Google Trends Compare Scraper (up to 5 keywords)

Google Trends' Compare view as data: 2 to 5 keywords on one scale, relative interest over time for all of them, Google's average per term, the leader and its share, and optionally which term wins in each country or region. Up to 200 comparisons per run, one row per comparison. You only pay for comparisons that come back with data.

Why this scraper

• The comparison, not five separate curves. Trends scales a comparison so that 100 is the strongest term's peak and every other number is relative to it. Fetching terms one by one loses exactly that; this actor keeps it, the way the Compare page does.
• Averages and a leader per row. Google's own average per term is delivered with a computed leader and its share of the total, so a keyword list can be ranked without opening a single chart.
• Who wins where. Switch on includeRegions and each comparison also carries the share of every term per country (worldwide) or per region (inside a country), with the winning term named. In testing on 13-09-2026 a worldwide comparison came back with 64 to 69 countries that had enough volume to score.
• Lightweight route. Two requests per comparison, about 10 KB for a 12-month weekly series of three terms.
• Only delivered comparisons are charged. A row with too little search volume comes back as a clear error record at no cost.

Who this is for

Brand and category managers comparing themselves with competitors. Market researchers sizing which option people search for most, per market. Content and SEO teams choosing between candidate topics. Anyone who wants Compare results for a list, not one at a time.

Who this is not for

Values are Google's relative 0 to 100 index within one comparison. They are not search volumes, and a number from one comparison cannot be compared with a number from another. For a single keyword's curve and related queries, use the Google Trends Batch Scraper from the same publisher; for what is trending right now, the Trending Now Scraper; for one keyword's regional map, the Interest by Region Scraper.

Input example

{
    "comparisons": ["bitcoin, ethereum, solana", "iphone, samsung galaxy, pixel"],
    "geo": "",
    "timeframe": "today 12-m",
    "includeRegions": false,
    "language": "en-US"
}

Separate keywords with commas (or |). Duplicates are dropped; 2 to 5 terms per row.

Output example

{
    "terms": ["bitcoin", "ethereum", "solana"],
    "geo": "worldwide",
    "timeframe": "today 12-m",
    "averages": [
        {"term": "bitcoin", "average": 43},
        {"term": "ethereum", "average": 7},
        {"term": "solana", "average": 6}
    ],
    "leader": "bitcoin",
    "leaderShare": 0.768,
    "pointCount": 54,
    "series": [
        {"time": "Sep 7 – 13, 2025", "timestamp": 1757203200, "values": [42, 12, 11], "isPartial": false},
        {"time": "Sep 14 – 20, 2025", "timestamp": 1757808000, "values": [44, 12, 10], "isPartial": false}
    ],
    "regions": null,
    "status": "ok"
}

With includeRegions on, regions holds one entry per country or region:

{"regionCode": "KR", "regionName": "South Korea", "shares": [{"term": "bitcoin", "share": 30}, {"term": "ethereum", "share": 20}, {"term": "solana", "share": 50}], "leader": "solana"}

A comparison that returns nothing produces an error record instead, and is not charged:

{
    "input": "qzxwv nonsense, zzzz zzzz",
    "status": "error",
    "errorCode": "NO_DATA",
    "error": "Empty time series for ['qzxwv nonsense', 'zzzz zzzz']: too little search volume"
}

Error codes: BAD_QUERY, NO_DATA, BLOCKED. INPUT_TRUNCATED appears once when your input is longer than this actor accepts.

Related actors from steadydata

• google-trends-batch: one keyword at a time, with related queries
• google-trends-regions: the regional split per keyword

Pricing

Pay per event: one comparison-fetched event per comparison delivered with data, whatever the number of terms and points inside it. No charge for rows that fail, no separate platform-usage surcharge.

FAQ

What do the numbers mean? Google's relative index for the whole comparison: 100 is the highest point of the strongest term in the period; everything else is scaled to it. averages are Google's averages per term on that same scale; leaderShare is the leader's average divided by the sum of all averages.

How many points will I get? Google picks the resolution for the timeframe: hourly for a day, daily for a month, weekly for 12 months, monthly for 5 years and longer.

Can I compare more than 5 terms? No; that is Google's limit. Split into several rows that share one anchor term.

Is personal data collected? No. Google Trends publishes aggregate, anonymised search interest.

What happens when Google changes something? Internal routes shift from time to time. The actor is monitored daily and fixed fast, and while it is broken you are not charged, because only delivered comparisons cost anything.