying to report. That last one is the common case, and the fix is to report from the preview rather than from a stale link. Seeing that refusal usually means the link you kept is older than your membership.

The Ten Reasons, and Which One to Pick

The reason list is fixed in the protocol, and picking the wrong one is the most common way a real complaint gets nowhere.

The full list

The documented reasons are spam, violence, pornography, child abuse, copyright, an irrelevant location group, impersonation, illegal drugs, disclosure of personal details, and other. Ten categories, fixed in the protocol, and unchanged in shape for years.

Two of those are narrower than they sound. Impersonation covers someone pretending to be a person or a brand, not someone you dislike, and the personal details reason covers publishing somebody's information rather than merely mentioning them. Both get used loosely and both lose weight when they are, because a queue full of misfiled reports is slower for everybody in it.

Why the choice matters

Reports are triaged. A queue for child abuse material is not the queue for unsolicited advertising, and putting a serious complaint into the wrong queue slows it down rather than speeding it up. Choosing the harshest available label to be taken seriously produces the opposite of the intended effect.

When to choose other

Other exists for things the list genuinely does not cover, and it is the reason most likely to come with a comment request. Use it when none of the nine fit rather than when several do, and write the sentence carefully, because in that branch the sentence is the whole report.

The temptation is to pick other because it feels safer than being wrong. It is not safer. A vague report about something that had an exact category on the list arrives with less context than the category would have supplied on its own. The category is free information, and declining to give it is declining to help.

Reporting Inside a Group or Channel

Groups have their own machinery, and using it is usually faster than going to the platform.

What an admin can do already

An administrator with the right permission can delete the message, remove the member and ban them, all without involving Telegram at all. That is minutes rather than days, and it is the correct first move in a room you or somebody reachable controls. Going to the platform first, in a room you can moderate yourself, is a slower version of the same outcome. Our guide on Telegram admin rights covers which permission does what.

Admins reporting members

The platform's own spam guidance states that group admins can also report users who post spam in their groups, and that all such reports are checked by human moderators. It is a separate lever from removing somebody, and the two are worth using together.

Removal protects your room and a report protects everybody else's. An admin who only removes is treating a platform level problem as a local one, which works right up until the same account arrives in the next group. Doing both costs one extra tap and is the only version that helps anybody outside your own room.

Reporting the group itself

A group or channel can be reported as a peer, which is the right move when the room exists for the thing you are objecting to. The distinction between a group and a supergroup does not change any of this, though it changes almost everything else about moderation, which our note on the Telegram supergroup covers in detail.

What does change things is whether the room is public. A public room can be reported by anybody who can open its link, while a private one generally needs you inside it. That is one of the few places where joining first is genuinely necessary rather than merely tempting. Weigh it against what joining exposes you to, and leave immediately afterwards if you do.

Reporting a Scammer or an Impersonator

This category has its own dedicated channel, which almost nobody knows about.

The handle for impersonation

The help pages carry a specific instruction. If a scammer is pretending to be you, contact the account named NoToScam. That is a direct route rather than the general queue, and it exists because impersonation reports need a real person to compare two profiles.

Note the framing. It is for somebody pretending to be you, so bring evidence that you are who you say you are rather than evidence that they are lying. The second follows automatically once the first is established, and only you can supply the first. The second follows from the first and the first is the part only you can supply.

What actually helps

Send the username or link of the impersonating account, screenshots showing the claim, and something that establishes your own identity in the same space. A verified channel, a website carrying the same handle, or an older account with history all work.

Skip the narrative. The person reading it is comparing two accounts and needs the two accounts, not the story of how you found out. Anything that is not one of those two things is delay dressed as thoroughness.

Scams that are not impersonation

Most scams are not impersonation, they are ordinary fraud, and those go through the normal report flow with the spam or other reason. Our guide on how Telegram scams work covers the shapes they take and which ones are worth reporting rather than blocking.

Reporting Illegal Content the Serious Way

For a narrow set of categories there are published addresses, and using them is not the same as tapping the button.

The addresses that exist

The help pages list an automated takedown address for abuse, a separate one for copyright complaints, and a dedicated one for child abuse material. Each goes to a different process, and using the wrong one costs time rather than getting forwarded quietly.

There is also guidance for users in the European Union, published for reporting under the Digital Services Act. It matters because a request made under that regime carries obligations for the platform that a tap on a button does not.

What to include

The instruction is explicit and people ignore it constantly. Include links to the content, either the public address or the handle, so the moderators can reach what you are describing. A screenshot without a link is a picture of a problem rather than a route to it. A description without a link is a report about nothing.

What the platform publishes about this

Telegram runs a public safety page with live counters, and the numbers are not small. It reports over twenty two million groups and channels blocked in 2026, and over three hundred and fifty thousand of those specifically for child abuse material. Those counters move while you watch them, which is an unusual thing for a platform to expose.

