Go-to-market teams run faster when demand signals sit next to copy, decks, and outbound drafts. Trends MCP connects Claude, Cursor, ChatGPT, and other MCP clients to search, news, Reddit, and web traffic trends so positioning and pipeline bets trace back to observable interest.
SaaS go-to-market work moves in tight loops: positioning drafts, outbound sequences, launch calendars, and weekly pipeline reviews. When those loops sit inside an AI assistant, the missing piece is often fresh demand data the model can cite without hallucinating.
Trends MCP supplies that layer. It exposes normalized trend series, multi-window growth math, and live ranked feeds across search, news, social, and web traffic signals through one MCP endpoint and the same contract on REST. Teams keep the assistant inside facts pulled from the API instead of guesses pulled from memory.
Search interest is a blunt but honest signal for whether a problem statement is heating up or flattening. News volume shows when journalists and newsletters pile onto a theme, which matters for enterprise categories where press drives inbound spikes.
With Trends MCP, an assistant can call get_growth on the exact phrases the team debates, compare preset windows such as 3M and 12M, and return the recent and baseline dates the math used. That makes weekly reviews less about opinions and more about whether the market’s language is drifting.
For deeper inspection, get_trends on Google Search or Google News gives the series behind the headline percent change. GTM leaders can see whether growth is a steady climb or a short burst tied to a single event.
Some B2B motions sell into practitioners who organize in public communities. Reddit subscriber growth for a named subreddit is a coarse but useful check on whether that community is attracting newcomers faster than usual.
The Top Websites style feed inside Trends MCP (labeled SimilarWeb Top Websites in product docs) helps teams name which properties dominate attention inside a category. That does not replace first-party analytics, yet it gives copywriters concrete examples when they explain where buyers spend time online.
Models route more reliably when prompts include explicit routing text. The project documentation recommends phrasing like “Using TrendsMCP, show 12M Google Search growth for {keyword}” so the client picks the MCP tool instead of opening a generic web search.
Keep keywords concrete. Broad tokens like “AI” blend enterprise, consumer, and academic meanings. Tight phrases like “SOC 2 automation” or “data activation platform” produce cleaner series.
The free tier includes a modest monthly request budget suitable for experiments. Each get_trends or get_growth call counts as one request for a single source and keyword. get_top_trends counts per feed page. Plan batch reviews accordingly.
Trend data can miss thin or brand-new phrases, and upstream gaps occasionally return data_unavailable. Good workflows treat empty responses as a signal to narrow the keyword or try an adjacent phrase rather than forcing a story.
For broader market context, see the market research page. For outbound and account research angles, sales intelligence trend data covers how revenue teams use the same signals. Teams using Cursor day to day can start from MCP server for Cursor.
Tools for this workflow
get_trendsPlot weekly or daily Google Search or Google News interest for a category, problem phrase, or competitor name before rewriting homepage hero copy.
get_trends(source="google search", keyword="customer data platform", data_mode="weekly")
get_growthMeasure quarter-over-quarter and year-over-year momentum for a positioning phrase or product category across search, Reddit, or Amazon where relevant.
get_growth(source="google search", keyword="revenue operations software", percent_growth=["3M", "12M", "YTD"])
get_top_trendsScan live Google Trends leaders or Reddit Hot Posts for timely campaign hooks that still fit brand risk guidelines.
get_top_trends(type="Google Trends", limit=25, offset=0)
FAQ