Google Trends Scraper — Compare Unlimited Keywords

Get Google Trends data through a simple API, without a browser. One result per keyword with everything in it: interest over time, interest by region and related queries (top & rising), plus Trending Now searches. Compare any number of keywords on one common 0–100 scale.

Fast, reliable and fairly priced: a keyword that Google refuses never crashes your run, you are never charged for empty or failed keywords, and the price per keyword is all-inclusive — no extra platform usage fees.

Why this Actor

• One keyword = one result = one charge. Time series, regions and related queries for a keyword come together in a single dataset item. You don't pay per data point or per data type.
• All-inclusive price. Compute and proxy costs are included. No start fee per run, no platform usage added to your bill.
• Unlimited comparison. Google Trends compares at most 5 terms at once. This Actor chains as many terms as you want through a shared anchor term, so commonScaleAverage and valueOnCommonScale are directly comparable across all of them.
• Fast and light. Pure HTTP calls to the same endpoints the Google Trends website uses — no browser. 7 keywords with all data took 48 seconds in our tests.
• Reliable by design. Short timeouts, automatic retries with a fresh session and a new proxy IP, and per-keyword isolation: one failure produces a free error item, the rest of the run continues.
• Insights, not just raw numbers. Average, peak value and date, latest value, trend direction (rising / stable / falling), top region, top and top rising related query — ready for a spreadsheet.

What you get

Item type	Content	Charged
keyword	All selected data for one keyword: summary columns + interestOverTime (time series), interestByRegion, relatedQueries, relatedTopics	1 result
trendingNow	One of today's trending searches for a country, with approximate traffic and news articles	1 result
noData	The keyword has too little search volume for this location and period	Free
error	The keyword failed after all retries	Free

Input example

{
  "searchTerms": ["python", "javascript", "rust", "golang", "java", "kotlin", "typescript"],
  "geo": "FR",
  "timeRange": "today 12-m",
  "compareOnCommonScale": true,
  "trendingNowCountries": ["FR", "US"]
}

• Location: ISO country code (US, FR), or region (US-CA, FR-J). Leave empty for worldwide.
• Time range: from the past hour to 2004–present, or a custom range such as 2024-01-01 2024-12-31.
• Search type: web, images, news, Google Shopping or YouTube.
• Category: any Google Trends category ID (0 = all).
• Data types: turn off the ones you don't need to make runs faster — the price per keyword stays the same.

Output example

A real result (location France, past 12 months, compared with 6 other languages — python is the anchor):

{
  "type": "keyword",
  "keyword": "kotlin",
  "geo": "FR",
  "timeRange": "today 12-m",
  "average": 60.1,
  "peakValue": 100,
  "peakDate": "2026-03-29T00:00:00Z",
  "latestValue": 29,
  "trendDirection": "falling",
  "trendChangePercent": -46.7,
  "commonScaleAverage": 1.87,
  "topRegion": "Île-de-France",
  "topRelatedQuery": "kotlin android",
  "topRisingQuery": "kotlin 2.3",
  "interestOverTime": {
    "average": 60.1,
    "peakValue": 100,
    "commonScale": {
      "comparedWith": ["python", "javascript", "rust", "golang", "java", "typescript"],
      "anchorKeyword": "python",
      "average": 1.87,
      "peakValue": 3
    },
    "timeline": [
      { "date": "2025-10-12T00:00:00Z", "formattedDate": "Oct 12 – 18, 2025", "value": 65, "valueOnCommonScale": 2, "hasData": true, "isPartial": false }
    ]
  },
  "interestByRegion": {
    "resolution": "REGION",
    "regions": [
      { "geoCode": "FR-J", "geoName": "Île-de-France", "value": 100, "hasData": true },
      { "geoCode": "FR-R", "geoName": "Pays de la Loire", "value": 86, "hasData": true }
    ]
  },
  "relatedQueries": {
    "top": [{ "query": "kotlin android", "value": 100, "formattedValue": "100", "link": "https://trends.google.com/trends/explore?q=kotlin+android&date=today+12-m&geo=FR" }],
    "rising": [{ "query": "kotlin 2.3", "value": 38150, "formattedValue": "Breakout" }]
  },
  "relatedTopics": null
}

average is on the keyword's own 0–100 scale; commonScaleAverage is on the scale shared by all compared keywords (here, kotlin is searched about 37× less than python in France).

Use the Overview view in the dataset to see one line per keyword, or export to JSON, CSV or Excel.

How the common scale works

Google normalizes each request so that its highest point equals 100. The Actor sends the first 5 terms together, picks the most searched one as the anchor, then sends the remaining terms in groups of 4 plus the anchor. The ratio between the anchor's average in the first group and in each later group converts every value to the first group's scale; the result is renormalized so the global maximum is 100.

Very rare terms compared with very popular ones may show values close to 0: Google only returns whole numbers, so precision is limited for small terms.

Pricing

Pay per result: one result = one keyword (with all the data types you selected) or one Trending Now search. noData and error items are free. The price is all-inclusive: no start fee, no platform usage charges. It gets lower on higher Apify plans — see the Pricing tab.

Your maximum cost per run is respected: the Actor stops cleanly when it is reached and keeps everything already collected.

Tips

• Proxy: the default Apify Proxy works well. If you see many retries in the log, select residential IPs in Proxy configuration. Google Trends rate-limits by IP; the Actor rotates to a new IP on every retry.
• Faster runs: turn off the data types you don't need.
• Monitoring: schedule the Actor and use trendDirection / latestValue to get alerted when a keyword starts rising.

FAQ

Is there an official Google Trends API? Google offers a Trends API only to a small number of approved testers. This Actor uses the public endpoints of the Google Trends website.

Is it legal? The Actor collects public, aggregated and anonymized search-interest statistics; it does not collect personal data. You are responsible for using the data in line with Google's terms and the laws that apply to you.

Why does a keyword return noData? Google has too little search volume for that term, location and time range. Try a wider location or a longer period. You are not charged for it.

Values differ slightly from the website? Google Trends works on a sample of searches, so values can vary a little between two requests.