Private beta for SEO and marketing teams
Can read your website?
Your website can say a crawler is allowed while its security settings still block the request. Fieldnote checks what each crawler receives and shows what is missing. The report is written for marketing teams, not just developers.
Check crawler access, late-loading content, schema and trust details before they affect how your company is described.
Join the private beta
Tell us which website you want checked. Joining the waitlist does not start a crawl.
Published crawler instructions
All four crawlers are allowed
Site security can still stop the request before the page is delivered.
Choose a crawler to see what it received.
Try the exampleClaude search crawler
Security challenge
The request reached a security check instead of the homepage content.
Next step: Ask the web team to review the site security rule for this crawler.
Example result from Fieldnote’s network. It does not recreate a vendor’s private crawler network.
Example based on a completed audit. Site names and page addresses are hidden.
Check the patterns, not 100 copies of the same page
This example maps the full site, then studies the page groups that can reveal a shared problem.
1,411
Page addresses mapped
17
Representative pages checked
80
Delivery and page-label checks
49
Trust questions reviewed
What Fieldnote adds
Tracking shows where your brand appeared. Fieldnote checks the site behind it.
A visibility report can show a lost mention or citation. Fieldnote checks whether your own pages were reachable, complete, clearly labelled and supported by visible trust details.
“Your crawlers are allowed”
Which crawlers received the page, and which were blocked or challenged
A long list of page errors
The few page types that matter, why each was checked, and how many similar pages may be affected
“This page uses JavaScript”
What arrives immediately compared with what appears only after a browser finishes loading
A single trust score
The author details, proof, sources, dates and business facts found or missing on each page type
A list of technical fixes
What happened, who should take it, what to do next and how to check the fix
Fieldnote puts these checks in one report: who reached the page, what content arrived, which product or author details were clear, what signs of trust were present and what to fix.
Inside the audit
Five questions the audit answers for your team.
For each question, the report names the page checked, shows the result and explains the next step.
01
Which pages matter?
Group similar pages and choose one useful example from each group.
02
Can crawlers get in?
Compare published instructions with the page each test receives.
03
Does the message arrive?
Check what arrives first and what the browser adds later.
04
Are facts clearly labelled?
Check the hidden labels for products, articles, prices and authors.
05
Is there proof?
Look for sources, people, dates, claims and business details.
01 · Pages checked
Check each page type once. Fix the pattern everywhere.
Fieldnote first works out what kind of website it is. Then it groups similar pages, including articles, service pages, product pages and category pages. It explains why each example was chosen and how many other pages use the same layout.
Page groups selected
17 examples chosen from 1,411 page addresses
Choose a page group to see why it was included.
Insight articles
185 pages
Why chosen: This page uses the same article layout as 184 other pages.
Signals: Article address, author area and shared content layout
Checks: Author details, sources, dates, page labels and content arrival
91% describes confidence in the page group, not the page’s quality.
What shoppers see
What the page tells machines
Choose a page detail to inspect.
Price
Missing
Customers see: $49 per month
Page label says: No price supplied
AI search has to infer a price that the page could state directly. Next step: Add the same price to the product label and visible page.
02 · Page labels (schema)
Tell AI search exactly what each page contains.
Schema markup gives machines explicit page details. It can identify a product, article, author, price or date even when the page design is complex.
Fieldnote checks that the right details exist, match what a visitor sees and arrive in the first version of the page.
03 · What arrives first
Does the full page arrive immediately, or does the browser add it later?
Some websites send the headline, copy, links and product facts immediately. Others send a mostly empty shell and rely on the visitor’s browser to add the useful content later.
The first approach is often called server-side rendering (SSR). The later browser work is client-side rendering (CSR). Both can work. The audit checks whether a crawler receives the copy, links and page details that matter.
Homepage check
Important copy arrives late
What the server sends first
No main page copy was found. The headline, product details and main links appeared only after the browser loaded.
Next step: Include the essential copy in the first version of the page.
Case study · Reviewed by Fieldnote AI
Choose a trust check to see the finding and proof.
Missing
Who created or reviewed this case study?
