SEO & AI search playbook for API companies

Searches to plan around

Query patterns and the intent behind them. We don’t quote search volumes: check your own Search Console for the ones you already appear for.

• “[capability] API”, “free [capability] API”

  Find a provider

  Say what the free tier includes in requests per month or similar, on the page.

• “[task] in [language]”, “[capability] [language] SDK”

  Implementation

  Language quickstarts with copy-paste code, one page per language you have an SDK for.

• “[provider] rate limits”, “[provider] pricing per request”

  Evaluation

  Put limits and prices in text on crawlable pages.

• “[provider] API alternative”

  Switching

  Common after a price change or outage. A migration guide with an endpoint mapping table helps.

Page families that work

• Endpoint reference

  /docs/api/messages/create

  One page per endpoint, generated from your OpenAPI spec but rendered on the server, with descriptions written by people.

• Language quickstarts

  /docs/quickstart/python

  Only languages with real SDKs or tested examples. Each has different install and auth steps.

• Use-case recipes

  /recipes/two-factor-auth

  End-to-end examples that combine endpoints for a real job.

• Migration and mapping pages

  /migrate/from-other-api

  Endpoint-by-endpoint mapping tables, which are factual and hard to fake.

Risky page types

• Code samples in 30 languages, auto-generated

  Identical pages with a different syntax highlighter are thin. Publish languages you test.

• Public API explorer URLs

  Try-it consoles with query parameters can create endless crawlable URLs. Keep them out of the index.

Technical pitfalls

1. 1

   Client-rendered reference docs

   Reference UIs that build the page from a spec in the browser can serve crawlers an empty shell. Google recommends server-side or pre-rendering; render the reference to static HTML at build time.

2. 2

   Hash-based routing in docs

   URLs like /docs#/messages/create are one page to Google. Google's JavaScript guide says to use the History API for routes, not fragments.

3. 3

   Every API version indexed

   v1, v2 and v3 of the same page compete with each other. Canonicalise older versions to the current one where content is the same, and mark deprecated pages clearly.

How AI assistants answer questions in this category

Developers ask coding assistants to write integration code. When an API isn't well represented in what the model learned, assistants can invent plausible-looking endpoints or parameters. Assistants that can fetch docs do better, if the docs are readable as text. The llms.txt proposal exists for this case, and a clean, server-rendered reference helps every tool, whether or not it reads llms.txt.

What tends to earn a citation

• Reference pages with the exact endpoint, parameters, types and an example response, in HTML.
• A changelog with dates, so assistants can tell current behaviour from old behaviour.
• A pricing page with per-request prices and limits written as text.

Check whether the AI crawlers you want can reach your site with our AI crawler checker.

Where links come from

• SDK repositories

  Official SDKs on GitHub and package registries link back to your docs. Set the homepage field on each package.

• Sample apps and templates

  Starter apps on platforms developers already use get forked and linked.

• API and developer directories

  List in developer tool directories with a direct link to your quickstart.

A 90-day plan

1. Days 1–30

   Make the reference crawlable

   • –Render reference docs to static HTML.
   • –Replace hash routes with real paths.
   • –Canonicalise duplicate versions.

2. Days 31–60

   Quickstarts and pricing

   • –Write quickstarts for your top three languages with tested code.
   • –Put limits and per-request pricing in text.
   • –Publish SDKs with homepage links.

3. Days 61–90

   Recipes and migration

   • –Publish five use-case recipes.
   • –Write a migration guide from your main competitor.
   • –Review Search Console for endpoint pages with impressions and improve their descriptions.

What to watch in Search Console

The Performance report gives clicks, impressions, CTR and average position by query, page, country, device and date. Traffic from Google's AI features is counted there too, under the Web search type (Google).

Indexed reference pages

The Page indexing report shows whether the render fix worked.

Queries with a language name

Tells you which language quickstarts to write next.

Time to first successful call

Not a Search Console metric, but the one that tells you whether docs traffic turns into users.

Frequently asked questions

Sources

Checked on September 23, 2026. Search patterns and page advice are our own judgement from running this playbook; we don’t quote search volumes or third-party statistics.

1. Google Search Central: JavaScript SEO basics
2. Google Search Central: How to specify a canonical URL
3. llmstxt.org: The /llms.txt proposal
4. Google Search Central: AI features and your website
5. Search Console Help: Performance report