Site Audit
How Fond's site audit is made, how to read a finding and its priority, what the coverage notices mean, and what approving or declining a finding does.
A site audit reviews your website for Google search, AI answers and shopping agents. The Fond team runs it, reviews the findings and sends you a link. The audit is not in the sidebar. To open it, you must be signed in as a member of the account that owns the website. Your job is to approve or decline each proposed change.
Approve or decline each finding
Under each finding, click Approve or Decline. Your answer saves when you click, and you can change it later. The count at the top shows how many findings you decided.
After you answer, you can type a note to our team and click Save note. If you change your answer, Fond saves the note with the new answer.
Approving a finding tells our team that you want that change. It does not change your site by itself. Our team sees each decision, its note and its time.
Any member of the account can answer. If two people answer the same finding, the most recent answer counts.
Read a finding
Fond groups the findings by area, with the most important first in each area. Page speed is not part of the audit. The areas are:
- Accessibility
- Content quality and search intent
- Technical SEO
- Information architecture
- Trust, evidence and business identity
- Agent readiness
Each finding shows a title, a priority and three labels:
- The Impact label shows how much the problem is likely to hold you back.
- High: it blocks or seriously harms discovery, indexing, accessibility or purchase across important pages.
- Medium: a clear but contained harm.
- Low: polish.
- The Effort label shows about how much work the change is.
- Small: under two hours, in one or two files.
- Medium: up to a day, or several files.
- Large: several days, or it needs new content, design or a decision from you.
- The Helps label shows who benefits: Google search, AI answers, shopping agents or shoppers.
Below the labels, the finding explains itself:
- the problem, under What it is
- what the problem costs you, and with whom, under Why it matters
- exactly what we would change, under The change
- what to send us first, such as a fact or a business decision, under What we need from you. This part appears only when we need it.
- the affected pages, under Pages, when the finding is about specific pages
- the evidence, page by page, under What we saw
Every finding cites what Fond measured it from. Page paths are links that open that page of your site in a new tab.
How priority works
Fond calculates priority from impact, effort, and how sure we are that the problem is real and the change helps. So it always agrees with the impact and effort shown, also after our team edits them.
The score is impact, times how sure we are (from 0 to 1), divided by effort. Impact counts 3 for high, 2 for medium and 1 for low. Effort counts 1 for small, 2 for medium and 3 for large.
- A high-impact finding that scores at least 1.4 is P0. A large change is never P0.
- Any other finding that scores at least 0.7 is P1.
- Everything else is P2.
So a P2 can be a large problem whose change takes a lot of work, or that we are less sure of.
Areas the audit could not cover
A Not audited notice at the top names an area that Fond could not review in this run. The audit says nothing about that area, good or bad.
If Fond could not display part of the stored audit, the page says so. Ask your Fond contact for the full copy.
How an audit is made
- Fond reads your homepage, the pages in your sitemap, the pages linked from them, your robots.txt and your llms.txt. It runs the Agent Readiness checks too.
- Fond opens one page of each kind, such as one product page, in a real browser at phone and desktop width. This checks accessibility and layout.
- AI models draft findings for each area from what Fond measured. A second model usually drops drafts that the data does not support and tones down overstated claims. Fond usually merges findings about the same problem.
- Before the audit reaches you, our team can edit any finding, including its impact, effort and confidence, or remove it.
What the audit does not show or measure
- Changes that only you or another company can make, such as a platform setting or a pricing change. The audit lists only changes that we can make to your site, also when we first need a fact or a decision from you.
- Every page. Fond reads a limited number of pages, so a problem on a page that Fond did not read does not show up.
- Every accessibility problem. The method cannot reliably find some, such as a keyboard trap or an invisible focus outline.
- Whether assistants mention you. That is AI visibility, on the Visibility page.
- Changes over time. An audit describes your site when Fond read it and does not update. A new audit comes with a new link. To keep watching the checks that AI agents depend on, use Agent Readiness, which Fond runs again on its own.
Agent Readiness
What Fond's Agent Readiness checks measure, how the passing count is worked out, what each check status means, how often checks run, and how to fix a failing one.
MCP Connector
Connect Fond to Claude or any MCP-capable AI tool and ask about your brand-visibility data in plain language. Your analytics, without the dashboard digging.