Google Trends Scraper & API (that actually works)

Get Google Trends data as clean JSON, CSV or Excel: interest over time, interest by region (country, state, metro, city), related queries (top and rising) and related topics for any keyword, comparison or pasted Google Trends link. Plus Trending now: what people are searching for right now in any country, with search volume, news articles and pictures.

It uses Google Trends' own data endpoints over plain HTTP (no browser), so runs are fast and cheap, and it was built around one goal: finish every run with data or a clear reason why not. No endless loops, no "succeeded with 0 results", and you are never charged for a result that failed.

What can this Google Trends scraper do?

• Interest over time: the 0-100 timeline (dates for daily/weekly data, date-times in your time zone for hourly and minute data), the partial last period flagged, plus average, peak and latest values.
• Interest by subregion: countries, regions / states, US metro areas (DMA) or cities (with coordinates). For comparisons you also get each term's share per location.
• Related queries: top and rising searches (including "Breakout"), with links.
• Related topics: top and rising topics, when Google provides them.
• Comparisons: up to 5 terms per comparison, as many comparisons as you like in one run.
• Paste Google Trends URLs: explore links keep their own terms, location, time range, category and search type.
• Trending now: the current trending searches for any country (or region such as US-CA), with approximate searches, when it started trending, related news and picture.
• Every location, time range (past hour to 2004-present, or custom dates), category, search type (web, image, news, YouTube, Google Shopping), time zone and language the Trends website offers.

Why this Google Trends API is reliable

Most Google Trends scrapers break the same way: Google answers 429 Too Many Requests, the scraper retries the same thing forever or finishes "successfully" with an empty dataset. This Actor is designed for exactly that:

• Cookie warm-up: every session first gets the cookie Google requires; without it the Trends API answers 429 to everyone.
• Same data flow as the website: explore request -> widget tokens -> widget data, the same calls the Trends website makes. Expired tokens are refreshed automatically.
• Retries with exponential backoff and jitter, and a fresh session and IP on every rate limit.
• Global adaptive rate limiter: slows down when Google pushes back, speeds up again when it doesn't.
• Automatic proxies: Apify datacenter proxies first, then an automatic switch to residential IPs if Google keeps blocking. You pay the same price per result either way.
• Circuit breaker and hard deadline: if Google blocks everything, the run stops early with a clear message instead of burning time; it also stops cleanly before the run timeout and keeps what it already scraped.
• Honest results: "not enough data" is returned as a normal result with hasData: false; anything that failed is listed in the run's status message and in RUN_SUMMARY. A run that gets nothing fails with the reason, it never pretends to succeed.

How to use it

1. Click Try for free / Start.
2. In Search terms, add one comparison per line. coffee, tea compares two terms; bitcoin on its own line is analysed alone.
3. Pick the location, time range and, optionally, a category and search type.
4. Choose what to extract under What to extract and press Start.
5. Download the results as JSON, CSV, Excel or HTML, or get them through the API.

For Trending now, set What to scrape to Trending now and pick the countries.

Input example

{
    "mode": "keywords",
    "searchTerms": ["coffee, tea", "matcha"],
    "geo": "US",
    "timeRange": "today 12-m",
    "category": "71",
    "searchProperty": "web",
    "includeInterestOverTime": true,
    "includeInterestBySubregion": true,
    "subregionResolution": "AUTO",
    "includeRelatedQueries": true,
    "timezone": "America/New_York"
}

Other ways to ask:

{ "exploreUrls": ["https://trends.google.com/trends/explore?date=now%207-d&geo=GB&q=bitcoin,ethereum"] }

{ "mode": "trendingNow", "trendingCountries": ["US", "GB", "JP"] }

Field	What it does
mode	keywords (default), trendingNow or both.
searchTerms	One comparison per line, up to 5 comma-separated terms. Topic IDs like /m/0dl9sd work too.
exploreUrls	Google Trends explore links (and trending links, which add a country to Trending now).
geo	US, US-CA, US-CA-807 (metro), GB-SCT, ... or WORLDWIDE.
timeRange	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, custom (+ startDate / endDate), or a raw range like 2024-01-01 2024-06-30.
category	Google Trends category id, e.g. 71 Food & Drink (0 = all).
searchProperty	web, images, news, youtube, froogle (Google Shopping).
subregionResolution	AUTO, COUNTRY, REGION, DMA (US metro areas), CITY.
includeLowSearchVolumeRegions	Also list locations with too little data.
includeRelatedTopics	Off by default (see FAQ).
trendingCountries	Countries / regions for Trending now.
timezone, language	Time zone for hourly and minute data and for Google's "today"; names in your language (en-US, de, ja, ...).
proxyConfiguration, residentialFallback, maxConcurrency, maxRequestRetries	Advanced; the defaults work.

