Content removal: what gets taken down, and what does not

Which legal grounds actually produce removals?

Every platform maintains a list of complaint types it is either legally obligated or operationally committed to review. A request that names one of those types enters a queue with a defined standard. A request that does not enters a general inbox, where it is triaged against nothing in particular. Volume is not a substitute for a ground — automated Instagram reporting tools resubmit the same complaint hundreds of times, and the decision still rests on whether the reported content breaches a named clause. The priced version of that offer is set against a platform's own published outcomes in what a WhatsApp ban service costs and what it reaches. On an end-to-end encrypted service the clause test runs on a narrower evidence set again, because only what the reporter forwards ever reaches a reviewer — the mechanic behind how to get someone’s WhatsApp banned. Where a platform draws its line at public content only, the ground has to be matched to a surface as well as to a clause — the six addresses that split that way on Telegram are mapped in which Telegram takedown route reaches which surface. Facebook splits the same decision by object before it looks at the ground at all, and which of the five Facebook objects a case turns on decides which form can reach it. On Instagram the same sorting runs across five separate forms, and choosing the ground before filing decides more cases than any amount of persistence afterwards. Where that filing is bought rather than done, which of the five Instagram objects the fee reaches is the question that prices the work. X sorts the same decision by who is asking before it looks at what is asked, and which instrument the filer has standing to use settles an X case before reach does. Snapchat adds a constraint none of the others impose: the ground still sets the ceiling, but the content proving it is usually deleted within 24 hours, so what a Snapchat case captures before filing decides more than which form it opens.

The pattern is consistent: the further down that table a URL falls, the more the outcome depends on discretion rather than obligation. That is why cases are triaged per URL rather than per client. A single dispute frequently contains URLs from three different rows, and each needs a different path. The defamation row is the slowest, and the legal route for defamation sets out which letter, notice or court order moves it, and the paper Medium, Nextdoor and review sites act on is mapped host by host. Intimate images, deepfakes and leaked subscription content sit in the second row, and which filing fits which kind of leak turns on who made the file.

What does the evidence package need to contain?

Evidence is captured before any contact is made, because contact is often what triggers the content to be edited, moved, or duplicated. A dispute where the original was never preserved becomes unprovable the moment the other side changes it.

• The full URL, including any query parameters that affect what is displayed.
• A timestamped screenshot showing the content in context, not cropped.
• An independent archive capture, so the record does not depend on your own screenshot.
• The account handle, display name, and profile URL of whoever posted it.
• Any prior correspondence, including refusals — refusals are useful in later escalations.
• For copyright claims, the original file with its creation metadata intact.

How long does a takedown actually take?

Timelines vary by ground and by platform, not by how urgently the request is phrased. Marking a request urgent has no effect on queue position anywhere we have filed.

Anyone quoting a fixed 24-hour removal for arbitrary content is describing an outcome nobody controls. The exception is a narrow one: content you own, hosted on a platform where you hold the account, can genuinely be deleted in minutes — but that is self-service, not a service worth paying for.

When is removal the wrong tool?

Three situations come up repeatedly where filing removal requests wastes months and makes the situation worse.

Accurate reporting by an established publisher

News organisations do not unpublish on request, and search engines will not deindex lawful journalism. Persistent demands sometimes produce a follow-up article about the demand itself, which is measurably worse than the original. These cases go to suppression from the start.

Court records and public filings

Public records are usually republished by dozens of aggregators within days of publication. Removing one aggregator does nothing while the source record remains public and the others keep syndicating. Sealing or expungement at the source, where legally available, is the only intervention that changes the underlying supply.

