International SMM

X (Twitter)

Tools for the platformActivepiecesAgorapulseAlbatoApiX-DriveAwarioBodyguardBrand24BrandBastionfull list ↓
RegionUSA and Canada

Platform · this deep dive is built from the official pages of the platform itself and of the tools that support it.

Overview

X is a service for short public posts that grew out of Twitter. Posts are visible without signing in, the feed is built around real time, and the platform's main income is advertising: in the terms of service the exchange is stated outright — "In exchange for accessing the Services, X and our third-party providers and partners may display advertising to you".

For an SMM specialist X is arranged differently from most platforms: there is no "business account" here as a separate entity — the ad manager and the API attach to an ordinary profile. In return the platform has a well-developed programming interface with pay-as-you-go billing and carefully documented limits.

An important feature of working with X in 2026: the help centre help.x.com is closed to automated requests — the server answers with a 403, and a browser is shown a "not a bot" check.

Who's here

X has no public pages with a user count. On the advertising page the platform describes its audience in words — "reach high-intent audiences", "connect with high-value audiences" — but there is not a single reach figure there.

The only numbers about scale that X publishes itself and without a login are in the global transparency report. The platform reports 181,701,993 user reports (excluding platform manipulation and spam), 4,191,171 suspended accounts and 10,131,247 posts removed or labelled; the share of violating posts, by the platform's own calculation, is 0.0178%. One caveat must be named: the report's heading assigns the data to the period "July to December 2024", while the caption on the reports table says "January to June 2024". The discrepancy is in the source itself.

The interface language of the home page x.com is English (verified 2026-07-28). Separately, X reports to the EU regulator: the DSA Transparency Report page states that the reporting is done by the Irish legal entity X Internet Unlimited Company (XIUC), and the latest report for the second half of 2025 is posted as a ZIP archive.

Getting started

The minimum age is set in the terms of service: "you must be at least 13 years old to use the Services". In the X app listing in the App Store (seller — X Corp.) the store rating is given as 16+ — that is a store restriction, not a platform one.

X has no separate procedure for "converting an account to a business one". The practical route for an organisation is: an ordinary profile → the ad manager → programmatic access.

Campaigns, according to the "How X Ads work" page, are run by "eligible accounts" on ads.X.com; the eligibility criteria are not disclosed there. For the API, according to the "Getting access" documentation, you need to create a developer account on console.x.com, accept the Developer Agreement and Policy, describe your use case and create an app. Keys are shown once — the documentation warns: "Credentials are only displayed once".

For advertisers in the US the platform promises on the advertising page "$500 in ad credit" on $500 of spend, with the caveat "New advertisers in the US. Terms apply.".

What you can publish

On the page for the create post method the text field is described without a maximum — the platform does not name one there.

The limits for attachments are documented in detail; below are the figures from the "Media best practices" page, which relate to uploading through the API:

  • How many attachments. "You may attach up to 4 photos, 1 animated GIF or 1 video in a Post".
  • Images. Up to 5 MB, formats JPG, PNG, GIF, WEBP. The same 5 MB limit is given in the media introduction.
  • Animated GIFs. Up to 15 MB, resolution up to 1280×1080, up to 350 frames, up to 300 million pixels.
  • Video. Up to 512 MB, duration from 0.5 to 140 seconds, resolution from 32×32 to 1280×1024, frame rate no higher than 60.

Polls are set in the same create-post method: from 2 to 4 options, each from 1 to 25 characters, with a voting period from 5 minutes to 10,080 minutes (exactly seven days). Up to 10 users can be tagged in media.

The publishing schema on the same page lists the service flags: paid_partnership (marking sponsored material), made_with_ai (marking generated content) and for_super_followers_only (a post for paying subscribers only).

What may not be published X describes in its rules on help.x.com — the page is closed off by a bot check. Among the open documents, the prohibitions are visible only in the terms of service and in the developer policy.

How to grow

What the platform says about the algorithm. On ad selection X writes on the "How X Ads work" page: delivery is tuned by your actions — "follow, post, search, view, or interact" — as well as by profile data, device location, IP address and installed apps. The same page lists the ways to opt out of personalisation, including the "Personalize ads" setting and the optout.aboutads.info tool.

Ad formats. On the advertising page X lists six: Amplify (ads alongside video), image and carousel ads (up to six cards), Collection and product ads, promotion of existing posts, dynamic product ads and "Boost existing posts" — the last marked "Additional features for Premium Business". The advertising help names Promoted Ads and Trend Takeover; all of them are "clearly marked with an “Ad” label".

