Google Trends Scraper & API: Interest, Related, Trending
Pricing
from $1.50 / 1,000 results
Reliable Google Trends data: interest over time, interest by region, related queries and topics, and Trending Now. Pay-per-event, no browser, partial results still succeed.
Keywords or topics to analyze. Each item is one request. Turn on Compare comma-separated terms to put up to 5 keywords on one shared 0–100 chart. Required unless you pass startUrls, a public spreadsheetId, or enable Trending Now.
[ "bitcoin"]When enabled, commas inside one search-terms line split it into a comparison group of up to 5 keywords (same as adding terms on trends.google.com). Extra keywords are batched into further groups of 5.
Paste explore URLs from trends.google.com. Each URL's q, geo, date, cat, and gprop values override the form fields for that job.
Optional. Load terms from a public sheet (first column, row 1 treated as a header). Accepted for drop-in compatibility with other Trends actors.
Predefined Google Trends period. Ignored when a custom range is set. Empty means the past 12 months.
Overrides Time range. Format: YYYY-MM-DD YYYY-MM-DD, for example 2024-01-01 2024-12-31.
Country or region code: US, GB, DE, US-CA, or empty for worldwide. Maps to interestBy / interestBySubregion / interestByCity depending on the resolution.
Google Trends category ID. Empty or 0 means all categories.
Which Google surface to measure, same as the Web / News / Images / YouTube / Shopping switch on Trends.
Which widgets to fetch. Leave the default for a complete explore pull. Trending Now does not need search terms.
[ "TIMESERIES", "GEO", "RELATED_QUERIES"]Granularity for interest by region. Auto uses COUNTRY worldwide and REGION inside a country.
Ask Google to include low-search-volume geos in interest-by-region results.
perDataType (default): one dataset row per term per widget. combined: one row per term in the familiar nested shape. flat: CSV-friendly one row per time point, region, or related item (billed per row).
Google tags automated sessions USER_TYPE_SCRAPER and mixes unrelated decoy queries (for example 'pet care tips' or 'how to cook pasta') into the rising list of every keyword. strict (default): move known decoys, cross-keyword repeats, and rising queries that share no word with the keyword or its top queries into suspectedDecoys. balanced: move only known decoys and repeats; keep unrelated entries but mark them unverified: true. off: return Google's raw rising list. Nothing is deleted in perDataType or combined mode; flat mode omits decoy rows, so they are not billed.
Language for Google's formatted labels, for example en-US or de. Also accepted as viewedFrom for drop-in compatibility.
Pull the current Trending Now list for a country. You can run this with no search terms. Uses the public batchexecute RPC, with the RSS feed as fallback.
ISO country code for Trending Now. Defaults to Geo, or US.
4, 24, 48, or 168 (7 days).
Cap Trending Now rows. 0 means all items Google returned.
Optional numeric category IDs to keep. Filtering is applied client-side; Google's RPC does not filter server-side.
Stop after this many search-term groups (URLs included). 0 means no limit.
How many keyword groups to process at once. Keep at 1 for the highest success rate and lowest 429 risk.
Retries per HTTP call on 429/5xx, with exponential backoff, jitter, cookie re-warmup, and proxy rotation.
Minimum pacing between Google Trends calls. 350 ms is a good default.
Apify Proxy. Default is datacenter (SHADER). The GOOGLE_SERP group only serves www.google.com Search/Shopping and cannot reach trends.google.com — if you select it, the Actor falls back to datacenter, then to direct egress.
Accepted for drop-in compatibility. This Actor never opens a browser, so there are no screenshots.
Accepted for drop-in compatibility. This Actor loads no pages.