Google Trends Scraper: Interest, Rising Queries & Trending Now

Get Google Trends data for any search term, as structured rows you can export or call through the API: interest over time, interest by region (state, metro or city), top and rising related queries, comparisons of more than 5 terms on one scale, and the full Trending now list with search volumes. It works like a Google Trends API: no browser, no API key and no proxies to set up, results in JSON, CSV or Excel.

Why this Google Trends scraper

• Built around Google's 429 errors. Google Trends refuses requests that come without its session cookie, and refuses an IP that asks too much. This Actor gets a fresh cookie when Google refuses, and moves to another IP (datacenter, then residential) when Google keeps refusing. In our October 2026 test runs, 199 of 200 term results came back complete (the incomplete one is why requests now get more retries), 150 terms took 100 seconds.
• Summary figures per term: average, peak and peak date, latest value, and the change from the start to the end of the period. No spreadsheet work needed to see whether a term is growing.
• More than 5 terms on one scale. Google compares at most 5 terms at a time. Here you can compare more (up to 500 per run; we tested 12): an anchor term links the requests, and each term gets its share of interest in percent (how it works).
• The full Trending now list: every search trending in a country (644 in the US and 395 in Brazil over 24 hours in our test), not only the 10 of the RSS feed, with search volume, growth, start and end time, categories and related searches.
• Regions at the level you need: countries, states/provinces, US metros (DMA) or cities with coordinates.
• Paste Trends URLs from your browser: an Explore page keeps its own terms, location, time range, category and property.
• No charge for unavailable term data. Terms Google has no data for are saved without a search-term charge. The Actor Start fee still applies per run.
• Run report and quality check, so a change on Google's side shows up as a warning (or a failed run, for schedules), not as silently empty results.

What you can use it for

• SEO and content planning: find rising queries before they peak, pick the wording people search for, time content for seasonal peaks.
• Market and product research: compare brands, products or categories on one scale and see which is gaining share.
• E-commerce: watch Google Shopping and YouTube interest for products; plan stock around seasonality.
• Regional targeting: see in which states, metros or cities a term is most searched, for ads and expansion.
• Newsrooms and social media: the Trending now list every few hours, filtered by category.
• Finance and data science: Google Trends as a time series signal, back to 2004.
• AI agents and dashboards: one clean row per term, ready for an LLM or a BI tool.

Quick start

1. Type your Search terms, one per line: pickleball, padel.
2. Add Locations as country codes (US, BR, GB), regions (US-CA) or leave empty for worldwide. Every term runs in every location.
3. Pick a Time range (past hour to 2004 – present, or custom dates).
4. Run it and export the dataset, or read it through the API.

{
  "searchTerms": ["pickleball", "padel"],
  "geos": ["US", "BR"],
  "timeRange": "past-12-months"
}

What you get

One row per term and location:

Field	Example
searchTerm, geo, timeRange, timeRangeResolved, resolution	padel · BR · today 12-m · 2025-10-25 2026-10-25 · WEEK
averageInterest, peakInterest, peakDate	51.1 · 100 · 2025-11-09
latestInterest, latestDate	45 · 2026-10-18 (the last complete week)
changePercent	1.1 (average of the last quarter of the period against the first quarter)
interestOverTime	[{ "date": "2026-10-18", "value": 45, "hasData": true, "isPartial": false }, …]
interestByRegion, regionResolution	[{ "geoCode": "BR-RS", "geoName": "State of Rio Grande do Sul", "value": 100 }, …] · REGION
relatedQueriesTop	[{ "query": "padel raquete", "value": 100 }, …]
relatedQueriesRising	[{ "query": "padel news", "increasePercent": 3600, "breakout": false, "formattedValue": "+3,600%" }, …]
comparedWith, scale, shareOfInterestPercent	In compare mode: the other terms, the scale used, the term's share
hasData, errors	false when Google has no data; sections that failed, if any
trendsUrl, scrapedAt	The same chart on trends.google.com

And one row per trending search:

Field	Example
trend, geo, rank	vasco da gama x corinthians · BR · 1
searchVolume, searchVolumeText	2000000 · 2M+ (Google's volume bucket)
increasePercent, active, startedAt, endedAt	1000 · true · 2026-10-24T17:20:00Z · null
categories, relatedSearches	[Sports] · ["corinthians x vasco", …]

Values are Google's relative interest, not search counts. 100 is the highest point of the chart for that term, place and time; 50 is half as popular; 0 means too little data. That is how Google Trends works for everyone. The values matched Google's own website on 51 of 54 weekly points we checked; the other 3 differed by 1, because Google samples its data (the same request a few minutes apart can differ slightly) and the last week was still in progress.

Pricing

Pay per result, with platform usage included. The Actor Start event costs $0.00005 per GB of memory, with a minimum of one event per run ($0.00005 at the default 512 MB). The examples below exclude this start event.

Plan	Per term result	Per trending search
Free	$0.002	$0.0005
Bronze	$0.001	$0.0003
Silver	$0.0006	$0.0003
Gold and above	$0.0004	$0.0003

A term result is one term in one location with every section you turned on (interest over time, regions, related queries): the sections do not change the price.

You get	Free plan
100 terms	$0.20
1,000 terms	$2.00
The top 100 trending searches in 5 countries, every day for a month	about $7.50

No search-term charge: terms Google has no data for (saved with hasData: false), and terms that failed. The Actor Start fee still applies. Apify's free plan includes $5 of monthly credit, enough for about 2,500 term results before start fees. Set a maximum cost per run and the run stops by itself when it gets there.

