
Google Trends Scraper
Interest over time, regions, related queries and Trending now
This Google Trends scraper turns the charts of Google Trends into rows you can sort, plot and compare: interest over time, interest by region and city, related queries, and the Trending now list of any country. Copying a chart from the website is fine for two or three terms. For a list of 200 product names, or a brand you check every morning, you want a table, and that is what each run gives you.
Try it on Apify Store See examples
- Countries
- Output
- JSON · CSV · Excel · API
- Monitoring
- New and changed only
- Updated
tsutseisoftware/google-trends-scraperdataset · item 1
| rowType | "term" |
|---|---|
| term | "bitcoin" |
| termType | "searchTerm" |
| geo | "US" |
| timeRange | "today 12-m" |
| startDate | "2025-11-12" |
| endDate | "2026-11-12" |
| resolution | "week" |
One row per term report or trending search. Download as JSON, CSV or Excel, or read it through the API.
Overview
What you get from Google Trends
Every search term becomes one report for a location and a time range. It holds the full time series, the average, the peak and its date, the latest value, and the change of the last four weeks against the four weeks before, in percent. Regions come at the level you pick: countries, states, US metro areas or cities with coordinates. Related queries come as top and rising, with breakout queries marked; related topics are filled in when Google returns them, which at the moment it usually does not. Web search, images, news, YouTube and Google Shopping all work, with categories, a language, a time zone and any time range you set.
Keep in mind that the numbers are relative. In Google Trends 100 is the highest point of one chart and 50 is half of it; they are not counts of searches. So every row says which scale it is on (its own, a comparison, or the common scale), and you can tell at a glance which values may sit side by side. A term with too little search volume comes back marked hasData: false, so missing data is easy to tell from low interest.
Terms can be typed one per line, read from a Google Sheet or a CSV link, or taken from Google Trends URLs you already have. The output is nested JSON for programs, or a flat format with one CSV row per data point for Excel.
In detail
More than 5 terms on one scale, daily data over years
The website compares five terms at most. The scraper puts any number of them on one common scale, so twelve phone brands or forty product names can be ranked in a single table. Each term gets an error bound in calibrationErrorPercent, usually between 1 and 4 %, because Google rounds every value to a whole number.
Google also switches to weekly points, and later to monthly ones, once a range is longer than about nine months. Turn on daily data and you get one day per point over 12 months, 5 years or since 2004, rebuilt on one scale with its own error bound. That is the series you need for seasonality and forecasting.
Trending now returns the list of a country for the last 4, 24 or 48 hours or 7 days, often a few hundred trends, with search volume ("500K+"), growth in percent, start and end time, categories, related searches and news for the newest ones. For scheduled runs, keyword watch saves a term only when its interest moved by your threshold or a new rising query appeared, and the new-trends mode saves only trends you have not seen. Alerts go to Slack, Discord, Telegram or a webhook. A term that cannot be read is not charged, and the status message names it.
Features
Why use this scraper
Everything Google Trends shows, in one run
time series, regions with resolution, related queries and topics, all search types (web, images, news, YouTube, Google Shopping), categories, any time range down to the minute, language and time zone.
More than Google shows
more than 5 terms on one scale (calibrated across batches, with an error bound per term), daily data over 12 months, 5 years or since 2004 (stitched windows on one scale), change of interest in percent, Trending now with up to ~450 trends per country.
Reliable and polite
Google's cookie handled, a steady pace (30 requests a minute), a refused request is asked once more after a pause from the same session; a term that cannot be read is never charged and the status says which ones. The run fails only when nothing could be read.
Excel-friendly
nested JSON for programs, or a flat output with one CSV row per data point.
Who uses it
Who uses the Google Trends scraper
SEO and content teams
Find rising queries around a topic and check which wording people search more, even for twenty variants on one scale.
Brand and marketing managers
Compare their brand with competitors week by week and get a Slack message when interest jumps by half.
Analysts and forecasters
Pull daily interest over five years for seasonality models, as a flat CSV with one row per day.
Retail and e-commerce
See which states, metro areas or cities search a product most before they plan stock or ad budgets.
Newsrooms and social media teams
Read the Trending now lists of several countries each morning, with volume, growth and the news behind each trend.