The barrier to entry. To the question "Is there a minimum budget to advertise?" the platform answers "No" and adds: "you're only charged when ads run". Targeting on the same page is described as "keywords, interests, followers, location". The same page carries the claim "10x ROAS average, never below 4x weekly" — with no methodology, period or sample: that is the platform's claim about itself, not a measurement.

Built-in statistics. The X API introduction lists the engagement metrics available: impressions, likes, reposts, replies and video views.

Path to monetization

This is the most closed spot in X's open sources: the terms of author payouts live behind a bot check. The pages help.x.com/en/using-x/creator-subscriptions and help.x.com/en/using-x/x-premium answer an ordinary request with a 403 and show a bot check in a browser.

What is nevertheless visible from open sources:

  • Paid subscriptions to an author exist. The create post schema has a for_super_followers_only field and a subscribers value in the reply setting — signs of a paid subscription to a specific author.
  • Subscriptions are sold through the app. In the X app listing in the App Store the in-app purchases include subscriptions to individual authors — "@elonmusk Subscription — $4.00" and "@FatKidDeals Subscription — $1.00" — as well as "Promote Post for $25" and "Promote Post for $50".
  • X Premium prices — according to the app store. From the same source: X Premium Basic — $4.00 a month and $42.00 a year; X Premium — $11.00 a month and $114.99 a year; X Premium Plus — $50.00 a month and $490.00 a year. These are iOS in-app purchase prices, which are usually higher than the site's prices because of the store's commission.
  • The terms of service say nothing about payouts: the words "monetization", "payout", "Premium" and "Subscriptions" do not appear in the text of the terms.

Tools and automation

The API access model. According to the introduction X works on a "pay-per-usage pricing. No subscriptions—pay only for what you use" basis: credits are bought in advance and drawn down as requests are made, with consumption visible in the developer console.

Prices. The X API pricing page gives the rates per unit of data: reading a post — $0.005, reading a user — $0.010, publishing a post — $0.015, publishing a post with a link — $0.200. Reading your own data — $0.001 per unit. Resources are deduplicated within a UTC day. The pay-as-you-go plan has a ceiling: 2 million posts read per billing month.

Request limits. The "Rate limits" section explains the model: 15-minute windows, separate limits per user (OAuth 1.0a or OAuth 2.0) and per app (Bearer Token), the remainder visible in the x-rate-limit-remaining headers, and an overrun producing a 429 with error code 88. The figures for specific endpoints are gathered on a separate page: publishing a post — 100 per 15 minutes on behalf of a user and 10,000 per day on behalf of an app; deleting a post — 50 per 15 minutes; likes — 50 per 15 minutes and 1000 per day; sending direct messages — 1440 per day; uploading media — 500 per 15 minutes on behalf of a user and 50,000 per day on behalf of an app; searching recent posts — 450 per 15 minutes. For the enterprise tier the limits are agreed separately.

What is available through the API. According to the introduction: posts (search, reading, publishing, threads), users, Spaces, direct messages, lists and trends by location; a filtered stream (up to 1000 rules) and full-archive search back to 2006 are named separately.

What may not be automated. The developer policy prohibits bulk follows, publishing identical content from several accounts, unsolicited automated messages without the recipient's consent, and paying users for actions on the platform. Stored copies of data must be kept in sync: deleted or changed content is to be removed within 24 hours. Sharing post identifiers is capped at 1.5 million per 30 days per person. The developer agreement additionally prohibits training foundation models on X data, embedding content through an iframe and reselling access "on a service bureau, rental or managed services basis".

The partner programme. The X Official Partner directory is described as an "invitation-only program": partners are selected after "an extensive evaluation" and undergo continuous re-evaluation. There are no names on the page itself — it refers to an external directory.

Limits and rules

Age. No younger than 13 under the terms of service.

Data collection. From the same document, plainly: "crawling or scraping the Services in any form, for any purpose without our prior written consent is expressly prohibited". Circumventing protections is banned too, including "jailbreaking" and "prompt engineering or injection". Access is permitted only through the published interfaces — "You may not access the Services in any way other than through the currently available, published interfaces that we provide". In practice this means a scheduling service must work through the X API and pay for its requests.

