How it works

A URL in, a verdict out, and a fix plan if you want one.

Four steps. Nothing is charged until you have already seen what we found.

1. You send a store URL

That is all we need. No theme access, no staff account, no admin login, no code on your site. We only read pages any shopper can already see.

2. We scan eight templates, twice each

Home, product, collection, cart, search, a content page, account login and a 404, at 1440×900 and 375×812. We run axe-core at a pinned version against WCAG 2.0, 2.1 and 2.2 Level A and AA, and record the version used.

3. You get the verdict, free

Whether the store conforms to WCAG 2.1 Level AA, which success criteria fail and at what level, and how many issues sit in each of the six categories. Plus one named example you can go and check yourself. This costs nothing and there is no obligation.

4. Unlock the full report, if it is useful

$20, $49 or $99, set by the number of confirmed issues you already saw in step three. You get every finding with its page and CSS selector, and a remediation plan with before and after code for each one.

The fix plan

Written to be handed to a person, or to an AI.

Each item in the plan is self-contained: what the problem is, who it blocks, where it is in your store, where it usually lives in a Shopify theme, the before and after code, how to confirm the fix worked, and roughly how long it takes.

That structure is deliberate. You can hand the file to a developer and they can work top to bottom. Or you can open your theme in an AI coding assistant and paste one item at a time, which works because each item already carries your selectors and its own verification step.

The fixes themselves are written by a person and reused across stores. Nothing in your plan is generated on the fly by a model, because a generated fix can be confidently wrong and you would have no way to tell.

The limits, up front

What a scan cannot tell you.

Automated testing detects roughly 30% of WCAG success criteria. That is not a limitation of our scanner specifically; it is true of every automated accessibility tool.

Keyboard operability, focus order, whether a cart drawer traps focus, and whether a screen reader can actually complete a purchase all need a person to sit down and try it. Our report says so on its own front page.

We also separate findings the tool is certain about from findings it cannot settle alone, and we never present the second group as failures or charge you for them.

What you get is an accurate floor, with every confirmed issue located precisely. Not a complete picture, and we will not claim one.

Get the free verdict