Telegram Channel Scraper & Keyword Monitor

Scrape posts, views, reactions, links and media from any public Telegram channel — no Telegram account, no API key, no phone number. Turn on keyword alerts and new-posts-only mode to monitor crypto, news, OSINT, brand mentions or competitors on a schedule — or get TGStat-style channel analytics (ERR, average views, posting frequency, best time to post, forward sources) for any public channel — and find similar Telegram channels in a niche from a few seed channels.

Why use it

• 🔓 No login or API keys — reads Telegram's public web preview (t.me/s/channel).
• 📈 Rich data — text, date, author signature, views, reactions (per emoji + total), engagement rate, outbound links, link previews (title + description), forwarded-from, reply-to link, edited flag.
• 🖼️ Media URLs — direct photo and video URLs, video thumbnails and durations.
• #️⃣ Hashtags & @mentions extracted from every post, plus polls (question, options, % per option) and file names.
• 🔥 Popular-posts filter — minViews keeps only posts above a view threshold (you don't pay for skipped posts).
• 🚀 Virality score (new) — every post gets viralityScore = its views ÷ the median views of the ~20 own posts published around it (age-matched, because Telegram views keep growing for weeks), plus isViral (2×+), viewsPerSubscriberPct and viewsStillGrowing. minViralityScore: 2 outputs only the posts that beat the channel's normal reach — works across channels of any size, unlike a fixed view threshold. Analytics rows add viralPosts, viralShare and maxViralityScore.
• 🔔 Monitoring — onlyNewPosts remembers the last post per channel, so scheduled runs return only fresh posts. You never pay twice for the same post.
• 🎯 Keyword alerts — output only posts mentioning your brand, tickers, CVEs, cities…
• 🔎 Full-history keyword search — searchFullHistory uses Telegram's own in-channel search, so a keyword finds matching posts from years back, not just among the latest posts. No need to download a whole channel to find 20 mentions.
• 💬 Comments — includeComments adds commentCount and the latest comments (text, date, reply-to, reactions; up to 100 per post) for channels with a public discussion group — audience reaction and sentiment, not just views. Commenter identities are not collected.
• 🔗 Direct post links — paste https://t.me/durov/300 to get exactly that post (views, reactions, media).
• 📉 Channel analytics — includeAnalytics adds one row per channel: posts per week, average/median views, ERR (views per subscriber %), engagement rate, views trend, best weekday & hour to post, content mix with average views per type (photo/video/text/link/poll), top 5 posts, top hashtags, mentioned channels, forward sources (the channels it reposts from — great for finding related channels) and linked domains. analyticsOnly skips post rows to compare many channels cheaply.
• 🧭 Similar / related channel finder — discoverRelatedChannels turns the channels your seeds forward from, link to (t.me/...) or @mention into ranked related-channel rows: title, description, subscribers, relevance score, which seeds reference it, last post date, recent average views and ERR. Every candidate is checked to be a public channel — users, bots and groups are never output. relatedChannelsOnly skips post rows. No login or subscriber-overlap data needed.
• 📊 Channel stats — title, description, exact subscriber count (10,480,121 — not the rounded "10.5M" of the preview page), photo/video/link counts.
• 📈 Subscriber growth tracking (new) — every run saves a snapshot per channel (subscribers + median views of recent posts). Schedule it daily and each channel row gets subscribersChange since the last run, subscribersChange7d / subscribersChange30d (count and %), avgDailySubscriberGrowth, growthTrend (growing / flat / shrinking), recentMedianViewsChange7dPct (is reach rising or falling?) and a subscriberHistory of the last 14 snapshots. With 2+ channels a GROWTH record ranks them by growth. TGStat-style growth charts for your own watchlist, without an account.
• 📚 History — paginates back through older posts (set maxPostsPerChannel).