Content where the requester has no standing

A common request is to remove content about a family member, a business partner, or an employee. Most removal channels require the affected person to file, or a documented legal authority to act on their behalf. Filing without standing gets the request rejected and can flag the account for abuse review. Meta grades that requirement by tier, and the tier that opens on self-reporting alone is mapped in which Facebook complaints only the target can file. The paid market that sells around that limit is priced and tested in our guide to what a twitter ban service can and cannot prove. LinkedIn draws the line tighter still on one route: only a Page super admin may report a false employer listing, which is why which LinkedIn route your evidence supports is settled before the form is opened. Instagram enforces the same rule through the document it asks for: its impersonation route is decided on a photo of the reporter holding government ID, so the report an impersonation case is actually decided on is not the one on the profile menu. X splits the same question route by route rather than by platform: three of its seven removal forms accept an authorised representative, one accepts nobody but the person depicted, and which X removal routes can be filed on your behalf is the first thing worth settling before paying anyone to file them. TikTok charges the same question to the buyer rather than the form, so what to check before paying a TikTok removal desk starts with who will sign the claim.

What monitoring after removal is for

Removed content reappears more often than most clients expect. Scrapers that copied the page before removal keep serving their copies. Archive services retain snapshots. Hostile posters re-upload deliberately once they notice a takedown landed.

Monitoring for 90 days after a successful removal catches the reappearance while the original evidence package is still fresh, which makes the second filing far faster than the first. Skipping this step is the most common reason a case that looked resolved is back within a quarter.

