Google Trends Scraper API — Fast, Reliable Interest & Related Queries

A fast, reliable Google Trends scraper API that returns clean interest-over-time data and top & rising related queries for any keyword, in any country — via the same unofficial public Google Trends API that powers pytrends. Built specifically to be reliable where the popular official Google Trends actor breaks: the most-installed Trends actor has ~10k users but only ~3.1★ because the endpoint is semi-protected and its runs frequently return errors or empty data. This actor is engineered around that fragility — realistic browser headers, optional Apify Proxy routing, one keyword pulled at a time, and a hard rule that you are never charged for a keyword that failed.

If you need a Google Trends API you can actually schedule and trust — for keyword research, market research or a data pipeline — this is built for exactly that.

Why this actor over the official Google Trends actor?

The official actor is popular but low-rated precisely because it breaks. Here's the honest comparison:

	This Google Trends Scraper API	The popular official Trends actor (~3.1★)
Reliability	Per-keyword isolation, consent-wall detection	Runs frequently error out or return empty
Related queries (top + rising)	✅ Included	Often missing or partial
Charging for failed keywords	❌ Never — failed keywords are free	You can pay for empty results
Speed	Plain fetch, no headless browser	Heavier
Proxy	Apify Proxy on by default	Varies
Output shape	One clean item per keyword	Varies

• Reliability first. Each keyword is queried independently, so one blocked or empty keyword never poisons the rest of your run. Consent/redirect walls are detected and reported instead of silently returning garbage.
• Speed. Plain fetch against Google's JSON API — no headless browser overhead. A keyword resolves in two lightweight requests.
• Related queries included. Not just the interest curve — you also get top and rising related search queries, the part most keyword-research workflows actually need.
• Fair pricing. Pay per keyword successfully returned. Failed keywords are written to a FAILED record and never charged.
• AI-agent ready. Flat input, clean JSON output. Works out of the box via the Apify MCP server for LLM pipelines and agents.

What you get for each keyword

{
  "keyword": "bitcoin",
  "geo": "US",
  "timeRange": "today 12-m",
  "interestOverTime": [
    { "date": "2025-07-06T00:00:00.000Z", "value": 61 },
    { "date": "2025-07-13T00:00:00.000Z", "value": 68 },
    { "date": "2025-07-20T00:00:00.000Z", "value": 100 }
  ],
  "averageInterest": 74.3,
  "peakValue": 100,
  "peakDate": "2025-07-20T00:00:00.000Z",
  "relatedQueries": {
    "top": [
      { "query": "bitcoin price", "value": 100 },
      { "query": "bitcoin etf", "value": 74 }
    ],
    "rising": [
      { "query": "bitcoin halving 2028", "value": 250 }
    ]
  },
  "scrapedAt": "2026-07-02T12:00:00.000Z"
}

One dataset item per keyword (interest curve + summary stats + related queries together), which is the most convenient shape for keyword research and for feeding an LLM.

How to use it

1. Add keywords — one or more search terms (e.g. bitcoin, ai agents, black friday).
2. (Optional) Set a geo — a country code like US, GB, DE, or a sub-region like US-CA. Leave empty for worldwide.
3. Pick a time range — past day, 7 days, month, 3 months, 12 months, or 5 years.
4. (Optional) Toggle related queries — on by default; turn off for a faster interest-only pull.
5. Run — or schedule it (e.g. weekly) to monitor how a keyword's interest and related queries evolve over time.

Use cases

Each of these is a real search buyers make:

• "Google Trends API for keyword research" — find rising related queries to target before they peak, and see seasonal interest for your head terms.
• "Measure demand for a product or brand over time" — quantify interest across regions and time for market research and go-to-market timing.
• "Time content and launches to a keyword's interest curve" — plan articles and campaigns around when demand actually rises.
• "Track a keyword's momentum week over week" — schedule the actor and watch a term's trajectory and its emerging related queries.
• "A reliable pytrends alternative I can schedule" — the same data source as pytrends, but hosted, scheduled, and with failed keywords never charged.

Pricing

Pay per keyword. You are charged once per keyword that returns real data (interest-over-time + related queries in a single pull). Keywords that fail — blocked, empty, or malformed — are recorded in the run's FAILED key-value record and are never charged. No subscription, no minimum.

FAQ

How is this different from the official Google Trends actor?

Same data source, different engineering priorities. The official actor is popular but low-rated because the endpoint is semi-protected and its runs often break. This actor detects consent/redirect blocks, isolates each keyword, routes through Apify Proxy by default, and refuses to charge for keywords that returned nothing.

Why does the official Google Trends actor fail so often?

Google's Trends endpoint is semi-protected: it can serve a consent or redirect wall to the request, and a heavier headless-browser approach hits those walls and returns errors or empty data. This actor is built defensively around that — realistic headers, proxy routing, per-keyword isolation, and honest failure reporting instead of silent garbage.

Is this a pytrends alternative?

Yes. It uses the same public Google Trends JSON endpoints pytrends uses, but hosted on Apify — so you get scheduling, retries, proxy routing and clean JSON output without maintaining a Python script, and you're never charged for keywords that failed.

Where does the data come from?

The public, unofficial Google Trends JSON API (/trends/api/explore → /widgetdata/multiline and /widgetdata/relatedsearches) — the same endpoints pytrends uses. No login, no personal data. See Google Trends.

What do the interest values mean?

Google Trends normalizes search interest to a 0–100 scale within each query, where 100 is the peak for that keyword over the selected time range. averageInterest and peakValue/peakDate are computed from that series.

Why do I sometimes get fewer results than keywords?

A keyword can fail if Google serves a consent/redirect wall to the request IP or if the term has no meaningful search volume. Failed keywords appear in the FAILED record (not charged). Keeping Apify Proxy enabled (the default) improves success rates.

Can I use a proxy?

Yes — Apify Proxy is enabled by default (useApifyProxy: true, no named group, so it works on any account). It routes requests through Apify's IPs to reduce blocking.

Can AI agents use this actor?

Yes. The input schema is flat and agent-friendly, and every Apify actor is consumable via the Apify MCP server, so agents can call it as a tool.

Which geos and time ranges are supported?

Any Trends geo code (country like US or sub-region like US-CA), and time ranges from the past day up to the past 5 years.