Google Trends Scraper

Get Google Trends data as clean JSON: interest over time, interest by region (country, state, metro or city), related queries (top and rising, including "Breakout"), related topics, and the Trending now searches for any country. SEO specialists, marketers, analysts and researchers use it to track keyword demand without copying charts by hand. Compare up to 5 keywords on one scale, or run hundreds of keywords one by one, across Web, YouTube, News, Image and Shopping search. You need no Google account or API key.

What you get

One dataset item per term per data type:

dataType	What it contains
interestOverTime	Full timeline: date (ISO, UTC), value 0-100, hasData, isPartial, plus Google's average
interestByRegion	Regions sorted by interest: geoCode, geoName, value 0-100 (and coordinates for cities)
relatedQueries	top and rising lists: query, value, formattedValue (for example +1,350%), isBreakout, link
relatedTopics	Same as related queries, with topic, topicType and topicMid
trendingNow	One item per trending search: rank, title, approxTraffic (for example 2000), startedAt, picture and related news articles

Every explore item also records the term, location, time range, category, property and, for compared terms, the other terms in the comparison.

Use cases

• Keyword research and SEO: find rising queries around a topic before they peak.
• Seasonality planning: pull 5-year timelines to time campaigns, stock and content.
• Regional targeting: see which states, metros or cities search most for a product.
• Brand and competitor tracking: compare up to 5 brands on one 0-100 scale on a schedule.
• News and content desks: collect Trending now searches with their news articles for several countries each hour.

How to use the Google Trends scraper

1. Open the Actor in Apify Console.
2. Choose a mode: Explore search terms or Trending now.
3. In Explore mode, enter search terms and turn on Compare terms if you want them on one scale. Set the location, time range, category and Google property.
4. Pick the outputs you want (interest over time, by region, related queries, related topics) and the region resolution. In Trending now mode, list the country codes.
5. Click Start, then download the results as JSON, CSV, Excel or HTML, or read them through the Apify API.

Input example

{
  "searchTerms": ["bitcoin", "ethereum"],
  "compareTerms": true,
  "geo": "US",
  "timeRange": "today 3-m",
  "property": "web"
}

Time ranges: now 1-H, now 4-H, now 1-d, now 7-d, today 1-m, today 3-m, today 12-m, today 5-y, all, or a custom range such as 2024-01-01 2024-06-30. Properties: web, youtube, news, images, froogle (Shopping). Region resolution: AUTO, COUNTRY, REGION, DMA or CITY.

Trending now: { "mode": "trendingNow", "trendingGeos": ["US", "GB", "DE"] }.

Output example

Interest over time:

{
  "term": "bitcoin",
  "geo": "US",
  "timeRange": "today 3-m",
  "category": 0,
  "property": "web",
  "comparedWith": ["ethereum"],
  "dataType": "interestOverTime",
  "average": 57,
  "points": 93,
  "timeline": [
    { "date": "2026-06-29T00:00:00.000Z", "timestamp": 1782691200, "formattedTime": "Jun 29, 2026", "value": 94, "hasData": true, "isPartial": false }
  ],
  "scrapedAt": "2026-09-29T19:43:25.630Z"
}

Related queries:

{
  "term": "bitcoin",
  "dataType": "relatedQueries",
  "top": [{ "query": "bitcoin price", "value": 100, "formattedValue": "100", "isBreakout": false, "link": "https://trends.google.com/trends/explore?q=bitcoin+price&date=today+3-m&geo=US" }],
  "rising": [{ "query": "british man recovers lost bitcoin", "value": 10750, "formattedValue": "Breakout", "isBreakout": true, "link": "..." }]
}

Trending now:

{ "dataType": "trendingNow", "geo": "DE", "rank": 10, "title": "enteignung", "approxTraffic": 1000, "approxTrafficText": "1000+", "startedAt": "2026-09-29T18:50:00.000Z", "news": [{ "title": "...", "url": "https://www.welt.de/...", "source": "WELT" }] }

Pricing

You pay per result saved, with no monthly rental. One result is a whole data type for one term, so a full 12-month timeline counts once. Empty results aren't charged: if Google has no data for a term, you get a log warning and no dataset item. Retries never charge twice, because the Actor skips results it already saved. Set a maximum spend per run and the Actor stops once it reaches that limit.

FAQ

Is it legal to scrape Google Trends? The Actor collects public, aggregated Google Trends data without logging in, and no personal data. Google's terms govern automated access, so review them for your use case. You're responsible for how you use the data.

Will Google block it? Which proxy should I use? Google Trends answers HTTP 429 "Too Many Requests" quickly, the main reason Trends scrapers fail. The Actor follows the same flow as the open-source pytrends and google-trends-api libraries: it gets an NID cookie from trends.google.com, calls the explore endpoint for widget tokens, then fetches each widget. Each session keeps its own cookie and a sticky proxy IP. On a 429 or a "sorry" page, it retires the session and retries on a new IP and cookie with exponential backoff (up to 60 seconds), up to 10 times by default. Use residential proxy, the default. Google blocks datacenter IPs quickly, and each request is only a few KB.

Can I get absolute search volume? No. Google Trends doesn't publish it.

How do I compare more than 5 terms? Google allows at most 5 per comparison. Put one common anchor term in each group of 5 and rescale the groups against it.

What are the limits?

• Values are relative 0-100 indexes, not search volumes. Values from separate runs, or from terms not compared together, aren't on the same scale.
• Related topics are often empty. In our tests in September 2026, Google's API returned an empty related topics list for every query we tried (web and YouTube, several time ranges), and it doesn't offer related topics at all when terms are compared. The option is off by default.
• Trending now returns about 10 searches per country. It uses Google's public Trending now RSS feed, which lists only the current top searches, not the full list or the historical archive.
• Low-volume terms may have no data. The Actor skips them with a warning.
• Google can still block a run. Retries with new IPs fix almost all 429s, but heavy blocking can make some terms fail. The log names them.
• Google can change these internal endpoints without notice.