Moderation. The figures are given above, in the audience section. The X Rules themselves live on help.x.com — behind the bot check. Reporting under the Digital Services Act is done by the Irish entity XIUC, with the reports posted on the DSA page.

Limits by time window, not by day. Schedulers used to daily quotas will meet a different model: X's basic beat is a fifteen-minute window, and daily ceilings are set only for some endpoints.

Who it's for

X is useful if you need a public feed that works without signing in, fast reaction to events and automation that is predictable in cost: the API rates and request limits are published down to the last cent, which is rare. The platform suits technology products, media outlets, public experts and customer support in an open channel.

X is a poor fit if you are planning to earn from the platform itself: the terms of author payouts sit behind a bot check, and the economics cannot be worked out in advance from open sources. Nor will it suit anyone who counted on free programmatic access: integration here is paid from the very first request.

Tools for this platform

The list is built from the tools' own facts: every line is a tool's statement that it supports the platform, with the source named. A tool missing from the list only means we have no such statement.

Scheduling and publishing:

Statistics and analytics:

Customer communication:

Automation:

Social listening:

Other tasks:

Verified data

The checked data this deep dive rests on.

Account and access

the app's age rating in the App Store is 16+

A store restriction, not a platform condition (the terms say 13)

source, checked 2026-07-28

What you can publish

«You may attach up to 4 photos, 1 animated GIF or 1 video in a Post»

source, checked 2026-07-28

images up to 5 MB, formats JPG, PNG, GIF, WEBP

The same 5 MB limit is confirmed on the media introduction page

source, checked 2026-07-28

animated GIFs up to 15 MB, resolution up to 1280x1080, up to 350 frames, up to 300 million pixels

source, checked 2026-07-28

video up to 512 MB, duration from 0.5 to 140 seconds, resolution from 32x32 to 1280x1024, frame rate no higher than 60

NEWER THAN THE PREVIOUS RECORD: the database held only the chunked upload requirements for video, with no numeric limits

source, checked 2026-07-28

the maximum video file size is 512 MB with media_category=amplify_video; an image 5 MB, a GIF 15 MB

source, checked 2026-07-28

a poll: from 2 to 4 options, each from 1 to 25 characters, voting period from 5 to 10,080 minutes; up to 10 users can be tagged in media

source, checked 2026-07-28

the publishing schema has the service flags paid_partnership (sponsored material), made_with_ai (AI-generated), and for_super_followers_only

source, checked 2026-07-28

Images. "Supported image media types: JPG, PNG, GIF, WEBP", "Image size: <= 5 MB". For a GIF animation the limits are separate: "<= 15 MB", "Resolution: <= 1280x1080 (width x height)", "Number of frames: <= 350", "Number of pixels: <= 300 million (width * height * num_frames)"

the limits are given for uploading through the API; a GIF counts as a separate kind of attachment and not as an image

source, checked 2026-08-04

Video. The obligatory requirements: "Frame rate: must be 60 FPS or less", "Dimensions: must be between 32x32 and 1280x1024", "File size: must not exceed 512 mb", an aspect ratio "between 1:3 and 3:1", "Pixel aspect ratio: must have 1:1". The recommended ones: "Video Codec: H264 High Profile", 30 or 60 frames per second, the resolutions 1280x720 (horizontal), 720x1280 (vertical), 720x720 (square), "Minimum Video Bitrate: 5,000 kbps", audio "AAC LC", "Minimum Audio Bitrate: 128 kbps"

some of the figures are obligatory ("must"), some are recommended — in the source these are different lists

source, checked 2026-08-04

Per post: "You may attach up to 4 photos, 1 animated GIF or 1 video in a Post". The maximum size by kind of media: an image 5 MB, a GIF 15 MB, a video 512 MB (with media_category=amplify_video). A video is uploaded in parts (chunked upload)

the number of attachments and the sizes — https://docs.x.com/x-api/media/quickstart/best-practices ; the procedure for uploading in parts — https://docs.x.com/x-api/media/quickstart/media-upload-chunked

source, checked 2026-08-04

video is uploaded in parts (chunked upload) with the total size and the media category specified

