For Claude, our site had one page. By the evening, five.
On the morning of 25 September 2026, as far as Claude's web search was concerned, Itria's website had a single page. By the evening, after we asked Brave to recrawl seven of them, it had five.
Claude's web search also runs through Brave. Anthropic, in its list of providers, names Brave Search for web search across all products, along with TurboPuffer. So what Brave's index knows about a site is part of what Claude can find.
From one page to five, in a day. With the same questions to Claude, itria.io appeared in the search results 4 times out of 8 in the evening, against 2 out of 8 in the morning, and with 5 different pages instead of one. In between, we asked Brave to recrawl seven of the site's addresses.
Cited: still zero times. Appearing among the results doesn't mean being cited in the answer: in both measurements Claude didn't cite itria.io. And a single pair of measurements doesn't prove the cause: we report it as an observation.
This is part of Itria's lab: things measured on our own site, with the real numbers, to understand how search engines and assistants read it. The service that grows out of it is visibility in search engines and AI assistants.
Why Brave matters for Claude
When Claude searches the web, it uses external providers. In Anthropic's Trust Center, the list of sub-processors, Brave Search appears with the function “Web Search” across all products, and alongside it is TurboPuffer, with the same function. A site that Brave's index barely knows is less likely to appear among the results Claude reads.
“Brave Search. Products: All products. Web Search” (Anthropic, Trust Center, list of sub-processors, read on 25 September 2026)
The method: nine questions to Claude, with web search
The probe asks Claude, with web search switched on, a fixed set of questions: five on food exports, two about choosing a supplier, one about the brand and one site:itria.io search. For each answer it records whether itria.io appears among the search results and whether it's cited. One run costs between $1.30 and $1.80 in API usage.
We ran it twice on the same day, 25 September 2026: at 11:23 and at 17:44 UTC. Between the two, after the first measurement, we asked Brave to recrawl seven of the site's addresses through its submission page: the home page, EX-1, the spec sheet template, the export guide, services, the Cruscotto and the English version.
Before and after
On the eight questions common to both measurements, itria.io appeared among the search results in 2 answers in the morning and in 4 in the evening. The number of different pages from the site seen in the results went from one, the Switzerland page, to five. For the brand question, no itria.io page came up in the morning; in the evening, two did.
| Question | Morning, 11:23 | Evening, 17:44 |
|---|---|---|
| Documents for exporting food to Switzerland | Switzerland page | Switzerland page |
| Who automates foreign buyers' enquiries with AI | No pages | Services |
| What Itria (itria.io) is and what it does | No pages | Services, spec sheet template |
| site:itria.io search | Switzerland page | Home, EX-1, Switzerland page |
| The other four questions | No pages | No pages |
| Citations in the answer | 0 | 0 |
How far this data can be trusted
Not very far, and we say so. It's a single pair of measurements, six hours apart. Web searches don't always return the same results for the same question. Two questions changed between the runs, which is why we compare only the eight they had in common. And Brave might have recrawled the site anyway, without our request: the timing doesn't prove the cause.
With those caveats, what the data shows is a direction: three of the seven pages submitted to Brave, the home page, EX-1 and the spec sheet template, appear in the evening results, along with services. Citations stay at zero. The next step is to run the same probe again days later, with the same questions.
The check, on your own site
The simplest check is on Brave's search: search for site: followed by your domain and count the pages that come up. If there are one or two on a site with dozens, Brave's index barely knows you. Brave has a public page for requesting a recrawl of an address, used one page at a time.
Questions and answers
Which search engine does Claude use for web search?
Anthropic, in its Trust Center, lists Brave Search among its web search providers, for all products, along with TurboPuffer. So what Brave's index knows about a site is part of what Claude can find.
The list was read on 25 September 2026.
Does asking Brave to recrawl your pages help you show up in Claude?
In itria.io's case, on the same eight questions to Claude, the site appeared among the results 4 times in the evening against 2 in the morning, and with 5 different pages instead of one, after we asked Brave to recrawl seven addresses.
It's a single pair of measurements and searches aren't deterministic: the cause isn't proven, and citations in the answers stayed at zero.
Does appearing in Claude's search results mean being cited?
No. Claude reads the search results and then chooses what to cite in its answer. In the two itria.io measurements of 25 September 2026, the site appeared among the results up to 4 times out of 8, and was never cited.
They're two different steps: getting found by the index, and giving the best answer to the question.
How do you check how many pages of your site Brave knows about?
Search Brave for site: followed by your domain and count the pages that come up. If there are one or two on a site with dozens, Brave's index barely knows the site.
Brave has a public page for requesting a recrawl of an address, to be used one page at a time.
How does Itria's probe of Claude's web search work?
It asks Claude, with web search switched on, a fixed set of questions: five on food exports, two about choosing a supplier, one about the brand and one site:itria.io search. For each answer it records whether itria.io appears among the results and whether it's cited.
One run costs between $1.30 and $1.80 in API usage, and it's repeated with the same questions to compare measurements.
Notes on sources
- Anthropic, Trust Center, sub-processors: Brave Search and TurboPuffer with the Web Search function. Read on 25 September 2026.
- Brave, page for requesting a recrawl of an address.
- Itria measurements: two runs of the probe on Claude's web search, on 25 September 2026 at 11:23 and 17:44 UTC; a recrawl request to Brave for seven addresses after the first run. The raw data is in our working archive.
- A single pair of measurements, with non-deterministic searches and two questions changed between runs: the data points in a direction and doesn't prove a cause.
See your own site the way search engines and assistants read it.
How many of your pages Brave knows, whether your email address can be read, whether your site answers addresses that don't exist: these are things that can be measured from the outside. Drop us a line about what's slowing you down: we'll make the first move, even if we never end up working together.