If you searched for a Google Trends API, you want Google Trends data in code: a curve for a term, a breakdown by region, the related queries. Google now has an official API for that, but few developers can use it yet. This post covers what the official API returns, how to apply, what to use until you are accepted, and what any of them can and cannot tell you. We read Google's pages and each provider's own pages on 8 November 2026.

TrendsFast is our product. It is not a Google Trends API, and we say below where it fits and where it does not.

Is there an official Google Trends API?

Yes. Google announced the Google Trends API on 24 July 2025 as an alpha, "available only to a very limited number of testers". In November 2026 its page is still titled Google Trends API Alpha, and it still says Google is accepting applications for alpha testers. Google has published no price, quota or terms for it, and no list of endpoints.

What Google says it returns:

  • About five years of history: a rolling window of 1,800 days, up to two days ago.
  • The interval you choose: daily, weekly, monthly or yearly.
  • Regions and sub-regions: countries and their sub-regions as ISO 3166-2 defines them.
  • Numbers you can combine: the website scales every request from 0 to 100 on its own, so two requests do not compare. The API uses one scale across requests, so to follow a term over time you fetch only the newest period instead of the whole range again.
  • More terms at once: Google says the API makes it easier to compare dozens of terms; the website compares a few at a time.

Neither Google page says which surfaces it covers: web search, news, images, shopping or YouTube.

How to apply, and what to expect

Google asks developers to apply through a form on the API's page. It prioritises developers "that know what they want to do, that can start doing it soon, and that are willing to provide feedback", and it opens access on a rolling basis.

Plan for a wait. In Google's own help community, applicants report weeks without an answer; one thread is titled "Google Trends API alpha access application — no response received". Apply with a concrete use case, and build on one of the options below in the meantime. If access comes, you change the data source, but its scale is not the website's, so old and new numbers do not line up.

What to use until you are accepted

SerpApi, SearchApi, DataForSEO and pytrends return the kind of number the Google Trends website shows; Glimpse adds an estimate of volume, and the official API has its own single scale. Each fact comes from the vendor's own pages on the date in the last column; where a vendor does not publish something, the table says so.

OptionWhat you getAccessPriceFreeWatch outChecked
Google Trends API (official)Google search interest by region and sub-region, daily, weekly, monthly or yearly.Alpha only, for accepted testers; developers apply through a form on Google's page.Not published.Not published.Not generally available, and search interest only.
SerpApi Google Trends APIGoogle Trends interest over time, by region, related topics and queries; separate Trending Now and autocomplete APIs.Yes, REST; up to five queries per search for interest over time.$25 a month for 1,000 searches; $75 for 5,000; $150 for 15,000.250 searches a month.Search only; social trends are outside its scope.
SearchApi Google Trends APIGoogle Trends interest over time and by region, related queries and topics, for web, image, news, shopping and YouTube search.Yes, REST; up to five queries per interest-over-time request, the key as a parameter or a bearer header.$40 a month for 10,000 searches, $100 for 35,000, $250 for 100,000; only successful searches are billed.100 free requests on sign-up, with no card.A plan may use at most 20% of its monthly searches in any hour; search interest only, not social posts.
DataForSEO Google Trends APIGoogle Search, News, Images, Shopping and YouTube search interest.Yes, REST; up to five keywords per task, standard or live mode.$0.0027 per standard task, $0.011 live; pay as you go from a $50 top-up.$1 trial credit.Relative search interest, not social posts and not search volume.
GlimpseGoogle search data, with a channel breakdown in the app.Yes, REST endpoints for search interest and enriched volume; weekly, daily or monthly series.Paid app access from $99 a month (stated on its own guide page); API pricing not published.Up to 10 searches a month in the app.Search demand only; social posts are not its data.
pytrends (unofficial)The Google Trends website's data in Python, through the site's own endpoints.A Python library. Archived on 17 April 2025; the last release, 4.9.2, is from April 2023.Free.Free.Read-only since April 2025, and its README always said it was only good until Google changed its backend. Its rate limit was never known.
  • If you need consistently scaled Google data you can merge across requests: apply for the official alpha, and build on a provider until you are in.
  • If you need a few hundred requests a month to try an idea: SerpApi's free searches.
  • If you need thousands of requests a month at a flat monthly price: SearchApi.
  • If you need the lowest price per request, and you can wait for results: DataForSEO's standard tasks.
  • If you need a number of searches rather than an index: Glimpse's enriched endpoint.
  • If you need what your niche discusses and what AI assistants answer, beside search: TrendsFast (our product).

