Google Trends API

A trends API is an interface that lets you query, programmatically, how search interest in a topic changes over time. Instead of opening Google Trends and comparing keywords by hand, you make a single HTTP request from your own application and receive the data as structured JSON, ready to store, chart or feed into your own reports. That makes it a practical building block for teams building SEO tools, content teams, agencies and anyone who tracks demand over time.

The Semust Trends API works through a single endpoint: you POST to data.semust.com/v1/trends and pick one of four reports with the type parameter — interest, regions, related or demographics. Interest, regions and related searches are Google Trends data; demographics is a clickstream-based estimate. Authentication is header-based and every response is JSON.

Search volume tells you how big a keyword is; trend data tells you which way it is moving. Seeing when interest in a product, a topic or a brand rises and falls helps you time your content, prepare for seasonal demand, and notice when a term is giving way to a newer one.

With search_type you choose where searches are counted: web, news, images, YouTube or shopping. A keyword that is flat on web search can behave very differently on YouTube or in shopping, and comparing them shows you which channel deserves the effort. Run these checks through an API and they can repeat on a schedule, not only when someone remembers.

Trend values are relative, not search counts. In the interest report, 100 is the highest point any of the keywords reaches in the period, and every other value is scaled against it. Because up to five keywords share that scale, you can compare them directly. The response also gives each keyword's average over the period, so a single request tells you which term leads and by how much.

Six periods are available, from the past day to the past five years. The interval follows the period: minutes for past_day, hours for past_7_days, days or weeks for the longer periods. The most recent interval may still be in progress; it is marked partial: true, so you know its value will change and can leave it out of your comparisons.

Pricing is flat per request: $0.0121 for interest, regions and related, and $0.00264 for demographics. The price does not change with the number of keywords, the period or the search type, so comparing five keywords over five years costs the same as checking one keyword for a single day.

A keyword with no data is still a valid answer — items is empty and values are null — and it is billed like any other request. Results are cached for 1 hour for past_day, 6 hours for past_7_days and 24 hours for longer periods; a cached answer is marked cached: true and billed at the same price as a live one. You are never charged for an error.

The regions report shows where interest comes from. For Turkey it breaks interest down by province, and every row carries the region's ISO 3166-2 code (such as TR-34), so you can join it to your own data or plot it on a map. Rows are sorted by the first keyword's interest, highest first.

Each row gives two views. interest puts every keyword on its own scale, where 100 is that keyword's strongest region; share says how the keywords split that region's searches, in percent. Together they show both where a keyword is strongest and where it wins or loses against the others. Region names come back in the language you choose.

The related report returns what people search for alongside a keyword: top and rising related searches, plus top and rising related topics, each topic with its category. It takes one keyword per request.

In the top lists, value is relative interest from 0 to 100. In the rising lists, it is growth over the previous period, in percent — which makes them a good source of fresh content ideas and of long-tail keywords that are only starting to gain interest.

Google Trends shows where and when people search, but not who. The demographics report fills that gap with interest by age group and gender, for up to five keywords. It is a clickstream-based estimate, not Google data.

Values run from 0 to 100 against the keyword's largest group, so they show how the groups compare rather than how many people search. The report costs $0.00264 per request. It does not use language or search_type, and sending either is rejected rather than silently ignored.

All you need is an API username and password, created from the API Access page in your account and sent in the SEMUST-API-USER and SEMUST-API-PASSWORD headers. Requests are standard HTTP POST and responses are JSON. Every error carries a stable code, a readable message and, where it helps, a hint; send Accept-Language: tr to receive messages in Turkish. A misspelled parameter is named in the error with a suggestion.

Requests run as tasks. If one takes longer than 60 seconds you get 202 with a task_id and collect the result from GET /v1/tasks/{task_id}; every response carries the task id in the X-Task-Id header, and results are kept for 72 hours after the task finishes. To retry safely after a timeout, send an Idempotency-Key: the same key returns the same task, and it is charged once.