Use cases

• Crypto & trading signals — track announcement and alpha channels, get new posts within minutes.
• News & OSINT — monitor breaking-news and conflict channels by keyword.
• Brand monitoring — find mentions of your product or company.
• Market research / influencer analysis — compare engagement (views, reactions) across channels.
• Content research — find what actually overperforms in a niche: minViralityScore: 2 over 10 competitor channels returns only their breakout posts (topic, format, hour), ready to copy the pattern.
• Competitor & channel growth tracking — a daily schedule over your and your competitors' channels shows who gains subscribers, who loses them, and whether reach keeps up (trackGrowth, on by default).
• Telegram ad buying — vet a channel before paying for a post: real reach (ERR), views trend, posting frequency and whether it mostly reposts others (includeAnalytics).
• Niche mapping & ad-placement lists — start from 3–5 channels you know, get the 20–100 related channels with subscribers and real reach (ERR), and spot active vs abandoned ones (relatedChannelsOnly).
• Audience sentiment — read what subscribers say under announcements, launches or news (includeComments).
• AI datasets — collect channel content for summarization, sentiment or RAG.

Input example

{
  "channels": ["durov", "@telegram", "https://t.me/s/bbcnews"],
  "maxPostsPerChannel": 100,
  "keywords": ["airdrop", "listing"],
  "minViews": 5000,
  "onlyNewPosts": true,
  "stateKey": "crypto-alerts"
}

Find every post that ever mentioned a topic, across the whole channel history:

{
  "channels": ["durov", "tginfo"],
  "keywords": ["privacy"],
  "searchFullHistory": true,
  "maxPostsPerChannel": 200
}

Posts with their comment threads (comment count + latest 30 comments):

{
  "channels": ["tginfo"],
  "maxPostsPerChannel": 20,
  "includeComments": true,
  "maxCommentsPerPost": 30
}

Channel analytics only — compare channels before buying ads (no post rows, one analytics row each):

{
  "channels": ["durov", "tginfo", "cointelegraph"],
  "maxPostsPerChannel": 150,
  "analyticsOnly": true
}

Find related channels in a niche — one row per related public channel, best first:

{
  "channels": ["meduzalive", "readovkanews", "tginfo", "durov"],
  "maxPostsPerChannel": 200,
  "relatedChannelsOnly": true,
  "maxRelatedChannels": 20,
  "minRelatedSubscribers": 5000
}

Output example

{
  "type": "post",
  "channel": "durov",
  "postId": 548,
  "url": "https://t.me/durov/548",
  "date": "2026-09-11T16:04:02.000Z",
  "text": "Telegram has become the sponsor of Codeforces ... #contest @codeforces_official",
  "author": "Pavel Durov",
  "views": 1990000,
  "reactions": [{ "emoji": "👍", "count": 51000 }],
  "reactionsTotal": 70360,
  "engagementRate": 0.0354,
  "media": ["video"],
  "mediaItems": [{ "type": "video", "url": "https://cdn4.telesco.pe/file/....mp4", "thumbnail": "https://cdn4.telesco.pe/file/....jpg", "duration": "0:20" }],
  "links": ["https://codeforces.com/blog/entry/156620"],
  "hashtags": ["#contest"],
  "mentions": ["@codeforces_official"],
  "linkPreview": "https://codeforces.com/blog/entry/156620",
  "linkPreviewTitle": "Telegram Returns as Title Sponsor of Codeforces!",
  "poll": null,
  "document": null,
  "forwardedFrom": null,
  "replyToUrl": null,
  "matchedKeywords": []
}

Every post also carries virality fields (real values for @durov post 527, the $200,000 contest announcement, June 2026):

{
  "views": 24300000,
  "channelMedianViews": 7040000,
  "viralityScore": 3.45,
  "isViral": true,
  "viewsPerSubscriberPct": 231.4,
  "viewsStillGrowing": false
}

With includeComments each post also has:

{
  "commentsEnabled": true,
  "commentCount": 137,
  "comments": [
    { "commentId": 1810549, "date": "2026-04-14T03:45:22.000Z", "text": "Great idea, finally ...", "replyToCommentId": null, "hasMedia": false, "reactions": [], "reactionsTotal": 0 }
  ]
}

Channel analytics row (type: "channel-analytics", real numbers from the latest 150 posts of @cointelegraph, shortened):

