Threads Topic Scraper — compare keywords by engagement

Put several keywords side by side on Meta Threads and get one row per topic, ranked by how much engagement a typical post in that conversation actually earns.

Post counts tell you what people are talking about. They do not tell you where a new post has a chance of being seen. A crowded topic where one viral thread absorbs everything is a worse place to enter than a quieter one where reach is spread — and that difference only shows up when you compare topics on the same axes.

What you get per topic

Field	Example
engagementRank	1
keyword, posts	ai agents, 20
medianEngagement	84
maxEngagement, totalEngagement	1912, 3140
topShare	0.61
verifiedShare	0.15
medianReplyRatio	0.09
withMediaShare	0.4
medianAgeHours	6.2
language, languageShare	en, 0.85

engagementRank is ordered by medianEngagement, not by total. Total engagement simply rewards whichever topic has more posts, which is not the question — you want to know what a typical post earns.

topShare is the share of all engagement held by the top 10% of posts. Near 1 means a couple of threads own the conversation and a new post is likely to be invisible. Lower means attention is distributed.

verifiedShare shows how much of the topic is being driven by verified accounts. A topic dominated by large accounts behaves very differently from one where ordinary posts still land.

medianReplyRatio is replies as a share of all reactions. High values mean people are arguing rather than simply agreeing — useful if you want discussion, and a warning if you do not.

medianAgeHours is the freshness of the sample. If the typical post is days old, the phrase is not currently moving.

Summary (key-value store, TOPIC_SUMMARY)

topics, topicsWithPosts, medianOfMedianEngagement, hottest, and empty — the keywords that returned no posts at all.

Input

{
  "keywords": ["ai agents", "vibe coding", "prompt engineering"]
}

Two or more keywords are required. One keyword is not a comparison, and the run is rejected rather than returning a single row that answers nothing. If you want the post list for one keyword, use Threads Search Scraper with engagement metrics instead.

Scope and limits — please read

Threads returns roughly 20 posts per search and this Actor does not scroll. Every topic is therefore measured on a sample of that size, and raising samplePerKeyword will not produce more. That is deliberate: all topics are compared on samples of the same shape, which is what makes the ranking meaningful. It is not an exhaustive count of everything posted about a topic and should not be read as one.

Keywords that return no posts are still returned as a row with posts: 0 and engagementRank: null. They are not silently dropped, so you can tell "nobody is talking about this" apart from "you misspelled it".

This Actor reads public search results only. It does not log in, does not access profiles that require authentication, and does not target private accounts.

Display names are not returned. The underlying data carries the @handle and whether an account is verified, but not the profile's real-name field — and this Actor's output is aggregate per topic in any case.

Pricing

Charged per topic, not per post — one search covers a whole topic, and the price reflects that.

Disclaimer

This is an unofficial tool and is not affiliated with, endorsed by, or sponsored by Meta or Threads. "Threads" and "Meta" are trademarks of their respective owners and are used here only to describe what this Actor reads. You are responsible for ensuring your use of the collected data complies with applicable law and with Meta's terms.