The quick-start page gives no numeric limits

source, checked 2026-07-28

Text length. The limit on the length of a post: "Posts on X can contain up to 280 characters". It is not bytes that are counted: the Latin alphabet, punctuation marks and ordinary characters weigh 1, emoji and CJK characters (Chinese, Japanese, Korean) — 2. Links are counted at a fixed length: "All URLs are wrapped with t.co shortener and count as 23 characters, regardless of the original length". Media attached through the official clients do not consume characters

subscribers on the paid plans do have longer posts, but this could not be verified against the official help centre: https://help.x.com/en/using-x/how-to-tweet-longer-posts returns 403 even though robots.txt does not forbid it

source, checked 2026-08-04

"Duration: must be between 0.5 seconds and 140 seconds" — that is, from half a second to 140 seconds

this is the limit for uploading through the API; announcements about longer clips for the paid subscriptions could not be found on the accessible official pages, the help centre help.x.com returns 403

source, checked 2026-08-04

Content

No general requirement to label all AI content was found. There is a requirement for one case — armed conflicts: "For any AI generated content depicting armed conflict, you must add a clear disclosure that it is AI generated"

source, checked 2026-08-04

No separate labelling tool for the creator at publication is described: the rule requires that one "add a clear disclosure that it is AI generated", that is, disclosure inside the post itself. The label is applied by the platform: "For posts containing media determined to have been significantly and deceptively altered or fabricated, X may add a “Manipulated media” label"

source, checked 2026-08-04

Partly: "To determine if media have been significantly and deceptively altered or fabricated, X may use its own technology or receive reports through partnerships with third parties". No mentions of C2PA or IPTC were found

source, checked 2026-08-04

"Media that is shared as authentic but is significantly and deceptively altered, manipulated, or fabricated in a way that fundamentally changes its meaning is prohibited, including media depicting a real person that has been fabricated or simulated, especially through use of algorithms or broader artificial intelligence". Not punished: "X will not take action to label or remove media that have been edited in ways that do not fundamentally alter their meaning, such as retouched photos or color-corrected videos"

source, checked 2026-08-04

For a missing label on an AI video about an armed conflict: "Effective March 3, 2026 users who post AI-generated videos of an armed conflict without adding a disclosure that it was made with AI will be suspended from Creator Revenue Sharing for 90 days. Subsequent violations will result in a permanent suspension from the program". For misleading altered content the platform "may add a “Manipulated media” label", and also labels or takes down the post

source, checked 2026-08-04

"Monetized content, like other content on X, must be in compliance with X’s rules on Safety, Authenticity, and Privacy. You must only monetize content that is original and authentic, or otherwise content that you have the rights to monetize". Plus mandatory disclosure for AI content about armed conflicts (see ai_disclosure_required)

source, checked 2026-08-04

Promotion

Judging by the addresses in search results, the platform has a whole help centre section, "Recommender systems" (help.x.com/en/resources/recommender-systems/: for-you-home-timeline-recommendations, search-recommendations, account-recommendations, trends-recommendations, conversations-recommendations), and a rules page, "About our approach to recommendations" (https://help.x.com/en/rules-and-policies/recommendations), and in the blog a post, "Twitter's Recommendation Algorithm" (https://blog.x.com/engineering/en_us/topics/open-source/2023/twitter-recommendation-algorithm). It was not possible to read a single one of them: they all return 403 or 402

the help centre and the blog of X are closed by protection from robots, and it must not be circumvented: help.x.com and blog.x.com return 403 with the splash page "Just a moment... Enable JavaScript and cookies to continue", x.com, business.x.com and transparency.x.com return 402, "Please contact the site owner for access.". On top of that the robots.txt of x.com itself contains the group "User-agent: * Disallow: /", that is, it closes the site to all robots. The existence of the pages is confirmed only by the addresses in search results, the content has not been checked and therefore has not been recorded

source, checked 2026-08-05

Advertising

promoting a post is sold as an in-app purchase: «Promote Post for $25» and «Promote Post for $50»

source, checked 2026-07-28

to the question «Is there a minimum budget to advertise?» the platform answers «No»; charges are made only when the ads are running

source, checked 2026-07-28

new advertisers in the US are promised «$500 in ad credit» on $500 of spend, with the caveat «New advertisers in the US. Terms apply.»

source, checked 2026-07-28

six formats: Amplify, image and carousel ads (up to six cards), Collection and product ads, promotion of existing posts, dynamic product ads, and «Boost existing posts» (additional features for Premium Business)

The mention of the Premium Business tier is the only trace of it in open sources

source, checked 2026-07-28

ads come as Promoted Ads and Trend Takeover, all marked with an «Ad» label; delivery is tuned by user actions, profile data, device location, IP address and installed apps