{
  "type": "channel-analytics",
  "channel": "cointelegraph",
  "subscribers": 332000,
  "basis": "latest-posts",
  "postsAnalyzed": 150,
  "periodStart": "2026-09-22T19:25:01.000Z",
  "periodEnd": "2026-10-06T13:30:03.000Z",
  "postsPerWeek": 76.3,
  "avgViews": 9681,
  "medianViews": 9630,
  "errPct": 2.9,
  "engagementRate": 0.0075,
  "viewsTrendPct": -9.7,
  "bestWeekdayUtc": "Wednesday",
  "bestHourUtc": 11,
  "byWeekday": [{ "weekday": "Wednesday", "posts": 26, "avgViews": 10080 }],
  "contentMix": [{ "kind": "photo", "posts": 123, "avgViews": 9699, "share": 0.885 }],
  "forwardedShare": 0,
  "withLinksShare": 0.987,
  "topLinkDomains": [{ "domain": "cointelegraph.com", "count": 148 }],
  "forwardSources": [],
  "topPosts": [{ "url": "https://t.me/cointelegraph/72272", "views": 15500, "kind": "photo" }]
}

For a news aggregator like @meduzalive, forwardSources lists the channels it reposts: meduza_news (11 posts), plottwist_m (3), meduza_books (2).

Related channel row (type: "related-channel", real result from the run above, shortened):

{
  "type": "related-channel",
  "channel": "betainfo",
  "url": "https://t.me/betainfo",
  "title": "Beta Info",
  "subscribers": 25600,
  "relevanceScore": 58,
  "seedChannels": ["tginfo"],
  "seedCount": 1,
  "forwardsFromSeeds": 0,
  "linksFromSeeds": 29,
  "mentionsBySeeds": 0,
  "examplePostUrl": "https://t.me/tginfo/4435",
  "lastPostAt": "2026-10-06T21:14:00.000Z",
  "daysSinceLastPost": 0.4,
  "recentAvgViews": 7154,
  "recentPostsMeasured": 17,
  "errPct": 27.9
}

Other related channels found from those 4 seeds: meduza_news (122K subscribers, 15 forwards), tandroidapk (258K), uranews (220K), contest (referenced by both tginfo and durov), plottwist_m, alfabank…

Channel record (type: "channel"): title, description, subscribers, photos, videos, links, verified, image.

Pricing

• $1 per 1,000 posts
• $2 per 1,000 channel-info records
• $0.01 per channel-analytics row (only with includeAnalytics / analyticsOnly; with analyticsOnly posts are not charged)
• $0.003 per related-channel row (only with discoverRelatedChannels / relatedChannelsOnly; with relatedChannelsOnly posts and channel info are not charged)

No subscription. In monitoring mode, already-seen posts are never charged again.

Scheduling alerts

1. Set onlyNewPosts: true, add keywords (optional), name the monitor.
2. Create a Schedule in Apify Console (e.g. every 15 minutes or hourly).
3. Connect Slack, Telegram bot, email, Google Sheets, webhook, Zapier or Make to receive new posts automatically.

Limits

Works for public channels that have web preview enabled. Private channels, groups and bots are not supported. Only publicly visible data is collected. Subscriber growth starts with your first run — past growth before that is not available from Telegram.

FAQ

Do I need a Telegram account, bot token or API ID? No. The Actor reads the public web preview at t.me/s/<channel>.

How long are media URLs valid? Photo/video URLs point to Telegram's CDN and contain a temporary token — download files soon after the run if you need copies.

How do I get only viral posts? Set minViralityScore to 2: only posts with at least twice the views of the posts published around them in the same channel are output (skipped posts are not charged). Unlike minViews, this works for a 5K-subscriber channel and a 10M one in the same run. In our test over the latest 100 posts, @telegram had 1 viral post, @tginfo 2 and @durov 12 (his June 2026 contest and UK-ban posts at 3–3.5× his usual reach). You can still use minViews (e.g. 50000) for an absolute threshold, and sinceDate for "top posts this week".