Related reading

• What an X removal service can be hired to file — the seven routes sorted by who X accepts the filing from, and the copyright takedown that can reverse itself.
• Which Facebook object a takedown quote is actually aimed at — profile, Page, group, Business Manager and ad account are removed and restored five different ways.
• The five published routes that end a YouTube channel — the only platform that prints its own thresholds, set against what a ban listing promises.
• The five LinkedIn removal routes and who may file each — sorted by what the filer can prove they own, including the one route with a published fourteen-day clock.
• When a YouTube complaint reaches the channel and not the video — the four routes that terminate a whole channel, and why three of them never touch the strike counter.
• Choosing the right route to remove a single YouTube video — which of the five forms a reviewer can actually grant, and the two lanes that now run against a legal deadline.
• The five TikTok account takedown routes, priced and ranked — and why the fastest one carries a statutory penalty for filing it dishonestly.
• What decides how fast a TikTok comes down — the three clocks behind the five routes, and the escalation ladder that survives a refusal.
• Which TikTok report reason fits the problem — the report path for an account, video, LIVE or comment, who learns your name, and where scam and rights claims go instead.
• What actually gets a TikTok account banned — strikes counted per item and per feature, the violations that end an account on the first strike, and the block and LIVE controls that need no review.
• Reporting a stolen TikTok video under copyright — which notice fits a copied video, logo or Shop listing, what the uploader sees about the reporter, and the counter-notice clock.
• Reporting a TikTok Shop seller, a bot ring or a duet — the route that reaches each target, the separate record a shop is scored on, and how to tell a ban from a deleted account.
• Reporting a fake or impersonating TikTok account — the in-app and browser report paths, the US form's ID photo and ten-link limit, the display-name test for fan accounts, and the appeal when a clone gets the original banned.
• Reporting a TikTok scammer, Shop seller or LIVE gifting scam — the report that fits each place a scam runs, what to save before a ban erases it, why Shop refunds stop at checkout and LIVE Gifts never come back, and the trademark and copyright forms for brand abuse.
• Who can delete a TikTok account, story or repost — the owner-only buttons for your own story, post and repost, the report routes for someone else’s account or video, and the old-account and deceased-user requests.
• Clearing out an old TikTok account you no longer use — deletion with or without the login, why inactivity only resets the username, and the duets, stitches and re-uploads that outlive the account.
• Filing a Twitter DMCA notice that X acts on — the six fields the copyright form checks, why about four notices in ten stall, the counter-notice window and the trademark route for names and logos.
• Reporting a fake X account that also doxxes or threatens — which of the three reports to file first, what X’s private-information rule leaves out, and why report counts do not decide suspensions.
• Reporting a Twitter clone that impersonates or scams — the impersonation form and its three filer choices, the four conditions a parody account must meet, trademark and copyright routes for brand copies, scam and mistaken reports, and the repair once the copy is suspended.
• Choosing a service to remove a fake website or profile — who publishes each takedown ranking, how to read a vendor’s speed claim, and the free desks that hold the switch for fake sites, profiles, apps and listings.
• Getting a phishing site or scam account switched off — which of the registrar, host, proxy, blocklists and platform holds each switch, the evidence each acts on, how to read a takedown vendor’s speed claims, and the report routes for eBay, OfferUp and Discord.
• Who may file the YouTube copyright takedown form, and every other notice — sixteen platforms sorted by the sender each route accepts, with Apple’s App Store dispute, dating-app impersonation reports and the limits of each.
• Taking down a YouTube channel, and the layers below it — the four routes that end a channel, what reaches private and unlisted videos, and who switches off a marketplace listing, a copied file, a proxied site or a domain.
• Ten copyright desks compared, starting with ComplaintsBoard — where a DMCA notice goes on YouTube, Dailymotion, Course Hero, DeviantArt, itch.io, eBay and the Wayback Machine, and what each desk does to the uploader.
• What a GitHub DMCA takedown service does, and how long a YouTube strike lasts — GitHub’s one-business-day window and fork rule, the 90-day strike and the shorter clocks inside it, and what eleven copyright desks publish about the person who files.
• Reporting Redbubble copyright infringement, and what thirteen desks make you prove — the proof behind a copyright, trademark or patent report on Redbubble, eBay, Amazon, LinkedIn, the Chrome Web Store and VK, and why only the listings you name come down.
• An Imgur copyright takedown request, and ten other desks by what was stolen — where a notice goes for a copied image, paid post, article, product design, game, stream or paste on Imgur, Pinterest, Patreon, Medium, LinkedIn, Redbubble, itch.io, Kick and Pastebin, and what each removal leaves online.