Output

You get one result per search term per comparison (and one per trending search). Results are shown in handy views in the Output tab: Overview, Interest over time (one row per date), Interest by subregion (one row per location), Related queries, Related topics and Trending now.

Keyword result (shortened)

{
    "resultType": "keyword",
    "keyword": "coffee",
    "comparisonGroup": ["coffee", "tea"],
    "groupIndex": 1,
    "geo": "US",
    "geoName": "United States",
    "timeRange": "today 12-m",
    "timeRangeLabel": "Past 12 months",
    "resolvedTimeRange": "2025-09-27 2026-09-27",
    "category": 0,
    "categoryName": "All categories",
    "searchProperty": "web",
    "timezone": "UTC",
    "language": "en-US",
    "exploreUrl": "https://trends.google.com/trends/explore?date=today%2012-m&geo=US&q=coffee%2Ctea&hl=en-US",
    "hasData": true,
    "averageInterest": 78,
    "peakInterest": 100,
    "peakDate": "2026-04-12",
    "latestInterest": 76,
    "timelineResolution": "WEEK",
    "timeline": [
        { "date": "2025-09-21", "timestamp": 1758412800, "formattedTime": "Sep 21 – 27, 2025", "value": 65, "hasData": true, "isPartial": false },
        { "date": "2026-09-27", "timestamp": 1790467200, "formattedTime": "Sep 27 – Oct 3, 2026", "value": 82, "hasData": true, "isPartial": true }
    ],
    "geoResolution": "REGION",
    "geoMap": [
        { "geoCode": "US-WY", "geoName": "Wyoming", "value": 100, "hasData": true },
        { "geoCode": "US-HI", "geoName": "Hawaii", "value": 50, "hasData": true }
    ],
    "geoMapCompared": [
        { "geoCode": "US-MT", "geoName": "Montana", "share": 74 }
    ],
    "relatedQueries": {
        "top": [
            { "query": "coffee near me", "value": 100, "formattedValue": "100", "isBreakout": false, "link": "https://trends.google.com/trends/explore?q=coffee+near+me&date=today+12-m&geo=US" }
        ],
        "rising": [
            { "query": "how to brew pour over coffee", "value": 3100, "formattedValue": "+3,100%", "isBreakout": false, "link": "https://trends.google.com/trends/explore?q=how+to+brew+pour+over+coffee&date=today+12-m&geo=US" }
        ]
    },
    "notes": [],
    "errors": [],
    "scrapedAt": "2026-09-27T19:38:39.941Z"
}

notes explains things like Not enough search volume or a resolution fallback. errors lists sections that could not be fetched; such results are stored free of charge.

Trending now result

{
    "resultType": "trendingNow",
    "geo": "US",
    "geoName": "United States",
    "rank": 1,
    "title": "cardinals vs 49ers",
    "approxTraffic": "10000+",
    "approxTrafficNumber": 10000,
    "pubDate": "2026-09-27T19:10:00.000Z",
    "picture": "https://encrypted-tbn1.gstatic.com/images?q=tbn:...",
    "pictureSource": "San Francisco 49ers",
    "newsItems": [
        { "title": "Ways to Watch and Listen: Cardinals vs. 49ers | Week 3", "snippet": null, "url": "https://www.49ers.com/news/ways-to-watch-and-listen-cardinals-vs-49ers-week-3-x8427", "picture": "https://encrypted-tbn1.gstatic.com/images?q=tbn:...", "source": "San Francisco 49ers" }
    ],
    "exploreUrl": "https://trends.google.com/trends/explore?q=cardinals%20vs%2049ers&date=now%201-d&geo=US&hl=en-US",
    "feedUrl": "https://trends.google.com/trending/rss?geo=US",
    "scrapedAt": "2026-09-27T19:25:09.067Z"
}

A complete sample dataset is in the Actor's source under assets/sample-output.json.

