Google Trends Scraper

Interest over time, interest by region, related queries and related topics for any search term, from Google Trends' own data API. Pick the country or region, the time range (presets or your own dates), a category and the Google property (web, images, news, YouTube, Shopping), and compare up to five terms in one chart like on the Trends page. Every term you ask for comes back as a row, and when Google has nothing the row says why (no_data, invalid, blocked) instead of an empty dataset. You pay only for rows that carry data.

No browser, no login, no cookies to paste: it calls the JSON endpoints behind the Trends page, paces itself the way Google tolerates, and rotates IPs before Google's rate limit or captcha wall shows up.

What you get

For each term, one row with:

• interestOverTime: the chart's points (time in ISO 8601 UTC, value 0–100, hasData, isPartial for the still-open last period), plus averageInterest.
• interestByRegion: every country, region, metro area or city with its score, code and, for cities, coordinates. The level is Google's default or the one you choose.
• relatedQueries and relatedTopics: top (0–100) and rising (growth in percent, isBreakout when Google says "Breakout"), each with its Trends link; topics carry Google's Knowledge Graph id.
• What was asked: term, comparedWith (the other terms of the same chart, when comparing), geo, timeRange, category, property, language, and the url of the same query on Google Trends.
• status and reason: ok, no_data (Google has no data for that term there and then), invalid (Google rejected the geo, time range or category; the reason quotes Google), or blocked (every IP route was refused; unpaid).

Input

field	what it does
searchTerms	One term per line. Each is queried on its own unless compareTerms is on.
urls	Explore links copied from Google Trends; each carries its own terms, geo, dates, category and property.
compareTerms	Put the terms in one chart, five per chart: values become relative to the most searched of the group.
timeRange	now 1-H, now 4-H, now 1-d, now 7-d, today 1-m, today 3-m, today 12-m (default), today 5-y, all.
dateFrom, dateTo	A custom range (YYYY-MM-DD) that replaces timeRange.
geo	Country (US) or region (US-CA) code; empty for worldwide.
category	Google Trends category id (0 all).
property	"" web, images, news, youtube, froogle (Shopping).
sections	Which of the four sections to fetch; fewer sections, fewer requests.
regionLevel	auto, country, region, city, dma (US metro areas).
language	Language of region names and topic titles (en, es, pt-BR).
residentialFallback	Retry through residential proxies when Google refuses the datacenter IPs. Default on.

{
    "searchTerms": ["bitcoin", "ethereum", "solana"],
    "compareTerms": true,
    "geo": "US",
    "timeRange": "today 12-m",
    "sections": ["interestOverTime", "relatedQueries"]
}

Output example

{
    "term": "bitcoin",
    "comparedWith": [],
    "geo": "US",
    "timeRange": "today 12-m",
    "category": 0,
    "property": "",
    "language": "en",
    "status": "ok",
    "reason": null,
    "interestOverTime": [
        { "time": "2025-09-14T00:00:00.000Z", "value": 25, "hasData": true, "isPartial": false },
        { "time": "2025-09-21T00:00:00.000Z", "value": 27, "hasData": true, "isPartial": false }
    ],
    "averageInterest": 41.3,
    "interestByRegion": [
        { "geoCode": "US-WY", "geoName": "Wyoming", "value": 100, "hasData": true, "latitude": null, "longitude": null },
        { "geoCode": "US-NV", "geoName": "Nevada", "value": 46, "hasData": true, "latitude": null, "longitude": null }
    ],
    "regionLevel": "region",
    "relatedQueries": {
        "top": [{ "query": "bitcoin price", "value": 100, "formattedValue": "100", "isBreakout": false, "url": "https://trends.google.com/trends/explore?q=bitcoin+price&date=today+12-m&geo=US" }],
        "rising": [{ "query": "bitcoin etf", "value": 1250, "formattedValue": "+1,250%", "isBreakout": false, "url": "https://trends.google.com/trends/explore?q=bitcoin+etf&date=today+12-m&geo=US" }]
    },
    "relatedTopics": {
        "top": [{ "title": "Bitcoin", "type": "Cryptocurrency", "mid": "/m/05p0rrx", "value": 100, "formattedValue": "100", "isBreakout": false, "url": "https://trends.google.com/trends/explore?q=/m/05p0rrx&date=today+12-m&geo=US" }],
        "rising": []
    },
    "url": "https://trends.google.com/trends/explore?q=bitcoin&date=today+12-m&hl=en&geo=US"
}

Why this one

Most Google Trends Actors drive a headless browser through the Trends page, which is why runs take minutes, time out, or end with "no results" when Google shows a captcha. This one:

• Reads the data endpoints directly, with the cookie Google hands out to anyone. A term with all four sections is five small requests, about a second each.
• Stays under Google's limits by design: one request per second per IP, several IPs in parallel, and a fresh IP well before the point where Google starts answering 429. A plain 429 is retried after a pause; a captcha wall retires that IP and moves on. Direct → datacenter proxy → residential proxy, and a route Google keeps refusing is abandoned for the run instead of retried forever.
• Says why a row is empty. no_data is Google's answer, invalid quotes Google's rejection of the parameters, blocked means the run was refused, and none of those is charged.
• Takes the query as you have it: terms, or explore links with everything already in them, custom date ranges, categories, properties, region levels down to cities, and up to five terms compared per chart.

Pricing

Pay per term row with data (status: ok). Rows with any other status are free. A term with all four sections costs about five requests of a few KB; running cost is a fraction of the price even through residential proxies.

Limits and notes

• Values are Google's relative 0–100 scores, never absolute search volumes: Google Trends does not publish those.
• Google returns related topics only for single-term queries; in a comparison relatedTopics is null.
• Google rejects some combinations with HTTP 400 (an unknown region, a custom range before 2004 or after today, a region level the country does not have such as DMA outside the US). Those rows carry status: invalid and Google's message.
• Hourly ranges (now …) are available for the past 7 days at most; daily and weekly points depend on the range, as on the site.
• Google may refuse every IP for a while during heavy use; those rows come back as blocked, unpaid, and a later run gets them.
• This Actor reads public, aggregated data that Google Trends shows to any visitor. No personal data is involved.