Opting out of personalisation — the «Personalize ads» setting and the optout.aboutads.info tool

source, checked 2026-07-28

the platform claims «10x ROAS average, never below 4x weekly»

There is no methodology, period or sample on the page. A claim about itself, not a measurement

source, checked 2026-07-28

campaigns are run by «eligible accounts» on ads.X.com; the eligibility criteria are not disclosed on the page

source, checked 2026-07-28

Analytics

public_metrics: "Public engagement metrics for the Tweet at the time of the request. Includes retweet_count, reply_count, like_count, quote_count, impression_count, and bookmark_count". non_public_metrics: impression_count, user_profile_clicks, url_link_clicks, engagements. Plus organic_metrics and promoted_metrics

four sets of metrics with different access conditions

source, checked 2026-08-05

"Non-public, organic, and promoted metrics are only available for posts created within the last 30 days."

the harshest stipulation of those found: for a post older than 30 days nobody will get the impressions from the non-public set, nor the link clicks, nor the profile clicks

source, checked 2026-08-05

The advertising API: "We do not support multi-segmentation" — a breakdown is possible only by one attribute at a time, and only from four of them: AGE, GENDER, METROS, PLATFORMS. Some of the ENGAGEMENT metrics "are not available at the account and funding instrument level". LIFE_TIME_VALUE_MOBILE_CONVERSION is "available only to advertiser accounts enabled for MACT". "Segmented data is not expected to roll-up 100% to the non-segmented data, due to how this information is derived"

the last stipulation explains why the sum over the breakdowns does not agree with the overall total — the platform admits it itself

source, checked 2026-08-05

The main API access: non-public, organic and promoted metrics — only posts no older than 30 days. The advertising API: synchronous requests — 7 days, asynchronous ones — 90 days without a breakdown and 45 days with a breakdown

the first part is from the page https://docs.x.com/x-api/fundamentals/metrics

source, checked 2026-08-05

The main API access has no demographic fields; a breakdown by age and gender exists only in the advertising API — the segments AGE, GENDER, METROS, PLATFORMS, one at a time

a creator without an advertising dashboard is not given demographics by the platform

source, checked 2026-08-05

public_metrics — a Bearer Token is enough. non_public_metrics and organic_metrics — "User context (owned posts only)", that is, only one's own posts. promoted_metrics — "User context (promoted posts only)", only promoted posts

there is no way at all to get someone else's statistics beyond the public counters

source, checked 2026-08-05

The advertising API: asynchronous requests make the metrics available "in a downloadable results file after the associated job has finished processing"

this is an export of advertising data; an export of the creator's ordinary statistics is not described in the open documentation

source, checked 2026-08-05

How the money works here

paid subscriptions to an author exist: the create-post schema has a for_super_followers_only field and a subscribers value in the reply setting

Indirect but unambiguous confirmation of the mechanic. The page contains no thresholds or payout shares

source, checked 2026-07-28

the X app listing in the App Store lists subscriptions to individual authors among its in-app purchases: «@elonmusk Subscription — $4.00», «@FatKidDeals Subscription — $1.00»

The app's seller is X Corp. This is store listing data, not the platform's help centre

source, checked 2026-07-28

iOS in-app purchase prices: X Premium Basic $4.00/mo and $42.00/yr; X Premium $11.00/mo and $114.99/yr; X Premium Plus $50.00/mo and $490.00/yr

App store prices, usually higher than web prices because of the commission. The web prices on help.x.com are unavailable

source, checked 2026-07-28

Pricing

reading a post $0.005, reading a user $0.010, publishing a post $0.015, publishing a post with a link $0.200; reading your own data $0.001 per unit

Resources are deduplicated within a UTC day

source, checked 2026-07-28

the pay-as-you-go plan has a ceiling of 2 million posts read per billing month

Material for monitoring services

source, checked 2026-07-28

Limits and restrictions

the platform names no daily limits: limits are set per endpoint on fifteen-minute windows. Separate limits apply per user (when acting on behalf of a user) and per app (when acting on behalf of an app); the remainder is visible in the response headers, and an overrun gives a 429 with code 88

Material for comparison: the scheduler's daily limit (25/50/100 posts) has NO direct counterpart on the platform — X counts differently

source, checked 2026-07-28

publishing a post — 100 per 15 minutes from a user and 10,000 per day from an app; deleting — 50 per 15 minutes; likes — 50 per 15 minutes and 1000 per day; direct messages — 1440 per day; uploading media — 500 per 15 minutes and 50,000 per day; searching recent posts — 450 per 15 minutes

