Google Trends Scraper — Reliable & Fast (No Browser)
Pricing
from $2.00 / 1,000 keyword reports
Google Trends interest over time, by region and related queries for hundreds of keywords, comparisons of up to 5 terms, and Trending Now with news. JSON endpoints, auto-retries and residential fallback. Pay only for keywords with data.
One keyword per line, e.g. coffee or iphone 17. Any language works (日本語, Deutsch, عربي…). Each term gets its own report. Duplicates (case-insensitive) are removed automatically and never charged twice. To compare terms on one scale use Compare terms instead.
Each line is one comparison of up to 5 comma-separated terms, e.g. coca cola, pepsi — exactly like the Compare box on Google Trends. All terms share one 0–100 scale. Each term in a group counts as one keyword report.
Paste explore URLs from your browser, e.g. https://trends.google.com/trends/explore?q=bitcoin&geo=US&date=today%203-m. The q, geo, date, cat and gprop parameters are used; other settings below apply to everything else.
A Google Sheet ID or URL (shared as Anyone with the link can view, or a Publish to web link), or any public CSV URL. Search terms are read from the first column (or the column set below). A specific tab is used when the URL contains gid=.
Header name (e.g. keyword) or 1-based column number. Empty = first column.
Skip the first row of the sheet/CSV.
Empty = Worldwide. A 2-letter country code (US, GB, DE), a region (US-CA, GB-ENG) or a US metro area (US-CA-807). Unknown codes are reported as invalid and not charged.
Google picks the granularity: minutes for 1–4 hours, hourly for 1–7 days, daily up to ~9 months, weekly up to 5 years, monthly beyond.
Overrides Time range. YYYY-MM-DD YYYY-MM-DD (e.g. 2024-01-01 2024-12-31), from 2004-01-01 on. Hourly windows within the last 7 days: 2026-10-20T08 2026-10-21T08.
Google Trends category id; 0 = all categories. Examples: 3 Arts & Entertainment, 5 Computers & Electronics, 7 Finance, 71 Food & Drink, 45 Health, 18 Shopping, 20 Sports, 67 Travel. Unknown ids fail fast with a clear message.
Which Google search property to measure.
0–100 values over the time range, with dates in UTC and in your time zone.
Countries (Worldwide), regions/states (country) or metros/cities, 0–100.
Google's default is countries for Worldwide and regions for a country.
Also list regions where Google has too little data (value 0, hasData false). Off = same list as the Google Trends website.
For comparisons, also fetch each term's own region breakdown (one extra request per term).
Top and rising related searches (25 each), with Breakout flags.
Google currently returns an empty related-topics list to all signed-out visitors — including real browsers on home connections — so this is usually empty (relatedTopicsStatus: notProvidedByGoogle). It fills in automatically if Google turns it back on. Never affects status or price.
Country codes (e.g. US, GB, IN) to fetch the live Trending Now list: term, search volume, % increase, start/end time, categories, related searches and up to 3 news articles per trend.
Time window of the Trending Now list.
0 = all (the US list often has 300–600 trends). Charged per trend.
Add up to 3 news articles (title, URL, source, published time, image) to each trend.
One row per search term: everything for a term in one item (like other Google Trends scrapers) — smallest dataset, best for APIs/JSON. Flat: separate rows for each date, region and related query (recordType column) — best for spreadsheets and BI tools.
Also output inputUrlOrTerm, searchTerm, interestOverTime_timelineData, interestBySubregion, relatedQueries_top/_rising, relatedTopics_top/_rising etc. in Google's raw format, so existing integrations keep working. One-row-per-term format only.
Process at most this many queries (0 = no limit). You can also set "Max cost per run" in Run options; the run stops cleanly when it is reached and unprocessed terms are not charged.
Interface language (hl) for formatted dates and Trending Now news, e.g. en-US, de, ja.
Google's convention: minutes west of UTC. 0 = UTC, 480 = US Pacific (PST), 420 = PDT, 300 = US Eastern (EST), -60 = Central Europe (CET), -330 = India. Affects localTime and formattedTime; date is always UTC. Google's data points themselves don't change with the time zone.
Queries processed at the same time. 3–5 is the sweet spot; higher values trigger more Google rate-limiting.
Each retry uses a fresh IP and cookies, with backoff.
Default Apify datacenter proxy is enough for most runs. RESIDENTIAL is the most reliable. GOOGLE_SERP cannot reach Google Trends.
{ "useApifyProxy": true}If Google keeps rate-limiting the datacenter IPs, switch to RESIDENTIAL automatically for the rest of the run (only if your plan includes it).