A first request

SerpApi and SearchApi both take a Google Trends request as one GET with engine=google_trends. This asks SerpApi for twelve months of US interest in two terms, with the key kept in your environment:

curl -G https://serpapi.com/search \
  --data-urlencode "engine=google_trends" \
  --data-urlencode "q=cold brew,iced coffee" \
  --data-urlencode "data_type=TIMESERIES" \
  --data-urlencode "geo=US" \
  --data-urlencode "date=today 12-m" \
  --data-urlencode "api_key=$SERPAPI_API_KEY"

Over twelve months the points are weekly, and each is scaled against the highest point in that one request. SearchApi takes the same engine and data_type at https://www.searchapi.io/api/v1/search, with the key as a parameter or a bearer header. DataForSEO takes up to five keywords per task: a standard task is posted and collected later, and a live one answers in the same request.

What the numbers mean

Google describes Trends data as "a largely unfiltered sample of actual search requests made to Google". Each point is divided by all searches in its place and time, then scaled from 0 to 100 by its share of all searches. Every Google Trends API inherits what that means:

  • It is not search volume. Google says two regions with the same search interest do not always have the same total searches.
  • Small terms read as zero. Terms with low volume appear as 0.
  • Some searches are left out. Trends removes repeated searches by one person over a short time, and searches made by Google's own products and services, which include internal searches by AI Mode and AI Overviews.
  • It carries noise on purpose. Google adds small random fluctuations to protect privacy, most visible on terms with little interest, and says Trends "is not a scientific poll".
  • Each website request is its own scale. Values from two requests cannot be stitched together; the official API's single scale is the exception.

DataForSEO's pricing page puts it plainly: "Google Trends values demonstrate keyword popularity on a relative scale, not in absolute values." If you need a count, Glimpse says its enriched endpoint returns absolute search volume over time; it does not publish an API price.

When search interest is not the question

A Google Trends API answers one question well: how interest in a term you already know moves on Google. An agent deciding what to post usually needs two more answers: what people in the niche discuss before they search, and what AI assistants say when a buyer asks. TrendsFast's Search product reads Google Trends interest and related queries for your brand's niche and asks ChatGPT, Claude, Gemini and Perplexity your category's buyer question, for 35 credits a call. Trending reads TikTok, YouTube, Instagram, Facebook, X, Reddit, Hacker News and Google Trends for 25. It does not return a raw Google Trends series for any term you choose, and a call costs more than one request to the APIs above: for that job, use them.

FAQ

Is there an official Google Trends API?

Yes. Google announced the Google Trends API on 24 July 2025 as an alpha for a very limited number of testers. In November 2026 it is still an alpha: developers apply through a form on Google's page, and Google admits them on a rolling basis.

Is the Google Trends API free?

Google has published no price, quota or terms for its alpha, and the Google Trends website is free. Providers that serve Google Trends data by API charge: DataForSEO $0.0027 per standard task, SearchApi $40 a month for 10,000 searches and SerpApi $25 a month for 1,000.

Does the Google Trends API give search volume?

No. Google Trends divides each point by all searches in that place and time and then scales it, so it shows relative interest, not a number of searches. The official API reports search interest too. Glimpse's API has an endpoint that returns absolute search volume.

Does pytrends still work?

pytrends, the unofficial Python library, was archived on 17 April 2025, and its last release, 4.9.2, dates from April 2023. Its README warned it was only good until Google changed its backend. A maintained provider, or the official alpha once accepted, is the safer choice.

How far back does Google Trends data go by API?

Google's alpha returns a rolling window of 1,800 days, about five years, up to two days ago. The Google Trends website goes back to 2004, and providers that read it, such as SerpApi, accept dates from 2004.

How we checked

On 8 November 2026 we read Google's API page, its announcement of July 2025 and its Trends help page, each provider's pricing and documentation pages, and the pytrends repository and its PyPI page. We quote only what they publish. We did not apply for the alpha, buy any plan or measure anyone's data, our own included.

Related reading