{
  "id": 13289256,
  "title": "A 10-Point Scorecard for Vetting Telegram Channels Before Adding Them to Your Collection Set",
  "url": "https://urgent.news/2026/10/10/a-10-point-scorecard-for-vetting-telegram-channels-before-adding-them",
  "topic": "tech",
  "section": "Tech",
  "published": "2026-10-10T01:29:19.000Z",
  "source": {
    "name": "Dev.to",
    "slug": "dev-to",
    "url": "https://dev.to/yuhehe/a-10-point-scorecard-for-vetting-telegram-channels-before-adding-them-to-your-collection-set-2hfd"
  },
  "original_language": "en",
  "account": "Creating a reliable Telegram OSINT collection set begins with carefully vetting the channels you add. To ensure the quality of your collection, I recommend using a 10-point scorecard that assesses candidate channels across five key criteria. Each criterion is scored on a scale of 0 to 2, with a minimum overall score of 7 for inclusion in the set. Channels are then reviewed again with scores between 5 and 6, while anything below that threshold is rejected.\n\nThe scorecard covers five main axes: origin, language consistency, cadence regularity, audience quality proxy, and link hygiene. To evaluate the origin of a channel, check if it creates original reporting or simply republishes content. Look for original photos, videos, and consistent watermarks, as well as posts that other channels quote and link to. Channels that only republish content without adding original reporting should score 0. Pure republishers are not useless, but they are not valuable sources either.\n\nLanguage consistency is another important factor. A well-run reporting channel should consistently report in one language with a clear target audience. If a channel switches languages frequently in an attempt to attract more engagement, it is likely a content farm. You can automate this assessment using a language-detect script to flag any language inconsistencies.\n\nCadence regularity assesses the posting frequency of the channel. Real operational channels, such as military administration, emergency services, or shipping feeds, tend to post in bursts with structured rhythms and quiet periods. In contrast, content farms typically post at a consistent, machine-like cadence throughout the day, including odd hours like 06:00 UTC. Analyzing the posting series using autocorrelation can help separate these types of channels effectively.\n\nThe audience quality proxy evaluates the quality of the channel's audience. For channels with more than 1000 subscribers, you can analyze the view counts provided in the public preview. Compute the view-to-subscriber ratio and assess the variance of per-post views. Genuine audiences will show high variance, with some posts being more important than others. Purchased or artificially inflated audiences, on the other hand, will display suspiciously uniform view counts across posts.\n\nLastly, link hygiene examines the outbound links shared by the channel. Scan the last 20 posts for links to suspicious domains, such as crypto giveaway sites, URL shorteners, or gambling affiliates. These types of links should score 0. Instead, look for links to primary sources like official statements, maps, or registries and award a score of 2 for those. This criterion ensures that the channel is sharing valuable and credible information, rather than spreading misinformation or promoting dubious content.\n\nBy applying this scoring system to each channel, you can create a curated collection set that is both high-quality and reliable. The scorecard can be implemented using publicly available information, without requiring any API keys or paid data sources. The full 11-point version, including additional rules for admin provenance, cross-post graph position, content-farm fingerprints, and more, is available in the Telegram and Web OSINT Bundle for $5. A free sample brief is also provided for reference. This approach ensures that your collection set delivers high-quality and trustworthy information, ultimately enhancing your alert quality.",
  "summary": "Every Telegram OSINT collection set starts as a pile of channels someone pasted from a Twitter thread. Most of that pile is junk, and the junk is expensive: every junk channel in your collection set adds review load, dedup noise, and false-alarm alerts forever. I classify candidate channels on five axes before they enter the set. Score each 0-2; keep at 7+, review at 5-6, reject below. 1. Origin…",
  "key_points": [
    "Scorecard assesses Telegram channels across origin, language, cadence, audience, and link hygiene",
    "Channels must score at least 7 overall to be included in collection set",
    "Origin evaluation checks for original reporting and watermark consistency"
  ],
  "editors_take": "This vetting process enables OSINT collectors to systematically assess Telegram channels and build a reliable collection set by filtering out low-quality or suspicious sources and prioritizing those with original reporting and credible information.",
  "illustration": null,
  "coverage": {
    "outlets": 1,
    "also_reported_by": []
  },
  "ai_generated": true,
  "disclaimer": "Summaries, key points and the editor’s take are written by software from other outlets’ reporting and may contain errors — always check the linked original."
}