SerpApi's RELATED_QUERIES and RELATED_TOPICS data types return rising and top keyword lists for a seed term. Each successful call costs one credit from the shared monthly pool, accepts only one query per request, and cannot be batched like interest-over-time. This page covers only related-query pricing: per-pull cost, expansion-loop math, and when keyword discovery belongs in the budget.
Keyword discovery is where trend API budgets quietly double. SerpApi exposes rising and top related queries through engine=google_trends with data_type=RELATED_QUERIES or data_type=RELATED_TOPICS. Every fresh pull costs one credit from the account's shared monthly pool. Unlike interest-over-time, related-query calls accept only one keyword per request. SEO teams that run expansion loops without modeling depth often exhaust a Starter allocation in a single afternoon.
For SerpApi's full Google Trends pricing across timeseries, autocomplete, and Trending Now engines, see SerpApi Google Trends API pricing. This page stays on related queries and related topics only.
SerpApi scrapes Google's related-queries panel and returns JSON. A typical response includes a related_queries object with two arrays:
rising: queries gaining momentum relative to the seed term, often labeled with percentage growth (for example +4,200%) or Breakouttop: queries with sustained co-search volume, scored 0-100 relative to the highest item in the setEach item carries query, value, extracted_value, link, and serpapi_link. The serpapi_link field points to a pre-built follow-up call, useful for chaining into interest-over-time without hand-building URLs.
The request shape is:
GET https://serpapi.com/search?engine=google_trends&q=coffee&data_type=RELATED_QUERIES&api_key=YOUR_KEY
Optional parameters mirror the main Trends engine: geo for country, date for time window (default today 12-m), gprop for search type ("" for web, youtube, news, froogle for Shopping, images), and cat for category ID. no_cache=true forces a live scrape and always costs one credit. Cached responses within one hour of an identical query are free.
data_type=RELATED_TOPICS returns Knowledge Graph topics instead of raw query strings. Each item adds title, type (Programming language, Company, and similar labels), and entity id fields. Pricing is identical: one credit per successful pull.
Related queries draw from the same credit pool as Google Search, autocomplete, Trending Now, and every other SerpApi engine.
| Plan | Monthly price | Included searches | Cost per fresh pull (if fully used) | Throughput cap |
|---|---|---|---|---|
| Free | $0 | 250 | $0.00 | 50/hour |
| Starter | $25 | 1,000 | $0.025 | 200/hour |
| Developer | $75 | 5,000 | $0.015 | 1,000/hour |
| Production | $150 | 15,000 | $0.010 | 3,000/hour |
| Big Data | $275 | 30,000 | $0.009 | 6,000/hour |
Result count does not change the charge. A response with 25 rising queries and an empty top array both cost one credit when the request succeeds.
Related queries are rarely a one-call workflow. Three common patterns show how credits multiply.
An SEO analyst seeds electric vehicles, pulls related queries, then pulls related queries for each of the 10 rising results to map a second-level cluster.
| Step | Calls | Credits |
|---|---|---|
| RELATED_QUERIES on seed | 1 | 1 |
| RELATED_QUERIES on each rising term | 10 | 10 |
| Total (discovery only) | 11 | 11 |
On Starter ($25, 1,000 credits), effective discovery cost is $0.275 for the batch. Add interest-over-time on 11 discovered terms and the sprint consumes 22 credits before any geographic or category filters.
SerpApi allows RELATED_QUERIES without a q value when cat and geo are set, returning top and rising queries for an entire category. Twenty category-by-country combinations at one pull each: 20 credits per week, roughly 80 per month. Fits Free tier (250 credits) with room for timeseries validation on a shortlist.
A content team pulls related queries for 100 seed keywords, then interest-over-time on the top 3 rising results per seed.
| Step | Calls | Credits |
|---|---|---|
| RELATED_QUERIES per seed | 100 | 100 |
| TIMESERIES per rising result (3 x 100) | 300 | 300 |
| Total | 400 | 400 |
Requires Developer plan ($75, 5,000 credits). Effective cost: $6.00 for the batch if no other engines share the pool. On Free (250 credits), the job stops after the first 250 calls.
For autocomplete credit math in the same SerpApi cluster, see SerpApi Autocomplete API pricing.
Both data types cost one credit and accept one query per call. The difference is what Google returns.
| data_type | Output | Best for |
|---|---|---|
| RELATED_QUERIES | Raw search strings (coffee near me, maca coffee) |
SEO keyword lists, content gap analysis |
| RELATED_TOPICS | Knowledge Graph entities (Python, TikTok) with type labels |
Brand monitoring, topic taxonomy mapping |
Teams running both on the same seed keyword pay two credits. A 50-keyword batch with both types needs 100 credits before any timeseries work.
Trends MCP does not expose a dedicated related-queries endpoint. Keyword momentum is available through get_growth with source: "google search" and percent_growth windows, or get_ranked_trends with sort: "wow_pct_change" for ranked breakout lists. Each call equals one request on the monthly allocation.
| Plan | Monthly price | Included requests |
|---|---|---|
| Free | $0 | 100 |
| Starter | $19 | 1,000 |
| Pro | $49 | 5,000 |
| Business | $199 | 25,000 |
For the 100-seed content calendar (growth check per seed, no second-level expansion), Trends MCP Starter ($19, 1,000 requests) covers 100 lookups with headroom. SerpApi Starter ($25, 1,000 credits) needs 400 credits if the pipeline pulls related queries plus timeseries on rising results.
Neither approach returns Google's exact rising-query panel. SerpApi wins when the workflow must mirror Google Trends UI related-query lists for client deliverables. Trends MCP wins when the question spans TikTok, Reddit, YouTube, or Amazon growth in the same JSON contract and ranked percent-change signals are enough.
SerpApi related queries fit SEO teams that export Google's rising and top panels into spreadsheets or BI tools and already pay for SerpApi on other engines. Category-level discovery without a seed keyword (cat plus geo, no q) also maps cleanly to SerpApi's parameter set.
SerpApi related queries are a poor fit when the budget counts credits per keyword and the pipeline runs multi-level expansion loops, or when the research question requires signals outside Google. Paying one credit per related-query pull on Google alone, then wiring separate vendors for TikTok and Reddit, stacks cost faster than a multi-source API with one request shape.
For the broader vendor comparison, see the trend data API pricing comparison. For a feature-level view beyond price, see Trends MCP vs SerpApi for trend data.
FAQ