The social media search engine
They search the open web. SocialQuery searches the live social web: one natural-language query, source planning, parallel fan-out, ranked evidence, and clustered themes.
Crawling data from the world's biggest platforms
The engine
One query in. One ranked, clustered answer out.
Watch a single question move through the pipeline: planned, pruned, fanned out, fused, reranked on what people said, and clustered into themes.
Query
what are people saying about ___?
Natural-language question
Universal Social Search Engine
01
AI plan
Turn a plain-language question into source-specific searches.
02
AI prune
Drop weak sources before they spend credits.
03
Parallel fan-out
Query the relevant social sources at the same time.
04
RRF fuse
Merge repeated evidence without duplicating the same story.
05
AI rerank
Use comments, transcripts, and engagement context to order results.
06
Cluster
Group the answer into themes your product can use.
Answer
How it works
How does one query search 17 places at once?
Three moves turn a natural-language question into a ranked, clustered answer.
01 - Plan and prune
An AI reads intent, expands the query into source-specific searches, and drops sources that will not help before they bill.
02 - Fan out and fuse
Every relevant social source runs in parallel, then repeated evidence is fused so one idea can rise across platforms.
03 - Rank on what people said
Comments, transcripts, posts, and engagement context shape the final ranking and clusters.
Why it is hard to copy
No single trick. The whole pipeline.
The advantage is the complete integration: planning, pruning, cross-source fusion, grounded reranking, streaming, and correct billing on partial failures.
Plan + prune
Two reasoning passes score the request before a provider runs.
Cross-source fusion
Weighted reciprocal ranking lifts the same signal across platforms.
Grounded rerank
Social details like comments and transcripts change the order.
Streaming ledger
Each source reports progress, credits, and final status.
Social vs open web
They search the web. We search social.
| Compare | SocialQuery | Open-web search |
|---|---|---|
| What it searches | Social platforms, forums, videos, comments, and social feeds | Public web pages and documents |
| What a result is | A real post, comment, video, or discussion with engagement context | A page, article, or crawled document |
| Ranking signal | Conversation, repetition, engagement, and source context | Page relevance and crawlability |
| Best for | Brand monitoring, social research, creator intelligence, and live public conversation | Web research, citations, and document discovery |
Try it
Ask anything. Watch it stream.
Run a universal social search in the browser. Results arrive as each source resolves, then collapse into the same final answer your API call receives.
Illustrative stream · planned
Reddit, X, YouTube, TikTok and more
Under the hood
Every one of those sources returns a different shape.
Before ranking, SocialQuery normalizes each result into a canonical object with source metadata, engagement fields, language, category, reach estimates, and request details.
See how the unified schema works{
"type": "post",
"platform": "reddit",
"text": "Customers are comparing our brand...",
"author": { "handle": "researcher" },
"metrics": { "comments": 42, "score": 318 },
"computed": { "engagement_rate": 0.037 }
}At a glance
One call. Every platform.
17
sources per search
17
sources at once
20
credits per API call
60
RRF constant
Frequently asked questions
Can't find what you're looking for? Talk to our team.
How is Universal Search different from Exa, Tavily, and Firecrawl?
Those products are strongest on open-web pages. Universal Search is tuned for social posts, comments, videos, forums, and live conversation surfaces.
How many platforms does one query search?
A query can fan out across up to 17 social and research sources, with pruning so irrelevant sources do not run.
How are results ranked across different platforms?
Results are fused across sources, then reranked with social context such as engagement, comments, transcripts, author context, and repeated evidence.
What does reranked on real comments and transcripts mean?
The final pass can use post text, comment context, transcript snippets, and source metadata instead of ranking only on page-level relevance.
Is it free?
New accounts start with 100 free credits, and the API reports credit usage with each response.
Can I stream results?
Yes. The API can return source events and partial rankings before the final clustered answer is complete.
Start searching
Search social like it's one place.
Run a universal social search in the browser, or call one endpoint from your code.
curl https://socialquery.dev/v1/search/everywhere \
-G --data-urlencode "query=what are people saying about our brand?" \
-H "x-api-key: $SQ_KEY" \
-H "Accept: text/event-stream"