- How sure we are
- 98%
- Team to involve
- Editorial
This percentage describes how clear the page evidence is. It is not a visibility score.
Found
A named client, project scope and measurable outcomes.
Missing or unclear
An author or reviewer with a relevant role and profile.
Why it matters
The work may be credible, but the page does not show who stands behind it.
Suggested change
Add the contributor’s name, role and profile to the shared case-study layout.
04 · People and proof
Can readers see who created the page and why they should trust it?
Fieldnote uses AI to review each selected page. For an article, it looks for an author, relevant experience, sources and dates. For a product or service page, it checks the proof behind claims, useful comparisons, buying details and disclosures.
The report uses three statuses: Found, Missing and Could not confirm. These checks cover the practical parts of E-E-A-T, and every result links back to the page evidence.
Article page · Visual evidence
Switch between the current page and a suggested change.
What we found
Problem marked on the pageInsights · 8 minute read
A practical guide to enterprise AI migration
No visible author or reviewer
The article makes useful claims, but no author or reviewer appears where readers would expect to find one.
Moving established systems to an AI-ready stack takes more than a platform decision. This guide covers the people, data and operating choices that shape the migration.
Why this was flagged
Readers cannot quickly tell who is responsible for the advice or whether a subject expert checked it.
This is a discussion example. Fieldnote does not change the live website.
05 · Screenshot and suggested change
Show your web team exactly what needs to change.
Screenshots show exactly where evidence is missing or unclear. When a visual suggestion helps, the report pairs the capture with a simple proposed layout your team can discuss.
The mockup gives the team a starting point for the fix. It does not change the live website.
What you receive
A clear brief for marketing and web teams.
The report does not stop at a score. Each issue names the pages that may be affected, the team to involve, the next step and the check that proves the change worked.
- 01The page groups reviewed and why each example was selected
- 02How sure we are that each page represents its wider group
- 03What the server sends first versus what the browser adds later
- 04Published crawler instructions beside what each crawler test actually received
- 05Trust evidence marked Found, Missing or Could not confirm
- 06Behind-the-scenes page labels for products, articles, authors, prices and dates
- 07How many similar pages may share the issue and what that means for marketing
- 08The team to involve, the next action and a clear way to check the fix
- 09A page capture with evidence highlighted and, where useful, a proposed fix mockup
- 10The source page, relevant excerpt and test result behind every finding
Questions
What is included in the beta?
What happens after I request access?
We are inviting teams in small groups while the crawler is in private beta. We will review the website you submit, then email you when a beta place and audit workspace are available. The public site does not start a crawl.
Does Fieldnote crawl every page?
It maps the site, groups pages that share the same purpose and layout, then selects a small number of examples. The report is clear about which page was checked and how many similar pages may share the same issue.
Why check more than the crawler rules file?
The crawler rules file says who should be allowed. A separate site security layer can still redirect, challenge or block a request. Fieldnote makes limited requests using published crawler names and records the result. This shows what happened from Fieldnote’s network, not what happens on a vendor’s private network.
What does the trust review look for?
The questions change with the page. An article is checked for an author, relevant experience, sources, dates and editorial responsibility. A product or service page is checked for support behind claims, useful comparisons, clear commercial details and disclosures. AI helps review the supplied page evidence, but every finding remains traceable to what was found or missing.
Is this an AI visibility tracker?
No. Visibility trackers show whether a brand appeared in an answer. Fieldnote checks the website itself. It tests whether important pages can be reached, whether the useful content arrives, whether page facts are clear and whether trust details are present.
What does the waitlist store?
The waitlist stores your work email, the public website origin you submit, the signup source, submission timestamps and a repeat-submission count in a private Cloudflare D1 database. There is no public signup list. The data is used only to review beta requests and contact you about access; it is not published or sold. Records are kept while the beta is being developed or until you ask for removal. Joining does not start a crawl or store crawler response bodies. If you later join an audit, limited public-page evidence may be sent to OpenAI for E-E-A-T review with response storage disabled. Do not submit private or confidential URLs.
Find out what crawlers receive from your site
See what AI crawlers may be missing.
Join the private beta for a report covering crawler access, page types, content delivery, schema and trust checks. The most important changes come first.