Google Trends API & Scraper + AI Trend Insights

Get reliable Google Trends data for any list of keywords:

• interest over time
• interest by region
• related queries (top and rising, including "Breakout")
• related topics
• multi-keyword comparisons
• today's trending searches

On top of the data, an AI analyst explains what the trend means: a plain-language summary, why each spike happened, where the trend is going, which keywords to target, and ready-to-use content ideas for your blog, YouTube channel, store or social media.

Failed keywords are never charged. The run finishes cleanly before your timeout instead of timing out.

Use cases

• Bloggers, YouTubers and creators: know when a topic peaks and get content ideas that match rising searches.
• SEO and keyword research: compare keywords on one scale, find breakout queries and regional demand.
• E-commerce and product research: validate demand and seasonality before you stock or launch a product.
• Marketing and agencies: weekly trend reports for clients, in any language (for example Indonesian).

Quick start

1. Click Start with the default input (pickleball, air fryer and a ChatGPT vs Gemini vs Claude comparison). It finishes in 1–2 minutes for under $0.10.
2. Put your own keywords, geo and timeRange. Turn aiInsights off if you only need the raw data.
3. Schedule it weekly and connect the dataset to Google Sheets, Make or Zapier.

Why this Actor

	This Actor	Typical Google Trends scrapers
Rate limits (429)	Every retry uses a new IP (Apify Proxy session rotation), exponential backoff, automatic escalation from datacenter to residential IPs	Retry on the same IP, or the run fails
Long runs	Time budget: stops starting new keywords before the run timeout and finishes with status SUCCEEDED	Run ends as TIMED-OUT and data is lost
Partial failures	A failed keyword never fails the run. Pieces fetched in earlier attempts are kept	One bad keyword can fail everything
Billing	Only successful reports are charged. Keywords Google has no data for are free	Charged per dataset row, errors included
Analysis	Trend direction, % change, robust spike detection, seasonality, breakout queries, plus AI insights	Raw numbers only
Output	One clean JSON report per keyword or comparison, with table views	Many fragmented rows

What you get per keyword

{
  "type": "keyword",
  "keywords": ["pickleball"],
  "geo": "US",
  "timeRange": "today 12-m",
  "interestOverTime": [{ "date": "2026-04-12T00:00:00.000Z", "formattedTime": "Apr 12 – 18, 2026", "values": { "pickleball": 100 }, "isPartial": false }],
  "interestByRegion": [{ "geoCode": "US-HI", "geoName": "Hawaii", "values": { "pickleball": 100 } }],
  "relatedQueries": { "pickleball": { "top": [{ "query": "pickleball paddle", "value": 100, "formattedValue": "100", "isBreakout": false }], "rising": [{ "query": "best pickleball paddles 2026", "value": 5000, "formattedValue": "Breakout", "isBreakout": true }] } },
  "relatedTopics": { "pickleball": { "top": [], "rising": [] } },
  "stats": {
    "pickleball": {
      "average": 47.3, "latest": 42, "peak": 100, "peakDate": "2026-04-12T00:00:00.000Z",
      "changePct": 7.9, "slopePerPeriod": 0.12, "direction": "stable", "seasonality": null,
      "spikes": [{ "date": "2026-04-12T00:00:00.000Z", "value": 100, "score": 4.1 }],
      "breakoutQueries": ["best pickleball paddles 2026"]
    }
  },
  "aiInsight": {
    "summary": "U.S. interest in pickleball peaked in mid-April 2026 and has eased since; the latest reading is 42/100.",
    "trendExplanation": "Interest climbs every spring with outdoor play and stays high into June…",
    "spikeExplanations": [{ "date": "2026-04-12", "explanation": "Likely the start of the outdoor season…" }],
    "contentIdeas": [{ "title": "Best Pickleball Paddles of 2026", "angle": "Compare by skill level…", "format": "Buyer's guide" }],
    "keywordOpportunities": ["pickleball paddle", "indoor pickleball near me"],
    "outlook": "Expect the next rise from late March; publish spring content in February.",
    "model": "gpt-6-luna"
  },
  "warnings": [],
  "attempts": 1,
  "exploreUrl": "https://trends.google.com/trends/explore?q=pickleball&date=today+12-m&geo=US",
  "error": null
}

The key-value store also holds a run report: REPORT (JSON) and REPORT_MD (Markdown). It lists rising and falling keywords, the biggest spikes, breakout queries, the AI summaries and any failures.

Input

Field	Description
keywords	One full report per keyword (up to 500)
compareGroups	Compare 2–5 keywords on the same 0–100 scale, e.g. "chatgpt, gemini, claude"
geo	US, ID, GB… or a sub-region like US-CA. Empty = worldwide
timeRange / customTimeRange	Past hour → 2004-present, or "2025-01-01 2025-12-31"
category, property	Google Trends category ID; web, image, news, Shopping or YouTube search
includeInterestByRegion, includeRelatedQueries, includeRelatedTopics	Choose the data you need
trendingNowGeos	Also fetch today's trending searches for these countries
aiInsights, insightLanguage, contentGoal	AI analysis on or off, its language (any, e.g. "Indonesian"), and the type of content ideas

Pricing (pay per event)

Event	Price
Keyword report (interest over time + region + related queries + related topics)	$0.004
Comparison report (2–5 keywords on one scale)	$0.008
AI insight (per report)	$0.02
Trending now row	$0.0005
Actor start	$0.001

• 100 keywords with full data = $0.40. Add AI insights for $2.00 more.
• Not charged:
  • keywords that failed after all retries
  • keywords Google has no data for
  • AI insights that could not be generated
• Platform usage (compute and proxy) is included in the price.
• Set Maximum cost per run in the run options. The Actor never starts work it cannot charge for within that limit.

Good to know

• The data is relative. Google Trends values go from 0 to 100, where 100 is the peak for that query, place and time. They are not search volumes. Use comparisons to put keywords on the same scale.
• Sampling. Google Trends is computed from a sample of searches, so two requests for the same query can differ by a few points. This is normal and happens on the website too.
• Related topics. Google currently returns "not enough data" for related topics on most queries, including on the Google Trends website. They are included whenever Google provides them, and a warning tells you when it doesn't. Comparisons never have related topics.
• Rising queries sometimes contain unrelated searches; this is how Google reports them. The AI insight is told to ignore clearly unrelated ones.
• AI insights explain spikes from the data, related queries and well-known events. Treat these explanations as hypotheses ("likely"), not confirmed causes.
• This Actor uses publicly available Google Trends data and is not affiliated with Google.

FAQ

How reliable is it? Every request that is rate limited or blocked is retried from a different IP, with increasing waits, and escalates to residential IPs when datacenter IPs are blocked. Before release we measured 0 failed reports out of 137 across 17 runs on the Apify platform (see the changelog).

Can I run it on a schedule? Yes. Use Apify Schedules (for example weekly) and connect the dataset to Google Sheets, Make, Zapier or your own API.

Can AI agents use it? Yes. This Actor is enabled for x402 agentic payments. An AI agent can run it through the Apify API or MCP server and pay per event in USDC on Base, without an Apify account or API key. See the Apify x402 docs.

More Actors by the same developer

• AI Visibility Tracker: does ChatGPT search, Grok and DeepSeek mention and cite your brand?
• Government Tenders & AI Bid Matcher: open tenders from EU TED, UK, SAM.gov, Canada, Australia, World Bank and UN, scored by AI for your company.