For most websites, a (manual) accessibility audit covers a representative sample of pages rather than every URL. Auditors typically evaluate between 10 and 30 unique page templates that reflect the full range of components, layouts, and content patterns across the site. The remaining pages are not ignored, they share the same templates and inherit the same issues identified in the sample. The right number depends on site size, template variety, and the depth of interactive functionality. Sampling is the standard approach because it produces accurate WCAG findings without the cost of evaluating thousands of duplicate pages.
| Site Type | Typical Pages Sampled |
|---|---|
| Small marketing site | 5 to 10 pages, often the full site |
| Mid-size informational site | 10 to 20 unique templates |
| Ecommerce site | 15 to 25 templates plus checkout flow |
| Web app or SaaS product | 20 to 40 screens covering core workflows |
| Large enterprise site | 25 to 50 templates across content types |
Why Sampling Works for WCAG Audits
Modern websites are built from templates. A blog has one article template repeated across hundreds of posts. An ecommerce site has one product detail template applied to every SKU. The accessibility issues present on one instance of a template will be present on every other instance.
Auditing 500 product pages would identify the same issues 500 times. Sampling one product page captures those issues once, which is what the remediation team needs to fix the underlying template.
The auditor’s job is to select a sample that represents every distinct pattern on the site. Done well, the sample produces a report that maps to the entire site even though only a fraction of URLs were evaluated.
What Counts as a Unique Page?
A unique page is one with a distinct template, layout, or set of components. Two product pages with different content but identical structure count as one unique page for audit purposes. A homepage, a category page, and a product page count as three.
Auditors look for variation in page templates such as homepage, category, article, contact, and checkout. They also consider interactive components like forms, modals, accordions, carousels, and filters. Media types including video players, image galleries, and embedded maps factor in, as do differences between authenticated and public views. Content patterns also matter, whether the site uses long-form articles, tabular data, or dashboards.
If a component appears anywhere on the site, at least one page containing it should be in the sample.
How Auditors Select the Sample
Sample selection is collaborative. The auditor reviews the sitemap, analytics where available, and the client’s priorities. High-traffic pages, conversion paths, and legally sensitive flows like checkout or account registration are almost always included.
From there, the auditor adds template variety. The goal is breadth, not duplication. A 20-page sample that covers 20 different templates yields far more value than a 50-page sample of near-identical product listings.
For web apps, the unit of analysis shifts from pages to screens or workflows. A SaaS audit might cover login, onboarding, the main dashboard, settings, billing, and three to five core feature workflows.
When Should Every Page Be Audited?
Full-site audits make sense in narrow circumstances. A small site with 15 unique pages and no template repetition is effectively audited in full at the sample stage. Government sites under heightened scrutiny, legal-sensitive properties, or sites with extensive custom-coded pages may also warrant deeper coverage.
For everything else, full-site auditing is cost-prohibitive without producing better results. The same WCAG issues appear in the same templates regardless of how many URLs the auditor opens.
How Sampling Affects Audit Pricing
Audit cost scales with the number of pages and screens evaluated, the depth of interactive functionality, and the WCAG version (2.1 AA or 2.2 AA). A 10-template informational site will cost less than a 30-screen web app, even if the underlying methodology is identical.
This is why auditors ask detailed scoping questions before quoting. Two sites with the same URL count can have very different audit scopes depending on how much template variety exists.
What About Scans?
Automated scans can crawl thousands of pages and flag a narrow set of detectable issues. Scans only flag approximately 25% of issues, which is why they cannot replace a (manual) audit. Scans and audits are separate activities with different purposes. Sampling discussions apply to manual audits, where a human evaluator reviews each page against the full WCAG criteria.
Frequently Asked Questions
How do I know if my audit sample is large enough?
The sample is large enough when it contains at least one instance of every unique template, component, and workflow on the site. If your auditor proposes a sample that excludes a key area like checkout, account creation, or a media-heavy section, push back and add it.
Can pages outside the sample still have issues that were missed?
Pages built from sampled templates inherit the findings in the report. Pages with templates not included in the sample are outside the audit’s coverage. Confirm with your auditor which templates were evaluated so the remediation team knows where the report applies.
Should the sample change for a re-audit?
A re-audit usually evaluates the same sample to verify fixes were applied correctly. If new templates or features were added since the original audit, those should be added to the re-audit scope.
Does sampling work for mobile apps?
Yes. For native apps, the sampling unit is screens and user flows rather than URLs. Auditors select a representative set of screens covering navigation, core tasks, settings, and any unique interactive patterns.
How many pages should be sampled for a small Shopify store?
A typical Shopify audit covers the homepage, a category page, a product detail page, the cart, the checkout flow, the account pages, and any custom landing pages. That usually lands between 8 and 15 templates depending on theme customization.
Sample size is a scoping decision, not a shortcut. The right number is the one that captures every distinct pattern on the site without duplicating coverage of identical templates.
Need help finding an auditor who scopes audits the right way? Contact the Accessibility Base directory to connect with vetted accessibility professionals.