cloro
Technical

Google Search API: Official vs SERP API

Ricardo Batista
Founder, cloro
8 min read
GoogleSERP APIComparison
On this page

The Google Search API question is confusing because three different products share the name. There is Google’s official Custom Search JSON API, there is DIY Google scraping, and there are third-party SERP APIs.

They do not return the same thing. The official API is structured but limited. DIY scraping can match the live SERP but is fragile. SERP APIs trade cost for coverage, parsing, localization, and anti-bot maintenance.

This guide compares the options, their limits, and when each one fits. For vendor selection, see best SERP APIs and the cheapest SERP API breakdown.

The three options

Three ways to get Google SERP data via API: the official Custom Search JSON API (closing to new customers by January 2027), third-party SERP APIs like SerpApi from $25 per 1,000 searches, and cloro's SERP API at $0.40 per 1,000 with AI Overview coverage — note the Search Console API returns your own analytics, not live SERPs

OptionWhat it returnsCostUse case
Google Custom Search JSON API (official)Results from a custom search engine you configure$5/1k queries, capped at 10K/dayEmbedding search on your own site
DIY Google scraping (unofficial)Real Google.com SERP, but you fight blocksProxy + infra costsOne-off projects, learning
Third-party SERP APIReal Google.com SERP, parsed JSON, managed infra$0.40–25 per 1k depending on providerSEO monitoring, rank tracking, competitive intel

Most “Google Search API” searches end up wanting option 3. People spend time evaluating option 1 first because it’s official and Google-branded, then discover it does not return Google.com results at all.

Option 1: Google Custom Search JSON API (official)

Google’s official offering is the Custom Search JSON API. The headline numbers sound competitive with SERP API pricing. The catch is what you get back.

What the Custom Search JSON API actually returns

The Custom Search API returns results from a custom search engine you configure in Google’s Programmable Search Engine console. You define which sites the engine indexes, optionally with weights and filters. Google’s own documentation confirms that the API runs requests against a Programmable Search Engine instance you create, not against Google.com’s full index by default.

What this means in practice comes down to three points:

  • Not the same as Google.com SERP. The custom engine is a curated subset. Ranking positions, ads, AI Overview, shopping cards, and featured snippets won’t match what a user sees on Google.com.
  • Designed for embedded search. The intent is “add a Google-powered search box to your own website”, and for that use case the API is genuinely good.
  • Capped at 10,000 queries/day. Even if you’re willing to pay, you can’t scale past roughly 300K queries/month per project.

For SEO monitoring or rank tracking, the Custom Search API is the wrong tool. The data isn’t comparable to SERP data at any volume.

Pricing and the 10,000-query daily cap

The pricing looks approachable at first glance. Google documents a free tier of 100 search queries per day, with additional queries at $5 per 1,000 up to a hard ceiling of 10,000 queries per day.

That ceiling is the real constraint. A rank tracker covering a few thousand keywords across multiple locations and devices blows past 10K daily queries almost immediately. The $5/1k rate never becomes the bottleneck, because the daily cap stops you first.

Google is closing the Custom Search JSON API to new sign-ups

There is a bigger reason not to build on it today. Google has closed the Custom Search JSON API to new customers, and its documentation states that existing users must transition off the product by January 1, 2027.

Google now points teams toward Vertex AI Search for web and enterprise search use cases. That migration path does not restore Google.com SERP data — it is another embedded/enterprise search product, not a rank-tracking source. If you are choosing an approach in 2026, starting on a sunsetting API with a fixed shutdown date is a poor foundation.

Option 2: DIY Google scraping (unofficial)

You can scrape Google.com directly with requests, puppeteer, or playwright against a proxy pool. The result is real Google SERP data with actual ranking positions, real AI Overview, and real shopping cards. The problem is everything you have to maintain to keep that working.

The anti-bot reality

Google fingerprints requests on the TLS handshake, headers, behavior, and IP reputation. A naive scraper gets captcha’d or blocked within minutes. We covered the depth of this in proxies for SERP scraping.

The operational overhead compounds from there. You’re maintaining proxy pools, handling captchas, parsing dynamic SERPs, and managing country-level targeting. Every time Google changes the SERP layout, your parser breaks and someone has to fix it.

Scraping public search results sits in a legal gray area rather than a clear yes or no. In the US, the 2022 hiQ Labs v. LinkedIn decision held that scraping publicly accessible data does not violate the Computer Fraud and Abuse Act. That ruling provides meaningful cover for scraping public SERPs.

