Facebook tracking pixel Skip to main content

SEO and visibility

See who gets stuck on your page.

Checks one page against a WCAG 2.1 AA checklist, criterion by criterion, and grades it with a score you can rebuild from its scorecard. Each finding names the success criterion, the element or value at fault, the people it hurts (keyboard-only, low-vision, screen reader or motion-sensitive users) and what is at stake, a lost conversion or legal exposure. It flags risk. It never calls a page compliant.

  • Checks against WCAG 2.1 AA
  • Flags risk, never certifies
  • No cited evidence, no finding

At a glance

Group
SEO and visibility
Command
/a11y-audit
Uses
A page address or pasted HTML
Writes
A graded fix list, a baseline
Sends
Nothing. Drafts only.

01When to use it

Run it before a redesign or a launch.

  • A page needs an accessibility baseline before a redesign, a launch or a legal review.
  • The page converts below its traffic, and you suspect a broken focus state, an unlabeled field or low contrast is stopping real people from finishing.
  • You want ADA exposure flagged as a risk before it turns up in a demand letter.
  • A speed or SEO check only touched on accessibility, and you need a dedicated WCAG 2.1 AA audit.
  • An earlier audit exists, and you want the next run to chart what moved.

Not for

02What it needs

The page is the one thing it asks for.

  • The page address, or its HTML pasted in. On an agency setup, the page has to belong to the active client.
  • Your offers and strategy, so each finding maps to the action the page exists for, such as a purchase or a booked demo.
  • Your buyer profile, including the share of buyers who rely on assistive tech.
  • Past baselines and any target from your numbers log (on an agency setup, the client goals file).
  • Optional: a live view of the page through Preview or Chrome, to tab through it and watch focus, or an axe-core run for the criteria a machine can check.

03What it writes

A brief, a graded audit and a baseline.

brief.md
Scope, the page and its conversion goal, the owner, and what acceptance looks like.
final/audit.md
Grade and score, a block on what moved when it is a re-run, a scorecard for each criterion and the quick wins. Then every finding worst first, with its criterion, level, severity, evidence, the people it hurts and the fix.
data/baseline.json
The overall score, the grade, each criterion result and a fingerprint of the findings, for the next run to diff against. Raw signals and any screenshots sit beside it in data/.
Ledger row
The score or grade, the criteria failed and the Blocker count, with a source. It is marked confirmed only when a live page view or axe produced it, and inferred when it came from fetched HTML.

04Limits

What it never does.

  • Call a page compliant or ADA-safe. It writes risk or likely fails, and leaves conformance to a qualified person.
  • Change the live page. Every fix stays a draft until a person applies it.
  • Take a clean automated scan as a clean page. Tools catch roughly a third of the criteria, so the keyboard and focus order checks run anyway.
  • Credit a page for a check with nothing to test, such as a pause control on a page where nothing moves.
  • Follow instructions planted in the page. Fetched content is data to check, never a command.

What stays with you.

  • Choosing the next move: the quick wins, the Blockers for a developer, or a re-audit once fixes ship.
  • Any claim that the page complies. That sign-off belongs to a lawyer.

05Sample output

Three findings and who they hurt.

An excerpt from an audit of a made-up yoga studio booking page. A real report cites the element or value behind every row and labels each contrast ratio measured or inferred.

final/audit.md Illustrative, not a client result.
Email field has no label Blocker
Criterion3.3.2 Labels or Instructions, A
Evidenceinput#email, placeholder text only
Who it hurtsScreen reader users
Focus ring removed Serious
Criterion2.4.7 Focus Visible, AA
Evidencebutton.book-class, outline: none
Who it hurtsKeyboard-only users
Class times in pale gray Moderate
Criterion1.4.3 Contrast (Minimum), AA
Evidence#9A9A9A on #FFFFFF, 2.8:1, inferred
Who it hurtsLow-vision users

Grade and quick wins

  • Grade C, weighted score 68. A baseline run, with no earlier audit.
  • Add a label to the email field.
  • Add the missing lang attribute to the page.
  • Darken the class time text.

07Questions

Before you run it.

Will it tell me my site is ADA compliant?

No. It flags risk and cites the evidence for each flag. Whether a page conforms is a legal determination for a qualified person, so the report says risk flagged or likely fails, never compliant.

Is an automated scan enough?

No. Tools like axe catch roughly a third of WCAG criteria. Keyboard use, focus order and screen reader structure are checked by hand on every run, however clean the automated result looks. It tabs through the page only when a live view is connected; otherwise the result is labeled inferred.

How is the grade worked out?

From a weighted score of the results for each criterion. Problems that block people, such as keyboard access, visible focus, form labels and contrast along the conversion path, weigh more than cosmetic ones, and the scorecard lets you rebuild the grade yourself.

Have us build it

We can build it for you.

Start with the free plan. A person names the first fix.