• Where a scam or impersonation report goes on 18 services — the report path on Fiverr, Grindr, marketplaces, payment apps, Binance, Roblox, Steam, Signal, X, YouTube and LinkedIn, and which desk on each can return money.
• Which account ID to quote in a scam or impersonation report — the identifier that survives a rename on Vinted, VK, Threads, Tinder, X, Bluesky, YouTube, Steam, Upwork and Shopify, and the report path on each, including Zelle and Australian scam-site takedowns.
• How a Vimeo copyright strike lands, and how it comes off — Vimeo’s three-strike rule with no expiry date, the counter-notification window, and how YouTube, X, Wattpad, VK and the Wayback Machine treat the account behind a removed copy.
• Reporting a Twitch stream or a YouTube channel for copyright — what one notice can name on Twitch, YouTube, Rumble, BitChute, Tumblr, Scribd, Substack, Patreon, Spotify and Shopify, and when strikes close the account.
• A Grindr catfish report, and seventeen other desks by what the fake took — where a fake profile, a scam payment, a scam ad or app, a marketplace fraud and a copied repository are each reported, and the deadline each desk sets.
• Impersonation takedowns, sorted by what each report does — which desks remove a fake profile, Page or copycat website, which only record the complaint, and who is allowed to file on Instagram, Threads, LinkedIn, Pinterest, dating apps and marketplaces.
• Which reports start on a computer, and which need the phone — reporting a YouTube ad on PC, an app on the App Store or Google Play, an eBay listing or counterfeit, art theft, game players and dating-app matches, with the email fallback for apps that keep Report on the phone.
• Who holds the delete button for a wiki page, a domain or a video — Fandom wiki and Wikipedia takedowns, fast domain takedowns and FBI seizures, YouTube videos, channels and comments, GitHub repos, eBay VeRO, Depop, craigslist and Fiverr listings, Discord servers, Facebook groups and reposted art.
• Having a YouTube video taken down, deleting your own, or finding one that went — the report, copyright and privacy routes for someone else's upload, Delete forever and channel hiding for your own, the warning-and-strike ladder, and what the Wayback Machine keeps of a removed video.
• What a YouTube report does after you send it — the four outcomes of a video or comment report, which YouTube routes tell the uploader who filed, reporting without an account or for privacy, and single-post reports on Twitch, Kick, Rumble, BitChute, Vimeo, Nextdoor, Wattpad and Upwork.
• Internet Archive and itch.io takedown requests, and what a granted one leaves behind — where a takedown goes on the Internet Archive, itch.io, Patreon, Substack, Pastebin, Mastodon, Discord and a registrar, and which copies, records and accounts each removal leaves in place.
• What each report desk sends back after you file — reporting a YouTube advert or an Airbnb listing, counterfeits on Amazon, eBay, Poshmark, Temu, Vinted and Apple, DeviantArt and Redbubble art theft, Discord raids, Signal, Kik and Mastodon, with the identifier each desk needs.
• Getting off ZoomInfo, and proof that a report worked — the ZoomInfo opt-out and California's DROP route, Steam trade bans, and what eBay, Fortnite, Discord, Mastodon and gossip-site desks send back after a report.
• What a scam report can do on OfferUp and Upwork — the report path on each from their own help pages, why OfferUp sales are final since shipping ended in 2025, which Upwork flag reason to pick, and the routes for Cash App, crypto, dating apps, Bluesky, Lemon8, Discord, Kik and Chrome extensions.
• Reporting a scam at every app it crossed — the report for each hop of a marketplace or romance scam, from OfferUp, a dating app, Vinted or Upwork to Signal, Discord, Zelle, Venmo and USDT, and why the payment report goes first.
• What one DMCA notice reaches on a website, on YouTube and on twelve other desks — how far each removal extends, from a single file to a fork network, and which desks forward or publish what you wrote.
• Reporting Telegram copyright infringement through [email protected] — which channels, bots and sticker sets a notice reaches, what to attach, and why private groups and mirror channels need other routes.
• The five WhatsApp number takedown routes and their ceilings — including the intellectual property claim most brand owners miss.
• Reporting a WhatsApp number that poses as a business — what a report sends, how Business app, Platform and verified numbers differ, and the scam and impersonation routes that reach Meta.
• The six Telegram takedown routes, and the boundary all six stop at — including the free EU notice route no ban seller mentions.
• How to report someone on Telegram, starting from the message — where Report sits on Android, iPhone and desktop, which official bot takes which report, and what happens to a reported account or a banned number.
• What a Telegram report bot can file, and what it only sells — the moderation bots, Telegram’s own intake bots and the paid storefronts, and why a Stars payment is the only one you can dispute.
• Which Reddit route removes which post — copyright notices, doxxing and impersonation reports, legal requests for defamation, and the copies that survive in Google, archives and mirrors
• Whether a reported Telegram user learns who reported them — what the reported side can see for each report route, why no report count bans a channel, and the notices that do carry a name
• How to report a Telegram account, group or channel — the Report path for each object, the email and police routes, and how to tell a blocked channel from one restricted on a single platform.