It is not a blanket permission slip, though. Google’s Terms of Service still prohibit automated access, which is a contract-law question rather than a criminal one. We covered the broader picture in is website scraping legal.

DIY scraping is viable for one-off research or learning exercises. For sustained operations, the proxy and anti-bot infrastructure becomes a side business you have to run. Most teams that try this move to a SERP API within 6 to 12 months.

Option 3: Third-party SERP APIs

Modern SERP APIs handle the scraping infrastructure for you. The list includes cloro, SerpApi, DataForSEO, Serper, ScraperAPI, ScrapingBee, and BrightData. You make a single REST call and the provider does the rest.

What you get back

The provider handles proxy rotation, captcha solving, fingerprinting, country targeting, and parsing. What comes back is clean JSON with the actual Google.com SERP — organic positions, ads, AI Overview, featured snippets, and shopping cards. This is what most teams asking about “Google Search API” actually want.

For buyers actively shopping the category, see our ScraperAPI alternatives and Serper alternatives breakdowns.

The trade-offs

A managed SERP API is not free of compromises. Three are worth weighing before you commit:

  • Cost. Providers charge $0.40 to $25 per 1,000 calls depending on provider and feature mix. We compared pricing in cheapest SERP APIs.
  • Provider lock-in. You’re committed to the provider’s response shape and uptime. Multi-provider redundancy is possible but adds complexity.
  • No legal cover. The provider scrapes Google on your behalf, but you’re still using the data they return. The legal posture is similar to DIY, except the operational risk shifts to the provider.

For SEO monitoring, rank tracking, competitive intelligence, or anything else that needs real Google.com SERP data at scale, this is the right answer.

Why the confusion exists

The “Google Search API” search term has accumulated meanings over more than a decade. The original Google AJAX Search API was deprecated in 2010. The Google Search Appliance, a physical search box for enterprises, was phased out from 2016 and fully discontinued by 2019. The Custom Search products that survived evolved into today’s Programmable Search Engine.

Add the modern third-party SERP APIs that compete on the same query, and the term now refers to at least three different things. That is why a single search phrase pulls up an official Google product, a stack of scraping tutorials, and a dozen commercial APIs all at once.

When someone asks for “the Google Search API”, the right clarifying question is simple. Do you need actual Google.com SERP data, or do you need to embed search on your own site? The answer picks the option for you.

Google Search API vs Google Search Console API

One more product gets swept into the “Google Search API” question, and it is worth separating out. The Google Search Console API is not a Google Search API in the sense most people mean, because it does not return live rankings for arbitrary queries.

The Search Console API returns your own property’s performance data. Google’s Search Analytics documentation describes it as a way to query clicks, impressions, click-through rate, and average position for a site you have verified in Search Console. In other words, it reports how your pages already performed, not where any page ranks right now.

That distinction matters when you are choosing a Google Search API for a project:

  • Search Console API answers “how did my verified site do in Google Search?” It is free, first-party, and limited to properties you own.
  • A SERP API answers “what does Google.com show for this query, from anyone’s perspective?” It covers any keyword, any location, and any competitor.

If you want to track your own click-through rate and average position over time, the Search Console API is the correct free tool, and no scraping is involved. If you want competitor rankings, share-of-voice, or AI Overview citations, the Search Console API cannot help — you are back to needing a SERP API.

People conflate the two because both are official Google interfaces with “search” in the name. But a Google Search API that returns the live SERP and a Search Console API that returns your own analytics solve completely different problems.

When the official Custom Search API is actually right

There are three legitimate use cases, all of them embedded-search rather than SERP-data:

  1. Embedded search on your own site. Your website needs a search box that returns results from your own pages or a small curated set of sites. Custom Search API is purpose-built for this.
  2. Internal index. You’re indexing a private corpus such as internal documentation or an intranet, and you want Google’s ranking quality without scraping Google.com.
  3. Light-touch web search. You need basic web search results for a small-volume use case (under 10K queries/day total) and don’t care about specific Google.com ranking signals.

For these, the Custom Search API works well — with the caveat that new projects should account for the January 2027 migration. For everything else, reach for a SERP API.

