SerpApi and DataForSEO both sell structured JSON for Google Trends Explore. Neither is a first-party Google API. Neither emits source tiktok. The useful difference is transport and money. SerpApi: GET https://serpapi.com/search?engine=google_trends. DataForSEO: POST https://api.dataforseo.com/v3/keywords_data/google_trends/explore/live (or Standard POST/GET). Docs for the shape of a third option, if a shared 0-100 across platforms is the next step, live on the API reference. Plan text for that option: pricing.
Synchronous search versus a task array
SerpApi opens a connection and returns interest_over_time as an object. async=true exists but the default is sync. Cached identical searches are free for an hour. DataForSEO wraps work in tasks[]. Live waits. Standard queues (docs mention up to 45 minutes on some Keyword Data copy). Pingback and postback are first-class.
What both still scrape
Interest over time, region maps, related topics, related queries, gprop / type switches for YouTube, Images, News, Shopping. Up to five keywords on a comparison. Values are relative to the peak in that request, like the website. If Google changes Explore, both vendors chase the same HTML.
Related queries on both
That widget is not a differentiator. Client libraries are. SerpApi ships first-party SDKs and a playground. DataForSEO ships a large v3 surface and a shared auth header across SEO endpoints. A team already on one vendor for SERPs will add Trends cheaper in engineering time than in cents.
Cost shape: monthly searches versus per-task
SerpApi sells monthly search allotments. A Trends call is one search. DataForSEO sells tasks. One explore task can carry up to five keywords. Per-task list prices on DataForSEO's own Keyword Data page were a fraction of a cent on Standard when checked in this run. Throughput limits differ (SerpApi publishes searches per hour per plan; DataForSEO publishes 2000 calls per minute). Model the real monthly mix, then open both pricing pages again.