Google Trends Scraper — compare any number of keywords on one 0-100 scale

Google Trends compares at most 5 keywords at a time and rescales every group of 5 separately, so a "100" in one group is not a "100" in another. This Actor fixes that: it repeats an anchor keyword in every group, rescales, and returns a single valueCommonScale column where all your keywords are directly comparable. Compare 8, 20 or 50 keywords on one chart.

Pay per result: $0.0015 (Free) / $0.0008 (Bronze) / $0.0005 (Silver) / $0.0003 (Gold and up). No start fee. A run that finds nothing costs nothing.

What you get

• Interest over time with value (Google's own 0-100 per group) and valueCommonScale (one scale for the whole run)
• Interest by region (countries worldwide, or sub-regions inside the geo country you set)
• Related queries, top and rising
• Flat JSON / CSV / Excel, ISO-8601 timestamps, query context in every row
• isPartial flag on the unfinished last data point, so your chart does not show a fake drop
• Change monitor (onlyNew): schedule the run, get only new data points

Quick start: compare 8 AI tools on one scale

{
  "mode": "interest_over_time",
  "keywords": ["chatgpt", "gemini", "claude", "copilot", "perplexity", "grok", "deepseek", "mistral"],
  "timeframe": "today 12-m",
  "geo": "US",
  "commonScale": true,
  "anchorKeyword": "chatgpt"
}

Output (one row per keyword per week):

{ "keyword": "claude", "date": "2026-09-27T00:00:00+00:00", "value": 61, "valueCommonScale": 18,
  "scaleAnchor": "chatgpt", "isPartial": false, "chunkGroup": 1, "geo": "US", "timeframe": "today 12-m" }

value is what Google shows for that group of 5. valueCommonScale is what you chart. (Example values illustrative; run it to see real data.)

Why a common scale matters

	Plain Google Trends / other scrapers	This Actor
5 keywords	one scale	one scale
8 keywords	2 groups, two scales, not comparable	one scale
50 keywords	10 groups, 10 scales	one scale

How it works: every group contains the anchor, so each group can be rescaled to the anchor's level in a reference group. Choose an anchor that is steady and not tiny compared with your other keywords. Google returns whole numbers, so a keyword at about 1/100 of the anchor loses precision. If your keywords differ by orders of magnitude, run them in two batches with a stronger anchor.

Honest status

• No data (Google has too little volume for a keyword): the run says so, and the keyword returns nothing and costs nothing.
• Blocked or throttled by Google: retries with a fresh residential session; if it still fails, the run ends with a partial status and the message says what failed. We never report a blocked run as a successful empty one.
• If Google throttles a multi-keyword comparison, the Actor fetches keywords in smaller groups and rebuilds the common scale where it can; rows that could not be rescaled are marked, not guessed.

Input

Field	What it does
mode	interest_over_time, interest_by_region, related_queries
keywords	up to 200 terms
commonScale	on by default for interest_over_time; output valueCommonScale
anchorKeyword	optional; default is your first keyword
timeframe	today 12-m, today 5-y, now 7-d, or YYYY-MM-DD YYYY-MM-DD
geo	US, PL, … empty = worldwide
category	Google Trends category id, 0 = all
maxItems	hard cap on records (and on your cost)
onlyNew	emit only data points not seen in earlier runs with the same input
proxyConfiguration	residential proxy strongly recommended

Cost examples

• 3 keywords x 52 weeks = 156 records = about $0.23 on Free, $0.05 on Gold.
• 8 keywords x 52 weeks = 416 records = about $0.62 on Free, $0.12 on Gold.
• Set maxItems to cap spending.

Use cases

• Market share of attention: brands, products, tools on one chart
• SEO and content planning: which topic is rising relative to your main keyword
• Weekly alerts with onlyNew + schedule
• Feeding an AI agent or BI dashboard with comparable numbers

FAQ

Can I compare more than 5 keywords? Yes. That is the point of commonScale. Are the numbers equal to the Google Trends website? value is Google's own number per group. valueCommonScale is derived and rounded to whole numbers. Why do values differ between runs? Google samples its data; small differences are normal. Do I need a proxy? Residential Apify proxy is strongly recommended; Google blocks datacenter IPs. Does it need a login or API key? No. Public, aggregated, anonymous data. Is it legal? The data is public and aggregated. You are responsible for using it in line with Google's terms and the law. Export formats? JSON, CSV, Excel, API, webhooks, schedules. Search volume numbers? Google Trends does not publish absolute volumes; indices only.

Integrations

Run via the Apify API, scheduler, n8n, Make, Zapier or MCP. Output has a defined schema.

If it worked for you

A short review on the Actor page helps others find it. If something went wrong, open an issue; we answer.

Unofficial. Not affiliated with Google LLC. "Google Trends" is a trademark of its owner, used only to describe what the Actor reads.

---END README---