Decision tree

Your use casePick
Embed search on my own siteGoogle Custom Search JSON API
Index a private corpusGoogle Custom Search JSON API (or a self-hosted alternative like Elasticsearch)
SEO rank trackingSERP API (cloro, SerpApi, etc.)
Competitive intelligence on SERPsSERP API
AI Overview citation trackingSERP API with AI Overview support (cloro)
One-off research project, low volumeDIY scraping (proxies + careful anti-bot work)
Sustained scraping operation, high volumeSERP API
AI engine coverage beyond GoogleSERP API with multi-engine support (cloro’s AI visibility API)

Pricing comparison (apples-to-apples where possible)

The Custom Search API and SERP APIs aren’t truly comparable, because they return different data. But for the specific case where someone is choosing between them on cost alone, here is the shape of it:

  • Google Custom Search: $5/1k queries, capped at 10K/day, returns custom engine results (not Google.com SERP), and closing to new customers.
  • cloro: $0.40/1k queries, no cap, returns actual Google.com SERP including AI Overview.
  • SerpApi: $25/1k queries on its entry Starter plan, dropping to roughly $9/1k at higher volume tiers, returns actual Google.com SERP.

If the data difference matters — and it usually does — the cost comparison is moot. The Custom Search API doesn’t return what you need at any price. If the data difference doesn’t matter, which is rare, the Custom Search API is still capped at 10K/day, and that kills most production use cases.

Bottom line

For 95% of teams asking about “Google Search API”, the answer is a third-party SERP API. That means cloro for the lowest per-call cost and AI engine coverage, SerpApi for established reliability at premium pricing, or one of the alternatives we compared in SerpApi alternatives.

Google’s Custom Search JSON API is the right call for embedded-search use cases such as a search box on your own site, and the wrong call for anything that needs actual Google.com SERP data. The data difference isn’t a coverage limitation that disappears at a higher pricing tier; it’s a different product — and one Google is now retiring.

If you’re starting fresh and need real Google SERP data at scale, the cloro SERP API is the cheapest defensible option with full Google feature coverage including AI Overview. For broader context, see best SERP APIs and cheapest SERP APIs.

Ricardo Batista

About the author

Founder, cloro

Ricardo is one of the founders and engineers behind its SERP and AI-search scraping infrastructure. Before cloro he scaled a financial comparison site to $7M ARR and ran the full-country operations of a unicorn to $65M ARR, then went back to building. He writes about search engine scraping, generative-engine optimization, and turning live search and AI-answer data into something teams can act on.

Frequently asked questions

Is there an official Google Search API?+

Yes, but with major limitations — and it is being wound down. Google's official Custom Search JSON API ($5 per 1,000 queries, capped at 10,000 queries/day) returns results from a custom search engine you configure, not the full Google.com index. It's designed for embedding search on your own site, not for SEO/SERP scraping, and Google has closed it to new customers with a migration deadline of January 1, 2027. For real Google SERP data — actual ranking positions, ads, AI Overview, shopping cards — you need a SERP API like cloro, SerpApi, or DataForSEO that scrapes Google.com directly.

Can I scrape Google legally?+

Scraping public Google search results sits in a legal gray area that depends on jurisdiction and methodology. The 2022 hiQ Labs vs LinkedIn ruling in the US established that scraping publicly accessible data isn't a CFAA violation, which provides cover for SERP scraping. However, Google's Terms of Service prohibit automated scraping, which creates a contract-law (not criminal) risk. Most teams use a SERP API provider that handles the scraping at scale, which shifts the operational risk to the provider. We covered the broader legality picture in is website scraping legal.

What's the difference between Google Custom Search API and a SERP API?+

Google Custom Search API returns results from a custom search engine — a curated subset of the web you configure. It does not return Google.com's actual SERP. SERP APIs (cloro, SerpApi, etc.) scrape Google.com directly and return the actual SERP — organic positions, ads, AI Overview, featured snippets, shopping cards, exactly what a user would see. For SEO monitoring, rank tracking, or competitive intelligence, you need a SERP API; the Custom Search API does not produce the data you need.

Why is Google's official Search API so limited?+

Google's strategic interest is in keeping users on Google.com, not enabling third parties to replicate or compete with Google's search experience. The Custom Search API is intentionally scoped to embedded-search use cases (your own site search, an internal index) rather than SERP-data use cases (rank tracking, competitive intelligence, AI training). The limitations are by design, not by accident.

How much does the Google Custom Search API cost?+

$5 per 1,000 queries above the free 100/day, capped at 10,000 queries/day. For comparison, SERP APIs that return actual Google.com SERP data cost $0.40-25 per 1,000 calls depending on provider, with no daily cap. The Custom Search API's per-query cost looks competitive on paper, but its data isn't equivalent to SERP API data, so the comparison is misleading.