How is viralityScore calculated? Post views ÷ channelMedianViews, the median views of up to 10 own posts published just before and 10 just after it (at least 5 needed). Comparing posts of a similar age matters because Telegram views keep growing for weeks: against the channel's overall average, old posts would always look viral. Forwarded posts show the original post's views, so they get no score. The very newest posts are compared mostly with older posts and score lower until their views catch up (viewsStillGrowing marks posts younger than 24 h), so for alerts on fresh posts use 1.5 instead of 2. With onlyNewPosts the channel's latest page is used as context, so even 1–3 new posts get a score.

What is engagementRate? Total reactions ÷ views for the post — handy for comparing channels of different sizes.

How is "Search full channel history" different from keywords alone? Without it, the Actor reads the latest maxPostsPerChannel posts and keeps those that contain a keyword. With it, Telegram's own search finds matches anywhere in the channel's history (e.g. Durov's posts about "privacy" back to 2022), and maxPostsPerChannel limits the number of matches. It also works with onlyNewPosts for alerts.

Can I scrape specific posts? Yes — add post links like https://t.me/durov/300 to channels. Each returns one post record.

Can it scrape comments? Yes — turn on includeComments. Works for channels that have a public discussion group (comments under posts); you get commentCount plus the latest comments (up to 100 per post, set maxCommentsPerPost; 0 = count only). For channels without comments commentsEnabled is false and, after 3 posts, the Actor stops checking that channel so runs stay fast. Comment rows contain text, date, reply-to and reactions only — no commenter names or usernames. There is no extra charge for comments.

What is ERR and how are the analytics calculated? ERR = average views per post ÷ subscribers (the standard Telegram reach metric; above ~20% is typical of a healthy channel, low single digits often means inactive or bought subscribers). All analytics are computed from the posts the run scanned (maxPostsPerChannel; with searchFullHistory — the matching posts, see basis). View metrics skip posts younger than 24 hours because their views are still growing. Weekdays and hours are in UTC. No AI is used, so numbers are reproducible.

How do I find similar / related Telegram channels? Turn on relatedChannelsOnly and give 3–5 seed channels from the niche. The Actor scans their latest posts (maxPostsPerChannel, 200+ recommended), collects every channel they forward from (weight 3), link to with t.me/... (weight 2) or @mention (weight 1), adds +5 for each extra seed referencing the same channel, checks that each candidate is a public channel and outputs the best maxRelatedChannels with subscribers and ERR. Feed the results back in as seeds to go one level deeper.

How is this different from "similar channels" by subscriber overlap? Subscriber-overlap tools need a logged-in Telegram account. This Actor uses only public posts: a related channel is one your seeds actually repost, promote or link to — which is exactly how Telegram cross-promotion and ad placements work. Channels that never interact with your seeds won't appear; use more seeds or more posts for wider coverage.

How does subscriber growth tracking work? Telegram does not publish a channel's subscriber history, so the Actor builds it: each run stores one snapshot per channel (exact subscribers and the median views of recent own posts older than a day) under your stateKey. The first run starts the history; from the second run on you get subscribersChange, and once the history covers about a week / a month, subscribersChange7d / subscribersChange30d (the snapshot closest to 7 days ago, ±2 days; 30 days ±6). Runs less than an hour apart replace the last snapshot instead of adding one. Up to 120 snapshots are kept per channel. Use a different stateKey per schedule, or the same one to share a history. Set trackGrowth: false to turn it off.

Does growth tracking cost extra? No. It adds one request per channel (the channel page with the exact subscriber count) and no extra rows.

Can it scrape private groups? No — only public channels that Telegram shows on its web preview.

More tools from the same developer

• Company Jobs Scraper — every open job from Workday, Greenhouse, Lever, Ashby, Personio & more, with salaries and new-job alerts.
• Workday Jobs Scraper — any myworkdayjobs.com careers site.
• App Store & Google Play Reviews Scraper — reviews, ratings and new-review monitoring.
• Shopify Products Scraper & Price Change Monitor — catalogs plus price-drop / restock alerts.
• Hacker News "Who is Hiring" Parser — monthly HN hiring threads as structured jobs.