Specific figures by endpoint; for enterprise clients the limits are agreed separately

source, checked 2026-07-28

the remainder is visible in the x-rate-limit-limit, x-rate-limit-remaining and x-rate-limit-reset headers; an overrun gives a 429 and error code 88

source, checked 2026-07-28

Word for word: "Rate limits control the number of requests you can make to each endpoint. Exceeding limits results in a 429 error until the window resets." The windows are 15 minutes and 24 hours. The limits by endpoint: POST /2/tweets — "Per App: 10,000/24hrs", "Per User: 100/15min"; POST /2/media/upload — "Per App: 50,000/24hrs", "Per User: 500/15min"; POST /2/media/upload/initialize, /append, /finalize — "Per App: 180,000/24hrs", "Per User: 1,875/15min". The response headers: x-rate-limit-limit, x-rate-limit-remaining, x-rate-limit-reset

The count is kept separately for each endpoint; with OAuth authorization the limit is per user, and with an app token it is per app. There is no breakdown by tariff (Free, Basic, Pro) on this page

source, checked 2026-08-04

Word for word: POST /2/dm_conversations — "Per App: 1,440/24hrs", "Per User: 15/15min, 1,440/24hrs"

This is the limit of the programmatic sending of private messages. The limit for sending by hand from the application is given on the help.x.com page closed by protection from robots and is not recorded here

source, checked 2026-08-04

Word for word: "Exceeding limits results in a 429 error until the window resets." The response on exceeding: HTTP 429 and the body {"errors": ({"code": 88, "message": "Rate limit exceeded"})}. The moment of removal is visible in the header x-rate-limit-reset (a Unix timestamp)

The slowdown is removed automatically at the end of the window — 15 minutes or 24 hours depending on the endpoint. No separate term of blocking is announced

source, checked 2026-08-04

It was not possible to check the daily limit on posts for an ordinary account: the help centre page "X limits" returns HTTP 403 and a robot-protection splash screen. Through programmatic access it is announced: POST /2/tweets — "Per App: 10,000/24hrs", "Per User: 100/15min"

X historically has a number for manual publication on the page help.x.com/en/rules-and-policies/x-limits, but it is closed by protection from robots: curl with an ordinary browser header receives 403 and a "Just a moment... Enable JavaScript and cookies to continue" page. The protection must not be circumvented, and therefore the number is not recorded. The robots.txt of help.x.com meanwhile permits everything

source, checked 2026-08-04

Through programmatic access, word for word: POST /2/users/:id/following — "Per User: 50/15min"; DELETE /2/users/:source_user_id/following/:target_user_id — "Per User: 50/15min". It was not possible to check the daily limit on follows for an ordinary account

The limits on following and on unfollowing are the same and are set with a window of 15 minutes. X announces the daily number of follows on help.x.com, which is closed by protection from robots

source, checked 2026-08-04

Restrictions

«crawling or scraping the Services in any form, for any purpose without our prior written consent is expressly prohibited»; access is permitted only through the published interfaces

Circumventing protections through «jailbreaking» and «prompt engineering or injection» is banned separately

source, checked 2026-07-28

the developer policy prohibits bulk follows, publishing identical content from several accounts, unsolicited automated messages without consent, and paying users for actions on the platform

source, checked 2026-07-28

stored copies must be synchronised within 24 hours of a deletion or change; sharing post identifiers is capped at 1.5 million per 30 days per person

source, checked 2026-07-28

the developer agreement prohibits training foundation models on X data, embedding content through an iframe, and reselling access «on a service bureau, rental or managed services basis»

Directly relevant to scheduling services and aggregators

source, checked 2026-07-28

the platform reports 181,701,993 user reports, 4,191,171 suspended accounts and 10,131,247 posts removed or labelled; the share of violating posts is 0.0178%

There is a discrepancy in the source itself: the heading assigns the data to July–December 2024, while the caption on the reports table says January–June 2024

source, checked 2026-07-28

Moderation

There is no public strike ladder in the documents available to us. The DSA report: «When we take enforcement actions, we may do so either on a specific piece of content (e.g., an individual post or Direct Message) or on an account» (verbatim); the decision takes into account «The user has a history of violating our policies» and «The severity of the violation» (verbatim) — the history of violations and the severity.

