Google Trending Searches — the one that doesn't rate-limit

Get the searches trending on Google right now, in any of 49 countries, with Google's own traffic estimate and the news stories driving each trend.

Why this Actor exists: most Google Trends Actors are built on Google's explore and widgetdata endpoints — the ones behind the interest-over-time charts. Those endpoints return HTTP 429 to datacenter IPs, which is where every Apify Actor runs. Measured on 2026-07-22: four attempts over 44 seconds from a freshly cookie-primed session, all 429. That is why Trends Actors on this Store carry 1.5 to 3.3 star ratings, and why runs come back empty.

This Actor uses Google's public trending RSS feed instead. No token handshake, no cookie priming, no proxy, no rate limit observed across 12 countries tested. It returns less than the others promise, and it returns it every time.

What you get

One row per trending search:

• the search term, and its rank in that country
• Google's traffic estimate, both raw ("200+") and as a sortable number
• when it started trending
• every news story Google links to the trend — title, URL, publisher
• a thumbnail and its credit
• the exact feed URL the row came from

What this Actor does NOT do

It does not return interest-over-time, the 0–100 popularity curve for a keyword you choose. That is the widgetdata endpoint described above, and it cannot be served reliably from Apify without residential proxies. If you need keyword history, this is the wrong Actor and you should not buy it.

This one answers "what is trending right now, where, and why" — for newsrooms, social and content teams, trend monitoring, and AI agents that need a live read on public attention.

Input

{
  "geos": ["US", "AU", "GB"]
}

Each country returns its current top 10, so three countries returns 30 results. Leave it as ["US"] for a single country.

Supported: AR AT AU BE BR CA CH CL CO CZ DE DK EG ES FI FR GB GR HK HU ID IE IL IN IT JP KE KR MX MY NG NL NO NZ PH PL PT RO RU SA SE SG TH TR TW UA US VN ZA.

An unrecognised code fails the run immediately rather than after the fetch — Google answers an unknown country with an empty 200, which would otherwise look like a successful run that returned nothing.

Output

Captured from a real run, 2026-07-22:

{
  "geo": "US",
  "rank": 1,
  "query": "nottm forest vs blackburn rovers",
  "approx_traffic": "200+",
  "approx_traffic_min": 200,
  "started_at": "Wed, 22 Jul 2026 02:40:00 -0700",
  "picture": "https://encrypted-tbn0.gstatic.com/images?q=tbn:ANd9GcTmlzigFLxgYOjaFmIQGM1kyvaa7SsZTC29Ydo10X7zJI9pVdLeeMezX5n4CoA",
  "picture_source": "Lancashire Telegraph",
  "news_item_count": 3,
  "news": [
    {
      "title": "Nottingham Forest view on Blackburn Rovers friendly and clear marker awaiting",
      "url": "https://www.lancashiretelegraph.co.uk/sport/26297524.nottingham-forest-view-blackburn-rovers-friendly/",
      "source": "Lancashire Telegraph",
      "snippet": null
    }
  ],
  "source_url": "https://trends.google.com/trending/rss?geo=US"
}

approx_traffic is Google's published estimate and is a floor, not a measurement — "200+" means at least 200. approx_traffic_min is the same value as an integer so you can sort and filter on it; it does not add precision Google did not give.

Notes on the data

• Google publishes exactly 10 trending searches per country. That is the feed's limit, not a setting.
• snippet is usually null — Google leaves it empty in the feed. It is kept in the shape so the schema is stable rather than dropped.
• Trends move fast. Schedule the Actor if you want a history; each run is a snapshot of that moment.

Reliability

The test suite includes a live check that the feed still returns 10 rows for US, GB and AU. If Google ever starts throttling this feed the way it throttles widgetdata, that check fails and the Actor gets fixed rather than quietly returning empty runs.