The same page states that moderation has combined user reports with proactive machine learning since 2015, and that the effort was extended with newer automated tools in early 2024. It also publishes a table of removals credited to outside organisations, which is a fairly unusual thing for a messaging platform to make public. Whatever else that page is, it is evidence that the button feeds a real pipeline.

Reporting an Advertisement

Sponsored messages have their own reporting flow, and one of its outcomes is not available anywhere else.

A separate call, a separate menu

Reporting an advertisement uses its own method with its own option list, structured like the message flow but not the same one. You pass the advertisement's identifier and the option you chose, and the server drives the steps. The identifier is generated for that impression, so reporting works even though the advertisement is not a message in your history.

The outcome nobody expects

One of the documented results means advertisements were hidden. Reporting an advertisement can therefore turn advertisements off for you rather than only registering a complaint, which is a quietly generous piece of design. It is also the only reporting path on the platform that changes something for you immediately.

It is worth knowing before you scroll past something you object to. The alternative most people use is ignoring it, and ignoring it is the one action that produces no result at all, for you or for anybody else.

Why this is not the same as reporting a channel

An advertisement is served into a channel rather than posted by it, so reporting the advertisement and reporting the channel are complaints about two different parties. Getting this backwards punishes a publisher for something the advertising system placed. Publishers do not choose which advertisement lands beside their post, and most of them never see it.

If the objection is the advertisement, use the advertisement flow. If the objection is the channel that carried it, that is a separate report and a separate decision, and making both is entirely reasonable when both are true.

What Happens After You Press It

The honest answer is that you find out nothing, but the process on the other side is documented.

A human sees it

The platform's own spam guidance says reports are forwarded to a team of moderators for review, and that all such reports are checked by human moderators. Automation triages, people decide. That order is worth remembering, because it explains why obviously wrong content sometimes survives a first pass.

That is why volume alone does not decide an outcome. A hundred reports on an account doing nothing wrong is a hundred pieces of the same weak evidence, and a single well chosen report on something clearly against the rules can be enough. Organising a wave of reports is therefore effort spent on the wrong variable.

How long it takes

No timeline is published, and anybody quoting you one is guessing. What is published is the scale of removals, which suggests a pipeline built for volume rather than for individual correspondence. Expect nothing on any particular day and you will not be misled by silence.

You will not be told

There is no notification, no case status and no outcome message. The only way you learn anything is by looking at whether the content or the account is still there, which is unsatisfying and also the entire design.

Treat the report as posted mail rather than as a conversation. Pressing it again the next day adds nothing except another identical item in a queue, and the temptation to do that is the single most common mistake in this whole subject.

What Reporting Does Not Do

Several things people expect from the button are handled somewhere else entirely.

It does not block anybody

Reporting and blocking are separate actions with separate calls. A report can be sent without blocking, and blocking can be done without reporting, and doing only one leaves the other job undone. The report is for everybody else and the block is for you, and most situations want both.

It does not delete the chat

The messages stay where they are on your side and on theirs. If you want them gone from your view, delete the chat yourself, and remember that deleting for both sides is a different option again.

There is a reason to wait before deleting anything. Once the conversation is gone from your device you cannot select messages from it, and the message report is the only path that carries evidence. Report first, tidy up afterwards. Deleting the chat before reporting is the mistake people make while trying to feel better quickly.

It does not remove you from anything

Reporting a group leaves you in the group. If the point was to get away from it, leaving is a separate action, and in a group the departure is visible to the members while in a channel it is not.

The order that avoids regret is report, then leave, then block if the person can still reach you privately. Doing it in the other order is how people end up unable to reach the evidence they meant to submit.

Blocking, Muting and Leaving Instead

For most nuisances, one of these is the correct tool and the report is not.

Blocking

Blocking stops the person reaching you and is instant and unilateral. It says nothing to the platform and nothing to them, which is exactly right when the problem is that you do not want to hear from somebody. Our note on how to tell if somebody blocked you covers what the other side can and cannot see.

Muting

Muting keeps the relationship and removes the noise, which suits a busy group more than a hostile one. It is reversible, invisible and immediate, and it solves the largest category of complaint people bring to the report button by mistake.

The test is simple. If the honest description is that there is too much of it, mute. If the honest description is that it should not exist, report. Applying that test honestly removes most of the traffic people send to moderators by accident.

Tightening who can reach you

Most unwanted first messages arrive because a setting allows them. Restricting who can find you by number, who can add you to groups and who can call you removes the problem at the source rather than one account at a time. Our walkthrough of Telegram privacy settings goes through each of those rows.

This is the only entry on the page that reduces the number of future reports you will need to make. Everything else deals with one account that has already reached you, which is work you can avoid rather than work you can speed up.

Being on the Other End of the Button

Telegram publishes a page written specifically for people whose accounts were limited after being reported, and it is the clearest description of the mechanism anywhere.

What a limit actually is

The page states that when users press the report button, the messages are forwarded to moderators, and if the moderators agree the account becomes limited temporarily. The account is not deleted and it is not silent.