• What to do when someone is impersonating you on Instagram — the identity-verified queue behind the impersonation form, and why the in-app tap files a weaker case.
• Getting a picture of you taken down on Instagram — the privacy form for a photo you appear in, the copyright form for one you took, and why your own old account has to be recovered before it can be deleted.
• What one Instagram report actually reaches — the Report option for a comment, post, photo, Reel or story, how Meta triages it, and why a comment that vanishes from your screen has not been removed.
• What an Instagram report sets in motion, on both sides — who is told what, where the decision shows up, and the two forms that pass your name to the person you reported.
• The six outcomes of an Instagram report — why a report can disable an account but never delete it, and which reason to choose for a bot, bought followers or a deceased person.
• Reporting a whole Instagram account, profile or page — the three account reasons, what an account review weighs against the strike record, why a group chat cannot be reported whole, and the routes left to an owner who was banned.
• What a single report on X can remove — the four outcomes of a review, including the deletion X makes the author perform, and the form that fits bots, ban evasion, counterfeits and already-deleted posts.
• Taking down a Facebook Page, post or profile picture — the admin’s deactivate-or-delete choice and its 14-day window, and the strike route that brings down a Page someone else runs.
• Filing an Instagram copyright report, and recognising a fake one — the six elements a claim needs, the name it passes to the person reported, and the WhatsApp and Telegram copyright scams that sell a withdrawal.
• Where a DMCA takedown notice goes for stolen work — the platform, host, Cloudflare and search routes for copied video, music, art, apps and listings, and who learns the filer’s name.
• Reporting a Discord user or server, and the same question elsewhere — why Discord judges messages and profiles rather than servers, what its Warning System does with a report, and the routes on Kik, Hinge, LinkedIn, eBay, Etsy and 4chan.
• Which Instagram account takedown route fits which harm — the four forms, their evidence standards, and what the paid ban trade sells instead.
• Proving a rival wrote a fake review — the conflict-of-interest rule on G2, Capterra, Trustpilot, Tripadvisor and twelve other desks, and the evidence that ties a reviewer to a competitor.
• What each Google removal reaches, surface by surface — reviews, fake listings, defamation, autocomplete, knowledge panels and doxxing, and what Bing and the archives keep afterwards.
• Which report route gets harassment and threats acted on — the category that sets the review queue, the evidence a reviewer needs, and the abuse takedown past the platform to hosts, Cloudflare and the Take It Down Act.
• Which Instagram form fits a fake profile, scam or brand abuse — impersonation versus the copyright and trademark forms, who has standing to file for a catfish, and the harassment form that no longer loads.
• Which X report route can end in a suspension — the account report versus the post report, the impersonation and trademark forms, what X’s notices do and do not reveal, and how the owner closes a suspended account.
• How to remove personal information from Google search results — the self-service paths worth trying before hiring anyone.
• Removing a dox from 4chan and the brokers it was copied from — what each forum and platform will take down, the BeenVerified, Spokeo, Radaris and Whitepages opt-outs, and what paid removal services leave untouched.
• How long content removal takes, by request type — the timeline data behind the ranges above.
• Which Facebook report gets harassment, threats and abuse removed — the category and policy tier each kind of abuse falls under, the rules that only count when the target files, Messenger and business Page reports, and the 90-day records window police rely on.
• Which Meta route fits a fake Facebook account, a scam or stolen work — the impostor form and its ID upload, scam Page, Marketplace and ad reports, the single copyright and trademark form, Rights Manager, and the defamation form’s ten-URL limit.
• How to report a Facebook account, Page or group, and what each report reaches — the profile, Page and group report buttons, the admin-only button that never reaches Facebook, groups paused after seven days without an admin, and the strike ladder behind every ban.
• What a Facebook account removal service can file, object by object — profile, Page, group, ad and Marketplace desks, who may file each, and how mass reporting of Facebook groups has played out since 2019.
• Which Facebook report reaches a fake that scams in your name — the impersonation report and impostor form, Marketplace scam reports for buyers and sellers, scam ads and the Page behind them, scam groups, and the clean-up once the copy is removed.
• Getting a fake Instagram account removed — which report fits a clone, a bot or a fake giveaway, the five reasons Instagram refuses a report, and the escalation order that follows a refusal.
• Which Instagram takedown request fits the harm — why no one can delete another person's account, the impersonation, copyright, trademark and privacy forms side by side, who learns your name, and the clocks set by Meta and by law.
• Reporting Instagram harassment so the right reviewer sees it — why DMs are reported from the chat rather than the profile, the 30-message review limit, threats and blackmail, the California form, and when the police come first.
