Google Trends Unlimited Compare

Compare any number of keywords on one Google Trends scale — not just 5 — and get everything for each keyword in one result.

Google Trends compares at most 5 terms, and every comparison is rescaled to its own 0–100 peak, so numbers from different comparisons can't be put side by side. This Actor splits your list into 5-term batches that all share one anchor keyword, then rescales every batch to that anchor. Every keyword lands on one comparable scale.

Accuracy check: stitched values vs. Google's own 5-term comparison differed by 0.26 points on average (max 0.7 of 100) in our test (US, 12 months).

What you get — one result per keyword

Field	Contents
interestOverTime	timeline on one scale shared by all keywords, average, anchor
relatedQueries	top and rising queries, with Breakout flags
interestBySubregion	states / provinces
interestByCity	cities (with coordinates)
interestByMetro	US metro areas (DMA)

Plus trending_now: today's trending searches per country, with approximate traffic and news links (one result per country).

You pay per keyword, not per data type or per day of the timeline. Pick only what you need in What to get.

Inputs

• Keywords — up to 500 per run, all on one scale
• Google Trends URLs — paste any trends.google.com/trends/explore?... link; its terms, country, time range, category and search type are used as-is
• Country/region, time range (now 7-d … today 5-y, or 2024-01-01 2024-12-31), category, search type (web, YouTube, News, Images, Shopping)

{
  "keywords": ["python", "javascript", "rust", "golang", "kotlin", "swift", "typescript"],
  "data": ["interest_over_time", "related_queries", "interest_by_subregion"],
  "geo": "US",
  "timeframe": "today 12-m",
  "startUrls": [{"url": "https://trends.google.com/trends/explore?date=today%203-m&geo=GB&q=tea,coffee"}],
  "trendingGeos": ["US"]
}

Output example

{
  "type": "keyword",
  "keyword": "rust",
  "geo": "US",
  "timeframe": "today 12-m",
  "interestOverTime": {"anchor": "python", "average": 29.9, "timeline": [{"time": "2025-11-09T00:00:00Z", "value": 31.0}]},
  "relatedQueries": {"top": [{"query": "what is rust", "value": 100}], "rising": [{"query": "...", "formattedValue": "Breakout", "breakout": true}]},
  "interestBySubregion": [{"geoCode": "US-WY", "geoName": "Wyoming", "value": 100}],
  "interestByMetro": [{"geoCode": "759", "geoName": "Cheyenne WY-Scottsbluff NE", "value": 100}]
}

Values are on the scale of the anchor batch: 100 = that batch's peak; terms more popular than it can exceed 100.

Reliability and fair billing

• Requests run in parallel (default 5, up to 10), each with its own proxy session, with automatic back-off on Google rate limits. In one test, 192 keywords × (timeline + related queries + subregions) finished in under 9 minutes, twice at the same time. Google does rate-limit heavy, repeated use from the same proxies; when that happens the run slows down and reports what it could not fetch.
• A keyword with no data at all is never billed.
• If one part could not be fetched after all retries, the result says so in incomplete (e.g. ["relatedQueries"]) instead of silently leaving it out.
• Every warning and error is listed in the run's Issues output (key-value store record ISSUES), saved during the run so it survives an abort.

Tips

• Anchor choice matters. Pick a steady, medium-popularity term related to your list.
• Keep the default Apify Proxy for large runs; Google rate-limits heavy use.

Limits

• Up to 500 keywords per run.
• Related topics are not offered: Google currently returns them empty through its public endpoints.
• Google returns no city data for some countries and time ranges; regions without data are left out.
• Values are relative interest (Google Trends), not absolute search volume.