What it loses is the ability to message strangers. The page is explicit that limited accounts can still message people who have their number saved as a contact, and can always reply to anyone who wrote first. So the account keeps working with everybody who already knows you, and stops working with everybody who does not. That single sentence explains most of the confused reports we see, because the account looks half working rather than blocked.

What counts as reportable

The wording is blunt and worth quoting closely. For private messages it does not matter what you send, as long as the receiver finds it unwelcome, and that includes a photo, an invite link or a simple hello.

It goes further and names the categories people mind most, which are unsolicited advertisements, links, invitations to groups or channels, random photos and anything related to commerce. Read as a warning it is stern. Read as a specification it is the most useful paragraph the platform has ever published about outreach. It names the exact contents that produce reports, which is the list any serious sender should be writing against.

Getting out of it

The route back is the spam bot, which reads the account's state and can lift a first offence. Our guide on fixing a spam restriction covers what it answers and what to do when it refuses.

What Operators Get Wrong About Reports

If you run outreach, this button is not a moderation feature you occasionally use. It is the input to the system that decides whether your accounts keep working, and three assumptions about it cost people real money. Teams that need this designed rather than survived usually end up with something built for them under Custom Solutions.

Reports are a rate, not an event

One report does not restrict an account and nobody's outreach survives a high report rate. What matters is the proportion of recipients who press the button, which is a function of targeting rather than of volume.

That is why a smaller, better matched list outperforms a larger one at the same send count. The people who wanted the message do not report it, and our note on avoiding bans in cold outreach covers the rest of that arithmetic. Every improvement in matching lowers the rate twice, once by removing a reporter and once by adding a reader.

The bar is your real audience

Every stranger you write to sees a strip offering to block you, report you, or add you as a contact, next to hints about when your account registered and which country its number came from. A new account with a mismatched number is arguing against itself before the message is read. Nothing in the message can undo what the strip above it already said.

This is the strongest practical case for accounts with history, and it is a case about the receiving screen rather than about limits. Which line sits behind which account matters here too, and holding that record is the job our Account Manager exists for.

Group admins are the other reporter

Most operators plan around recipients and forget that an administrator can report a member for posting in their room. That is a second, entirely separate source of the same restriction, and it is triggered by behaviour that never reaches a private chat.

Anything that posts into rooms you do not control needs the same care as anything that messages people directly. That applies to Mass DMs and to Member Adder in equal measure, and the discipline is the same one, which is to be wanted before being loud. The rooms you post into have moderators, and moderators have the same button your recipients do.

Frequently Asked Questions

Does the person know I reported them?

Nothing in the documented flow tells them, and no message appears in the chat. If their account is later limited they learn about the restriction rather than about who reported it, so the two facts are separate from their side. Nothing appears in the chat and nothing is added to your profile.

How many reports does it take to get someone banned?

There is no published number and anybody quoting one is guessing. Reports are reviewed by human moderators rather than counted, so a single well chosen report on clear evidence can outweigh a pile of vague ones. Organising a group of people to report the same account is effort spent on the wrong variable.

What happens after I report someone on Telegram?

The report goes to a moderation queue where automation triages and people decide. You receive no notification, no case number and no outcome, so the only way to see a result is to check whether the account or the content is still there. Reporting the same thing again the next day adds a duplicate rather than urgency.

Can I report someone without joining their channel?

Yes for a public channel, because opening the link shows a preview carrying the same report entry. A private channel generally needs you inside it, which is one of the few cases where joining first is actually necessary. Leave again as soon as the report is sent if you joined only to make it.

Does reporting also block the person?

No. Reporting and blocking are separate actions and doing one leaves the other undone. If you want them unable to reach you, block them as well, and do it after reporting so the messages are still available to select. Doing it in the other order can leave you unable to attach the evidence you meant to send.

What do I do if someone is impersonating me?

Telegram's help pages point to a dedicated account for this rather than the general queue. Send the impersonating handle, evidence of the claim, and something that establishes your own identity, because a moderator is comparing two profiles. Skip the narrative of how you discovered it, since none of that helps the comparison.

Which reason should I choose?

Pick the specific one that fits, since reports are triaged by category and the wrong queue is slower rather than safer. Reserve the other option for things the list genuinely does not cover, and expect that branch to ask you for a written comment. Picking a harsher label than the situation deserves makes the report weaker rather than stronger.

Why was my own account limited after messaging strangers?

Because recipients pressed the report button and moderators agreed. The platform states that for private messages it does not matter what you sent, only whether the receiver found it unwelcome, and a plain hello qualifies. The same page lists advertisements, links and invitations as the contents people mind most.

Can group admins report members?

Yes. The platform's spam guidance says admins can report users who post spam in their groups, and that those reports are checked by human moderators like any other. It is a separate lever from removing or banning somebody locally. Using both is the only version that protects rooms other than your own.

How do I report illegal content properly?

Use the in-app button for speed and the published takedown addresses for anything serious, including a separate address for copyright and one for child abuse material. Always include links to the content, because a description without a link is unactionable. There is a separate route for users in the European Union reporting under the Digital Services Act.