Product market intelligence from demand, attention, and news signals

Product and strategy teams build conviction faster when search demand, community scale, encyclopedic attention, and news intensity can be queried as structured series in one place. Trends MCP supplies those channels to assistants and to backend jobs through MCP and REST.

What people search for when they want "market intelligence" inside an AI client

Typical queries read like multi platform evidence requests. A PM might ask for category demand on Google Search, learning intent on YouTube, community momentum on Reddit, and background attention on Wikipedia for the same entity list. Without tools, models guess at co movement. With Trends MCP, each source returns a time ordered payload the assistant can chart or tabulate.

A simple evidence stack for a PRD narrative

Use get_trends on the core problem phrase across google search and youtube, then add wikipedia for the named product category leaders. When the space is noisy in the press layer on news volume and news sentiment for the same entities referenced in the PRD. If the thesis includes commerce, add amazon for the SKU level language buyers actually type.

When growth math matters more than a single curve

get_growth answers the board style question directly: did interest rise over YTD, 6M, and 12M windows on a chosen source. Keeping windows in one call per source and keyword respects quotas and reduces copy paste between slides.

What this page deliberately leaves out

Trends MCP does not replace primary interviews, pricing experiments, or warehouse telemetry. It strengthens external demand and attention chapters with reproducible pulls. Teams should still attach their own internal metrics beside any external series shared in a decision memo.

Nearby pages that extend the same workflow

Product managers who want role framed examples can read Trend data for product managers. Analysts who need a wider survey of workflows should scan AI powered market research with live trend data. Teams who worry about false positives when one channel spikes can use Multi source trend validation without a dozen logins as a checklist style companion.

Common questions

Start with Google Search for category demand, YouTube search for how people look for explanations, Reddit for community scale using a real subreddit name without the r prefix, and Wikipedia for information seeking spikes on key entities. Add news volume and news sentiment when the category faces regulatory headlines.
Treat 0 to 100 values as comparable within a single source over time rather than as literal shares across platforms. When the assistant writes a summary, ask it to keep per source language explicit so readers do not merge TikTok hashtag volume with Google Search volume as if they shared units.
Yes. POST https://api.trendsmcp.ai/api mirrors the MCP operations described in the public docs. Services should read the Bearer token from a secret store, cap concurrency, and log the as_of timestamps that appear on top trend responses.
The free tier includes a hundred requests per month with no credit card. Teams often reserve daily top trend pulls for a small set of feeds and spend the rest on weekly deep dives for a fixed keyword basket. Upgrade paths are listed on the pricing page when volume grows.
Get Product market intelligence from demand, attention, and news signals in 30 seconds
Free tier includes 100 requests per month. No credit card required. Works with Claude, Cursor, ChatGPT, Raycast, and every MCP client.
Get your free API key