Skip to content

Notion Sites SEO: search engine indexing and what crawlers get

Notion Sites SEO explained: the search engine indexing toggle, paid-plan title and description, custom domains, and a curl test of what crawlers receive.

By Crawlito · 8 min read

In short

  • Notion says a Site becomes searchable on the web when you select Search engine indexing and turn on Discoverable on the web, under Share, then Publish.
  • Notion says a Site can take up to four weeks to be indexed and appear in search results.
  • Editing the page title and description used for SEO and share previews requires a paid Notion plan.
  • A custom domain is a paid-plan add-on at $10 per month per domain, or $8 on annual billing, and must use a subdomain such as www pointed at external.notion.site.
  • Notion does not document how much page content is in the raw HTML of a published Site, so test it with curl before assuming AI crawlers can read it.
On this page

Notion Sites SEO comes down to one switch and a few paid-plan fields. A published Notion Site appears in search engines only after you select Search engine indexing and turn on Discoverable on the web in the publish settings, and Notion says indexing can then take up to four weeks. Paid plans add an editable title and description, a custom domain add-on and Google Analytics. What Notion does not document is how much of each page arrives in the raw HTML, which is what most AI crawlers read, so that part you test yourself.

This post covers the documented settings, what is missing from them, and a curl check that shows what a non-rendering crawler receives from your Site.

Everything below comes from Notion's help pages on publishing, site customization and availability and pricing and custom domains, at the time of writing.

SettingWhere it isPlanWhat it does for search
Search engine indexingShare, Publish, Search engine indexing, toggle Discoverable on the webFree and paidMakes the Site searchable on the web
Page title and descriptionShare, Publish, alongside Search engine indexingPaidUsed in share previews and for SEO
Share preview imageSite customization, Share preview, Upload imagePaidThe preview uses the Site's cover image and title unless you upload your own image
Custom slugPublish settingsPaidLetters, numbers and hyphens, up to 60 characters
notion.site domainSettings in the sidebarOne on Free, up to five on paidYour Site's address if you have no custom domain
Custom domainSettings, Sites, then New domain next to DomainsPaid plus add-onServes the Site on your own hostname
Google AnalyticsSite customization, Google AnalyticsPaidAdds your measurement ID

Two details from the same pages matter for structure. Publishing a page publishes its subpages by default, and slugs do not carry over to subpages: Notion says home pages cannot have slugs and a slug set on one page is not inherited by the pages under it. Edits to page content update the live Site automatically, while customization changes go live after you select Publish changes.

How to turn on search engine indexing

  1. Open the page you published as a Site.
  2. Select Share, then Publish.
  3. Select Search engine indexing and toggle on Discoverable on the web.
  4. Select Publish changes if you also edited customization settings.

Notion's publishing guide states that "Notion Sites can take up to four weeks to be indexed and appear in search results." Treat that as Notion's estimate for search engines generally; it is not a promise from Google or Bing, and Notion does not say the toggle submits your Site to any search engine.

The help pages do not say what the toggle changes technically. It could be a robots meta tag, an X-Robots-Tag header, robots.txt rules or something else. That matters because robots.txt and meta robots behave differently: robots.txt stops crawling, while a noindex tag lets a crawler fetch the page and then keeps it out of the index. You can see which one your Site uses with the commands in the next sections.

What Notion does not document

At the time of writing, Notion's help pages on publishing, customization, custom domains and site management do not cover:

  • editing robots.txt, or what the default robots.txt contains;
  • an XML sitemap, or whether one is generated;
  • a per-page noindex setting separate from the Site-wide toggle;
  • canonical tags, or which URL is canonical when the same Site is reachable on notion.site and a custom domain;
  • redirects from an old notion.site address to a custom domain.

On that last point, the Manage your Notion Sites page warns that after you rename a notion.site domain, links using the previous domain "will no longer work". It does not mention a redirect, so plan a domain change as you would any site migration: expect old links to break and update the ones you control.

Because these are undocumented, do not guess. Fetch them from your live Site and record what you find.

Custom domains: what the setup requires

Notion's custom domain guide sets out the rules:

  • Only workspace owners on paid plans can connect a domain, and each domain needs the custom domain add-on: an additional $10 per month, or $8 per month with annual billing. You can connect up to 25 domains.
  • Notion does not sell domains. You buy one from a registrar first.
  • You must include a subdomain. Notion suggests www and recommends redirecting the root domain to it at your DNS provider.
  • You add a CNAME record pointing to external.notion.site. and a TXT record Notion gives you, then select Verify.
  • On Cloudflare, Notion says to turn off proxy status for the record.

The DNS records look like this:

www.example.com.   CNAME   external.notion.site.
www.example.com.   TXT     (value shown in Notion's domain settings)

The root redirect is your job: only the subdomain points at Notion, so an apex without a redirect leaves anyone who types or links the bare domain, crawlers included, without your Site. After setup, confirm the redirect with curl:

bash
curl -sI https://example.com/ | grep -iE "^(HTTP|location)"

You want a 301 or 308 with location: https://www.example.com/. Google's redirects guide treats both as permanent and uses them as a signal for which URL is canonical.