Output fields
What data can you get
Term report (one row per term and location; rowType: "term"):
| Field | Example |
|---|---|
term, geo, timeRange, category, property | bitcoin, US, today 12-m, 0, web |
interestOverTime | [{"date": "2026-11-08", "time": "2026-11-08T00:00:00Z", "value": 18, "isPartial": true, "hasData": true}, …] |
averageInterest, peakInterest, peakDate, latestInterest | 33, 100, 2026-02-01, 22 |
recentAverage, previousAverage, changePercent | last 4 weeks against the 4 before: 24.5, 30.25, -19 |
interestByRegion, topRegion, regionResolution | [{"geoCode": "US-WY", "geoName": "Wyoming", "value": 100, …}], Wyoming, region |
comparedRegionShare | comparisons: each term's share of the searches per region (%) |
relatedQueriesTop, relatedQueriesRising | [{"rank": 1, "query": "bitcoin price", "value": 100}], rising with isBreakout |
relatedTopicsTop, relatedTopicsRising | topics with topicId, title, type |
scale, comparisonId, comparedWith | own, comparison or calibrated |
scaleFactor, anchorTerm, calibrationErrorPercent | common scale of more than 5 terms |
isStitched, stitchWindows, stitchErrorPercent | daily data over long ranges |
hasData, failedDataTypes, isComplete | false = too little search volume |
url | the same chart on trends.google.com |
Trending search (rowType: "trending"): title, geo, rank, searchVolume (500000), searchVolumeText (500K+), increasePercent (1000), startedAt, endedAt, isActive, trendCategories (Technology, Sports…), relatedSearches (the trend breakdown), newsArticleCount, news (title, url, source, imageUrl), imageUrl, url.
Every row has id (stable), source, url and scrapedAt; unknown values are null, empty lists [], dates ISO 8601.
Step by step
How to use it
- Enter Search terms (one per line), or a Google Sheet / CSV URL, or Google Trends URLs, or countries for Trending now.
- Pick the Location, Time range and Data to read; optionally a comparison mode, category, search type or daily data.
- Click Start. Each term takes about 8 seconds (4 requests at a polite pace); 100 terms take about 15 minutes.
- Download the results as JSON, CSV or Excel, or use the API. With Output format: flat, the Output tab also links CSV files with one row per data point.
- For monitoring, schedule the run daily with keyword watch or only new trends, and add a Slack, Discord, Telegram or WhatsApp notification.
Examples
Ready-made inputs
Copy an input into the JSON tab of the actor in Apify Console and run it, or change it to your needs.
01Compare products or brands on one scale (more than 5)
InputJSON
{
"searchTerms": ["iphone", "samsung galaxy", "google pixel", "xiaomi", "oneplus", "motorola", "nothing phone"],
"compareMode": "all",
"geo": ["US"],
"timeRange": "today 12-m",
"dataTypes": ["interestOverTime"]
}
02Daily search interest over 5 years (seasonality, forecasting)
InputJSON
{
"searchTerms": ["air conditioner"],
"geo": ["US"],
"timeRange": "today 5-y",
"dailyData": true,
"dataTypes": ["interestOverTime"],
"outputFormat": "flat"
}
03Where is a topic searched most: cities and metro areas
InputJSON
{
"searchTerms": ["pickleball"],
"geo": ["US"],
"timeRange": "today 12-m",
"dataTypes": ["interestByRegion"],
"regionResolution": "DMA"
}
04Trending now in several countries, with news
InputJSON
{
"trendingCountries": ["US", "GB", "DE"],
"trendingHours": "24",
"trendingSortBy": "searchVolume",
"maxTrendsPerCountry": 25
}
05Daily alert on new trends or keyword jumps, to Slack
InputJSON
{
"searchTerms": ["your brand", "competitor brand"],
"geo": ["US"],
"timeRange": "today 3-m",
"watchKeywords": true,
"alertChangePercent": 50,
"alertOnRisingQueries": "any",
"slackWebhookUrl": "https://hooks.slack.com/services/..."