The detailed ladder (warning, read-only, temporary and permanent suspension) is described on help.x.com, which is closed by bot protection; transparency.x.com is allowed (User-agent: * with an empty Disallow).

source, checked 2026-08-07

The DSA report: «Our content moderation systems blend automated and human review paired with a robust appeals system that enables our users to quickly raise potential moderation anomalies or mistakes» (verbatim). The specific appeal forms are hosted on help.x.com, which is closed by bot protection; the report states that affected users are sent a notice with the context of the actions taken.

The block appeal form (help.x.com/forms) is unavailable for checking — the host is closed.

source, checked 2026-08-07

The DSA report (the period October 2024 to March 2025): the median time to respond to account suspension complaints (Account Suspension Complaints Median Time to Respond) across EU countries ranges from 1.1 to 37.1 hours (for most countries, single-digit hours).

The data covers only the EU under the DSA; no global standard for the period is announced.

source, checked 2026-08-07

«you must be at least 13 years old to use the Services» — no younger than 13

The page is served only with scripts enabled; it was read in a browser

source, checked 2026-07-28

access to the service is granted in exchange for the display of advertising: «In exchange for accessing the Services, X and our third-party providers and partners may display advertising to you»

source, checked 2026-07-28

reporting under the Digital Services Act is done by the Irish entity X Internet Unlimited Company (XIUC); the report for the second half of 2025 is posted as a ZIP archive

We did not download the archive, so the European audience figures have not been verified

source, checked 2026-07-28

API access

access on a PAY-PER-USAGE model: «pay-per-usage pricing. No subscriptions — pay only for what you use», credits are bought in advance and drawn down as requests are made; an enterprise tier is announced separately

source, checked 2026-07-28

Capabilities. REST interface v2: search, reading and publishing of posts, work with users, Spaces, direct messages, lists and trends by location

source, checked 2026-07-28

the platform has an X Official Partner programme: a directory of services built on the X API (analytics, moderation, ad management). Participation is BY INVITATION, after an «extensive evaluation», and members undergo continuous re-evaluation

The link was found in the markup of the introduction page, not guessed

source, checked 2026-07-28

a developer account is created on console.x.com, the Developer Agreement and Policy must be accepted and a use case described; keys are shown once

Sign-in methods: Bearer Token, OAuth 1.0a, OAuth 2.0

source, checked 2026-07-28

the filtered stream supports up to 1000 rules; full-archive search covers posts back to 2006

source, checked 2026-07-28

the X Official Partner programme is described as an «invitation-only program»; selection follows «an extensive evaluation» and members undergo continuous re-evaluation; there are no names on the page

Confirms the earlier fact in the database; the partner list is still not served statically

source, checked 2026-07-28

Languages

the interface language of the home page: en

The language is taken from the markup attribute, the versions from hreflang

source, checked 2026-07-28

metrics

A video view in the advertising analytics follows the MRC standard, verbatim: «The video_total_views metric within the VIDEO metrics group will report on any views which are at least 50% in-view for 2 seconds, as per the MRC standard» (at least 2 seconds with at least 50 % of the player in view); the former definition is preserved as a separate metric, video_3s100pct_views (100 % of the player for at least 3 seconds). For the organic view_count of a video no threshold is published, verbatim: «Aggregated across all posts containing the video. A video reposted in multiple posts has one total view count»

The help centre help.x.com returns 403 (bot protection) — the API documentation was used instead; docs.x.com is open to everyone in robots.txt. The organic quotation comes from https://docs.x.com/x-api/fundamentals/metrics.md

source, checked 2026-08-07

X publishes no reach metric; in its place are impressions, which are expressly defined as non-unique, verbatim: «Count of times the post appeared on a user's screen. Not unique—the same user viewing twice counts as two impressions»

partial: there is no definition of reach either in the open API docs (docs.x.com was read) or in the help centre (help.x.com — 403); the definition of impressions is recorded as the nearest metric

source, checked 2026-08-07

In the open documentation engagement is given only as a formula: Engagement Rate = engagements/impressions; the composition of the actions is not spelled out — the engagements field is described tautologically («Total number of engagements»)

partial: the list of actions that count as engagement exists only in the closed help centre help.x.com (403); read: docs.x.com/x-ads-api/analytics.md and docs.x.com/x-api/fundamentals/metrics.md

source, checked 2026-08-07

Catalogue section: all similar See also: catalogue index · region: USA and Canada · scheduler comparison · find by situation · platform restrictions