Pricing

Pay per result, platform usage included:

Event	Price
Keyword result (one term of one comparison, all selected widgets)	$0.004 ($4 per 1,000)
Trending now search	$0.001 ($1 per 1,000)
Actor start	$0.00005 (Apify's standard start fee)

• Compute and proxy costs are included; the price is the same when the Actor has to switch to residential proxies.
• Results with failed sections, and anything beyond your maximum charge per run, are not charged.
• A term with not enough search volume is a real answer from Google and counts as a result.
• Apify subscribers on higher plans may get lower prices; the Pricing tab shows your price.

Examples: tracking 50 keywords weekly is about 200 results a month, $0.80. A daily Trending now check of 20 items for 5 countries is about 3,000 items a month, $3.

Use it through the API, schedules and integrations

Everything in the Console works through the Apify API: start a run and get the dataset items in one call with the run-sync-get-dataset-items endpoint (copy the ready-made URL from the API tab of this Actor). You can also schedule runs (e.g. daily trending searches), send results to Google Sheets, Slack, Make, Zapier or n8n, and call the Actor from AI agents through the Apify MCP server.

from apify_client import ApifyClient

client = ApifyClient("<YOUR_APIFY_TOKEN>")
run = client.actor("<ACTOR_ID>").call(run_input={"searchTerms": ["coffee, tea"], "geo": "US", "timeRange": "today 5-y"})
for item in client.dataset(run["defaultDatasetId"]).iterate_items():
    print(item["keyword"], item["averageInterest"], len(item["timeline"]))

FAQ

What do the numbers mean? Google Trends values are relative, not search counts: 100 is the peak popularity for the selected terms, location and time; 50 is half as popular; 0 means not enough data. In a comparison all terms share one scale.

How do I compare more than 5 terms? Google allows 5 per comparison. Put one "anchor" term in every comparison (e.g. anchor, a, b, c, d and anchor, e, f, g, h) and rescale each group by the anchor's values.

Why does a term have hasData: false? Google has too little search volume for that term, location, time range and filters. The run still succeeds and the result says so in notes.

Why are related topics empty? Google currently returns an empty related-topics list for most searches. The option is off by default so you don't wait for empty data; switch it on if you want to check.

What does "Breakout" mean? A rising query grew by more than 5,000%.

Why do values differ slightly from the Trends website or between runs? Google computes Trends from a sample of searches, so the same request can vary slightly over time. Short time ranges vary most.

Can I get data for a US city or metro area? Yes: set Location to a metro code (e.g. US-CA-807 for San Francisco-Oakland-San Jose) or set Subregion resolution to DMA or CITY. Metro areas come with Google's numeric metro (DMA) code, e.g. 807; cities come with coordinates instead of codes, like on the website.

Do I need proxies? No setup needed: Apify Proxy is used automatically. If you prefer, choose your own proxy groups or proxy URLs under Advanced settings.

What happens when Google blocks requests? The Actor slows down, retries with fresh sessions and IPs, and switches to residential IPs if needed. If Google still refuses, the run stops early, keeps everything it scraped, tells you what failed, and does not charge for failed results.

Is scraping Google Trends legal? The Actor collects aggregated, public, non-personal statistics that Google displays to any visitor. You are responsible for how you use the data and for complying with Google's terms; if unsure, ask a lawyer.

Is this an official Google product? No. This Actor is not affiliated with, endorsed by or sponsored by Google.

Limitations

• Google Trends allows at most 5 terms per comparison and returns at most 25 top and 25 rising related queries per term.
• Minute and hour-level data exist only for recent ranges (past hour to past 7 days); long ranges return weekly or monthly points, as on the website.
• Trending now returns what Google's public feed lists (usually 10-20 searches per country) with approximate search volumes; there is no history.
• Metro areas (DMA) exist for the United States only.
• Related topics are often empty because Google does not provide them for most searches at the moment.

Support

Found a problem or need a field that is missing? Open an issue in the Issues tab with your run link and it will be looked at quickly.

Changelog

• 1.0 (2026-09-27): first release. Keyword analysis (interest over time, by subregion, compared breakdown, related queries and topics), explore URLs, Trending now, automatic proxies with residential fallback, pay per result.