• Reporting one Instagram post, comment or story — the button and the reason for each item, the 30-message chat limit, the Report a problem dead end, report history and reviews, and which form fits when someone is using your photos.
• Removing defamation by letter, notice and court order — the defamation test and time limits, what a cease and desist letter can and cannot do, when a copyright notice is faster, and how Section 230, the UK and the EU change what a host must do.
• What each review site will remove, and what a removal service may file — the removal grounds on Trustpilot, Amazon, Airbnb, Capterra, DealerRater and the BBB, what Ripoff Report arbitration buys, and the tactics the FTC’s review rule penalizes.
• Proving a review is fake, desk by desk — fake, false and defamatory reviews sorted by route, the evidence Tripadvisor, Glassdoor, Avvo and Angi accept, and when only a court order removes a review.
• Which Facebook protection tier covers your complaint — Meta’s three-tier ladder, the public-figure threshold, and the appeal reporters are not told about.
• Why Facebook counts strikes, not reports — the published restriction ladder, and where report volume sits in it.
• How many reports an Instagram ban takes — none on their own: the strike rules Meta publishes, why the Facebook penalty ladder does not run on Instagram, and why new accounts follow a disabled one down.
• Getting a player banned on Roblox, Discord or Steam — which bans a server, channel or game owner can issue directly, what each platform publishes about report volume, and the five things that actually trigger a ban.
• What report volume does on Instagram — Instagram's own statement on the question, and how to recover if you were the target.
• How Telegram judges a reported message — what the panels sell, set against what moderators actually weigh.
• What a paid or free report bot actually costs — donated logins, cookie-stealing tools and copied vendor pages, what game studios say about report floods, and when a parody account or gripe-site post can really be removed.
• Why a Snapchat takedown turns on the delete clock — Snap’s own retention table, and the one preservation route that is not the report button.
• Why duplicate Snapchat reports get deprioritised — the public web form behind every Snap reporting bot, and how fast Snap already moves.
• Which Snapchat report gets an account disabled — impersonation, scam, fake-profile, bullying and group-chat reports matched to the rule Snap enforces, and what paid ban services cannot do.
• Where a Snapchat report goes, and what a ban looks like — the report path from a chat, Search, a Snap or the web, what Snap’s review can do, and why nobody outside the account can tell a ban from a deactivation.
• The TikTok removals nobody reported — how much enforcement happens before a report exists, and what that leaves for one.
• What a TikTok report bot actually sends — replay scripts and fake-account fleets, what the GitHub repositories carry, the seller red flags, and the appeal path when an account is mass reported.
• Who may file each Twitter takedown request — the nine X intakes sorted by signer, the documents each asks for, and what the reported account is told about the filer.
• What happens after a YouTube DMCA notice is filed — the review and follow-up questions, what the uploader learns about the claimant, and the 10-business-day counter-notice clock.
• When an X parody account or a YouTube channel comes down — the four parody labels X checks, the one-year channel appeal, and which policy label each harassment desk reads.
• Reporting Twitch harassment by where it happened — stream, chat, whispers or another site, the evidence each report needs, and why report volume does not ban anyone.
• The five instruments that reach an X account — who has standing to file each one, and which two put X on a deadline set in law.
• Why X discounts report volume — what bulk reporting does to the buyer's own account, and what removes a post instead.
• What bulk WhatsApp reports return — the grievance figures WhatsApp publishes itself, and the routes that move a case.
• What user flags actually remove — why exceptionally high flagging volumes are set aside, and what is left holding a channel to account.
• When to report a whole account instead of a post — the account-level reasons, where YouTube, Twitch, Discord, Grindr, Bluesky, Kik, LinkedIn, Steam and DeviantArt hide the profile report, and the evidence a reviewer can act on.
• Where a Discord ban bot’s reach ends — server-only ban permissions, why a paid account ban service cannot reach Trust & Safety, and which desk removes domains, listings and wiki pages.
• Why a LinkedIn ban bot bans no one — the five products sold as mass reporting, what each one costs the buyer, and the single reports that remove a profile, post or video.
• When a fake Instagram account copies you — the report that fits what the copy took, the escalation after a refusal, and the repair work that removal leaves undone.
• Fake sellers, copycat apps and cloned profiles — how eBay, Poshmark, Depop, Google Play, the App Store and the Chrome Web Store remove copies, and the proof of the original each desk asks for.
• What a takedown service files, and what a ban panel sells — the three products sold as takedown, what each platform says about report volume, and the seller questions that expose a scam.
• Reputation management — what to do when removal is refused.
• Account recovery — for content posted from an account you have lost control of.

Questions we are asked before every engagement