}
06YouTube search interest from a Google Trends URL
InputJSON
{
"startUrls": ["https://trends.google.com/trends/explore?q=minecraft,roblox&geo=DE&date=now%207-d&gprop=youtube"]
}
Output
Output example
Dataset itemJSON
{
"id": "trends:bitcoin:us:today_12-m:0:web",
"rowType": "term",
"source": "google-trends",
"term": "bitcoin",
"termType": "searchTerm",
"geo": "US",
"timeRange": "today 12-m",
"startDate": "2025-11-12",
"endDate": "2026-11-12",
"resolution": "week",
"category": 0,
"categoryName": "All categories",
"property": "web",
"scale": "own",
"hasData": true,
"interestOverTime": [
{"date": "2025-11-09", "time": "2025-11-09T00:00:00Z", "value": 40, "isPartial": false, "hasData": true},
{"date": "2026-11-08", "time": "2026-11-08T00:00:00Z", "value": 18, "isPartial": true, "hasData": true}
],
"interestOverTimePoints": 53,
"averageInterest": 32.66,
"peakInterest": 100,
"peakDate": "2026-02-01",
"latestInterest": 17,
"latestDate": "2026-11-01",
"recentAverage": 16.0,
"previousAverage": 18.75,
"changePercent": -14.7,
"interestByRegion": [{"geoCode": "US-WY", "geoName": "Wyoming", "value": 100, "hasData": true, "latitude": null, "longitude": null}],
"regionResolution": "region",
"topRegion": "Wyoming",
"relatedQueriesTop": [{"rank": 1, "query": "bitcoin price", "value": 100, "formattedValue": "100", "url": "https://trends.google.com/trends/explore?q=bitcoin+price&date=today+12-m&geo=US"}],
"relatedQueriesRising": [{"rank": 1, "query": "spcx stock", "value": 17350, "formattedValue": "Breakout", "isBreakout": true, "url": "https://trends.google.com/trends/explore?q=spcx+stock&date=today+12-m&geo=US"}],
"failedDataTypes": [],
"isComplete": true,
"url": "https://trends.google.com/trends/explore?q=bitcoin&date=today%2012-m&geo=US",
"scrapedAt": "2026-11-12T13:30:00Z"
}
API
For AI agents and API users
Each row is flat at the top level with stable ids; nested lists hold the data points. For tables use outputFormat: "flat" and read interest-over-time.csv, interest-by-region.csv and related-queries.csv from the run's key-value store (linked in the Output tab and listed in SUMMARY). SUMMARY also lists every term with its status (ok, noData, refused, failed).
FAQ
Questions about the Google Trends Scraper
Is it legal to scrape Google Trends?
The actor reads only aggregated, anonymous statistics that Google Trends shows to every visitor without logging in, at a polite pace, and it never solves CAPTCHAs or other access checks. Google's terms of service may restrict automated access and reuse of the data, so you are responsible for how you use it; if in doubt, consult a lawyer. This actor is not affiliated with, endorsed by or connected to Google.
How is it different from Apify's own Google Trends Scraper?
On top of the time series, regions, related queries and topics, categories, custom ranges, start URLs and Google Sheet input that both have, this actor adds: more than 5 terms on one common scale, daily data over long ranges, the Trending now list with volume, growth and news, all search types (images, news, YouTube, Shopping), region resolution down to cities, multiple locations per term, flat CSV output, keyword watch with alerts, new-trend monitoring and notifications, and a clear status of what could not be read (not charged).
Why are the numbers not search volumes?
Google Trends publishes only relative interest (0-100). The only absolute numbers are the rounded search volumes of trending searches ("200K+").
Can I compare two separate reports?
Only reports on the same scale: the terms of one comparison (comparisonId) or of one common scale (scale: "calibrated"). Each term on its own has its own 100.
Do I need a proxy?
Usually not: Google Trends answers at the actor's pace without one. If a run reports refused terms, try again later or use Apify Proxy.
Can I use it via API, Make, Zapier or n8n?
Yes, like any Apify actor: run it through the Apify API or an integration, schedule it, and read the dataset in JSON, CSV or Excel. Notifications can also post to any webhook.
Changelog
What changed in the Google Trends Scraper
- 2026-11-16 · New: Notifications can also go to WhatsApp (WhatsApp Cloud API of your Meta business account), to up to 5 numbers and with your approved message template so they arrive at any time; wrong notification settings now stop the run at its start, before anything is read.
- 2026-11-16 · Fixed: A URL, code or search the website rejected or did not answer is reported as not read (try again later), never as not found or as a page that does not exist.
- 2026-11-15 · New: Send error reports in the Privacy section: turn it off and the actor sends nothing to its developer when a run fails.
- 2026-11-15 · Fixed: Discord and Slack webhooks are accepted in every form you can copy them in, including Discord beta clients and links in angle brackets; a channel link is refused with advice where to copy the webhook.
Support
Missing a field or a filter?
We read every message and usually answer the same day. Small changes to an actor are free; a custom scraper gets a fixed quote.
Apify Store
Full documentation, free trial and pricing. Open an issue on the actor's page.
Open on Apify StoreFeedback form
A short note about this actor, no account needed.
Send feedbackCustom scrapers
Another website, field or format: write what you need and how often.
info@tsutsei-software.com