A custom domain also unlocks one cosmetic control: Notion's customization page says the Notion watermark (the "Built with Notion" mark in the header settings) can only be turned off when the Site uses a custom domain.

How to test what crawlers receive with curl

Google renders JavaScript. Its JavaScript SEO guide describes crawling, rendering and indexing as three phases and notes that "not all bots can run JavaScript". OpenAI, Anthropic and Perplexity do not document JavaScript rendering for their crawlers, so the safe assumption is that an AI crawler gets the raw HTML only, as covered in raw vs rendered HTML. Notion does not say whether a published Site sends its text in the first response or builds it in the browser, and the answer may change as Notion updates the product. So test your own page.

Step 1: check the response and robots signals

bash
URL="https://www.example.com/your-page"

curl -sIL "$URL" | grep -iE "^(HTTP|location|x-robots-tag|content-type)"
curl -s "https://www.example.com/robots.txt"

Look for the final status code, any X-Robots-Tag header, and what robots.txt says. Run the same commands once with Discoverable on the web on and once with it off: the difference shows you how Notion implements the toggle on your Site.

Step 2: save the raw HTML as a crawler would get it

bash
curl -sL --compressed \
  -A "Mozilla/5.0 AppleWebKit/537.36 (KHTML, like Gecko); compatible; GPTBot/1.4; +https://openai.com/gptbot" \
  "$URL" -o notion-raw.html

grep -io '<meta name="robots"[^>]*>' notion-raw.html
grep -io '<title>[^<]*</title>' notion-raw.html
grep -io '<meta name="description"[^>]*>' notion-raw.html

The user agent string is OpenAI's documented example for GPTBot; the version may change. The output tells you whether the title and description you set in Site customization reach crawlers, and whether a robots meta tag is present.

Step 3: search for your body text

Copy a plain sentence from the middle of the page, without quotes or symbols, and search for it:

bash
grep -c "a plain sentence from the middle of your page" notion-raw.html

Then check one sentence from a toggle block and one from a linked subpage's title. The full walkthrough, including headers and redirects, is in how to see your page as an AI crawler with curl.

How to read the result

What you find in notion-raw.htmlWhat it meansWhat to do
Body text, headings and links presentNon-rendering crawlers can read the pageSpot-check other page types, such as databases
Title and description present, body missingCrawlers get metadata but not contentTreat the body as invisible to most AI crawlers
Almost no text, mostly script tagsThe page is built in the browserSee the empty HTML problem
A robots noindex tag or headerThe page asks not to be indexedCheck the Discoverable on the web toggle

Test more than one template. A plain text page, a page with a database view and a page with toggles can behave differently, and Notion publishes subpages along with their parent.

When Notion Sites is the wrong tool

If the test shows your text is missing from the raw HTML and AI search matters for that content, the settings above cannot change it: Notion controls the rendering. The practical options are to keep Notion for internal or low-stakes pages and move key pages, such as pricing, product and documentation, to a platform where you control the HTML, robots.txt and sitemap. For help content specifically, knowledge bases for AI search covers what a public answer page needs once it is reachable.

If you want a Site kept out of search, leave Discoverable on the web off and confirm it with Step 1. Notion's publishing page also notes that a page whose general access is set to Anyone on the web with link can stay reachable after you unpublish the Site, so check general access too. Not being indexed is not the same as being private: at the time of writing, Notion's FAQ says pages cannot be password protected.

Frequently asked questions

Are Notion Sites indexed by Google by default?

Notion's instructions say to select Search engine indexing and turn on Discoverable on the web to make a Site searchable, so check that the toggle is on. After that, Notion says indexing can take up to four weeks.

Can I edit robots.txt or add a sitemap on Notion Sites?

Notion's help pages do not document either at the time of writing. Fetch /robots.txt on your Site to see what is served. If you own a custom domain and verify it in Search Console, you can request indexing of individual URLs with the URL Inspection tool, which Google says does not guarantee indexing.

Do I need a paid plan for Notion Sites SEO?

The indexing toggle is available on the Free plan. Editing the SEO title and description, the share preview image, Google Analytics and a custom domain require a paid plan, and the custom domain also needs the add-on.

Can AI crawlers read my Notion Site?

That depends on whether your page's text is in the raw HTML, which Notion does not document. Run the curl test above on each type of page before relying on it.

Can I use my root domain with Notion Sites?

Notion requires a subdomain such as www and recommends redirecting the root domain to it at your DNS provider.

Check one published page

After you turn on indexing and connect your domain, run a key page through the free check to see whether the main AI crawlers can fetch it and what the raw HTML contains. It needs no account and gives you a second view alongside your own curl results.

Filed under Platform guides. Topics: notion, noindex, raw-html, rendering, ai-crawlers