Options

{
  "searchTerms": ["coca cola", "pepsi", "red bull", "celsius"],
  "compareTerms": true,
  "geos": ["US"],
  "timeRange": "custom",
  "startDate": "2024-01-01",
  "endDate": "2026-06-30",
  "category": "Food & Drink",
  "searchProperty": "web",
  "regionResolution": "DMA"
}

• Locations: US, BR, GB (countries), US-CA, BR-SP (regions), US-CA-807 (a US metro), or empty for worldwide.
• Time range: past hour, 4 hours, day, 7 days, 30 days, 90 days, 12 months, 5 years, 2004 – present, or custom dates. Google picks the point size: hourly up to 7 days, daily up to about 9 months, weekly up to 5 years, monthly beyond.
• Category: a name as trends.google.com shows it ("Food & Drink", "Finance") or a category ID (71). Empty = all categories.
• Google property: web search, image search, news search, YouTube search or Google Shopping.
• Sections: turn interest over time, interest by region, related queries and related topics on or off. Fewer sections make runs faster; the price per term stays the same.
• Region level: Google's default (countries for worldwide, states for a country), cities, or US metros (DMA). A level a location doesn't have falls back to Google's default, and regionResolution says which one was used.
• Google Trends URLs: paste trends.google.com/trends/explore?q=… (up to 5 terms, compared as on the website) or trends.google.com/trending?geo=…&hours=…&category=…&status=active.
• Language changes region and topic names (pt-BR, de...), not the data.

Related topics are off by default: in October 2026 Google returned no related topics for any search we tried, on its own website too ("not enough data"). Turn them on if Google brings them back. Google gives them for single terms only, not inside comparisons.

Comparing more than 5 terms

Turn on Compare terms on one scale and all terms share one 0–100 scale, like Google's Compare button, plus a shareOfInterestPercent per term (its part of the total interest of all compared terms).

Google takes at most 5 terms per comparison, each comparison on its own scale. With more terms, the first request takes 5; its most searched term becomes the anchor and is repeated next to 4 new terms in every further request. Each further request is scaled so the anchor matches its first-request values, then everything is rescaled so the top point of any term is 100.

What to know:

• Values in compare mode can have decimals.
• A term far smaller than the anchor gets few distinct values, as on Google's own chart.
• scale says how each term was placed: shared; shared-low-precision when the anchor peaked under 10 in that request (a much bigger term was in it), so the factor rests on few values; own when the anchor had no data in that request and the term kept its own scale.
• Our test: 12 soft-drink and energy-drink brands across 3 requests, shares adding up to 100%.

Trending now

{
  "trendingGeos": ["US", "BR", "JP"],
  "trendingHours": "24",
  "trendingCategories": ["17", "18"],
  "trendingActiveOnly": true,
  "maxTrendsPerGeo": 100
}

• Past 4 hours, 24 hours, 48 hours or 7 days, per country.
• Categories: Sports, Technology, Entertainment, Politics, Business and Finance and 14 more, as on trends.google.com/trending. A trend can have several.
• Only active leaves out trends Google marks as ended.
• Google lists them by relevance; Max trending searches per location keeps the first ones. In our tests Google listed 644 for the US and 395 for Brazil over 24 hours, 117 for Germany over 4 hours and 2,568 for Japan over 7 days.
• searchVolume is Google's bucket (200K+ = 200000), not an exact count.

Trending now and search terms can run in the same run.

Run report and quality check

Each run writes a REPORT to its key-value store: status per term and location (OK, PARTIAL when a section failed, NO_DATA, FAILED, SKIPPED when a run limit was reached), requests made, how many Google refused and which IP route was used.

A quality check flags runs where every term failed, where Google's response no longer has the expected parts, where no term had data, or where Trending now came back empty. With Fail the run on a quality drop, such a run is marked as failed, so a scheduled run e-mails you instead of saving empty rows.

Integrations and API

Run it from the Apify API, the JavaScript or Python client, or schedule it in Apify. Results go to Google Sheets, Make, Zapier, n8n, Slack, webhooks, or anywhere through the API.

curl -X POST "https://api.apify.com/v2/acts/roomy_tweet~google-trends-scraper/run-sync-get-dataset-items?token=YOUR_TOKEN" \
  -H "Content-Type: application/json" \
  -d '{"searchTerms": ["pickleball"], "geos": ["US"]}'

FAQ

Is this the official Google Trends API? No. Google's own Trends API is an alpha for a limited group of testers; this Actor reads the same data the Google Trends website shows, for any term.

Why are the values 0–100 and not search volumes? Google Trends never publishes search counts, only interest relative to the highest point. Trending now has volume buckets (200K+, 2M+).

Why do two runs give slightly different numbers? Google computes Trends from a sample of searches and refreshes it, so values can move by a point or two between requests. Long time ranges and popular terms move least.

A term has hasData: false. Google has too few searches for it in that place and time. Try a broader term, a bigger location or a longer time range. These rows are not charged.

Can I get daily data for a whole year? Google gives daily points for ranges up to about 9 months; longer ranges come weekly. Use a custom range of up to 9 months for daily points.

How many terms per run? Up to 500 search terms and 1,000 term results (terms × locations) per run, plus up to 50 Trending now locations.

Is scraping Google Trends legal? The Actor reads public, aggregated data that Google shows to anyone without login. It contains no personal data. You are responsible for how you use the results.

Something broke or a field is missing. Open an issue on the Actor's Issues tab with the run link; the REPORT usually shows what happened.