--- title: "What It Actually Costs to Fix an ADA Problem on Your Website" url: https://guavawebcompliance.com/articles/ada-website-compliance-cost description: "What does ADA website compliance cost? We scanned 31 pages across nine small-business sites: a median of 71% of a site's detected failures sat in a single rule." published: 2026-09-17 updated: 2026-09-17 --- **Nobody can quote what it costs to make a website ADA compliant without knowing how much work is on the site, so we measured the work on nine of them.** On 31 pages across nine ordinary small-business websites, a median of 71% of the accessibility failures our scanner detected on a site sat in one rule, and each site's top three rules covered a median of 100% of them. The pattern is a few mistakes cut into a template and printed on every page that uses it. Work that repeats like that can be counted, and counted work can be priced up front. One of the nine told mobile browsers to turn zooming off. The same viewport tag shipped on all three pages we scanned there, and the scanner returned the same line all three times: *"user-scalable=no on `` tag disables zooming on mobile devices."* [Deque](https://dequeuniversity.com/rules/axe/4.10/meta-viewport), which publishes the rule our scanner runs, rates that critical, maps it to [WCAG 1.4.4 Resize Text](https://www.w3.org/TR/WCAG21/#resize-text), and calls it "problematic for people with low vision who rely on screen magnifiers." The fix is one deletion. One plate, one correction, every page at once. ## How much work is on an ordinary small-business site? The nine came off the "Restaurants, Food and Drinks" category of the [Portland Metro Chamber's public member directory](https://community.portlandmetrochamber.com/membership-directory/Search/restaurants-food-and-drinks-249449), independent businesses that each hired whoever they hired. All nine had at least one detected violation. The median site failed three distinct rules that automation can detect. Four of the nine had their entire detected scope in one or two. Defects concentrate because templates repeat. A mistake in a header, a photo slider or a form is published on every page that uses it, so the scan reports it once per page and a developer fixes it once. The most widespread defect in this run was a link with no [accessible name](https://guavawebcompliance.com/articles/what-is-an-accessible-name), on six of the nine sites. Two of the nine failed on the same control: the site search box, with nothing naming it for a screen reader. ## So what does the fix cost? Your number depends on your site. Our free report counts the instances it detects on the pages it scans and prices the remediation off that count, so the price arrives on the report instead of at the end of a discovery call. [Run your site through it](https://guavawebcompliance.com/free-scan): about a minute, no login, no credit card, no sales call. ## What does that price leave out? A scan finds a floor. Nine of the 31 pages came back with zero detected violations, and five of those carried findings the scanner handed to a person to decide. Six more pages returned an error or timed out, and those count as unknown, never as passes. "Three detected rules" means three rules a machine can catch on the pages it reached, and automation reaches part of WCAG 2.1 AA. What it detects is what anyone can price in advance. The rest takes a person sitting down with the site, whoever you end up hiring. --- ## The fine print - **Where the scan numbers come from.** A run of 38 pages across 10 businesses on 2026-09-08, selected by taking every 8th surviving listing in the directory's own order, with the exclusion rules and the sample fixed before any site was opened. 31 pages across 9 businesses produced usable results. Six pages returned an error body or timed out and are counted as attempted and reported as unknown. A seventh was the same document as another page on the same site and was dropped so its findings were not counted twice. The sample, the full 217-card frame with each listing's disposition, and the raw results are kept with the run in our research repo, which is not public; the checkable part is the method above. - **This is not a prevalence rate.** One chamber, one category, one metro, one date, nine businesses with data. It says nothing about how often restaurants, small businesses or Portland sites fail these rules. - **Medians, per business.** One gallery page in the run carried more than half of all the instances we found, which skews any corpus-wide percentage. The figures above are medians across the nine businesses, which that page cannot move. - **No business is named, on purpose.** The unit here is the aggregate. - **The viewport finding is the scanner's own message and count.** Our tool records a truncated serialization of the page element, not the bytes the server sent, which is enough to characterize what it found and not enough to quote as markup. - **A scan is a floor, never a verdict.** Automated testing reaches only part of WCAG 2.1 AA. Keyboard traps, focus order, whether alt text is *meaningful*, and form error recovery need a person. Zero detected violations is not a passing page: five of the nine clean pages here carried findings flagged for human review. - **How our price is built.** The report prices the violation instances it detected on the pages it scanned. Findings the scanner cannot decide are listed separately and left out of both the count and the estimate, so the estimate covers less than the site's real work. - **We sell ADA remediation.** This is our own scanner, our own data, and our own interest in you having a problem. The method is stated in full, which is the part you can argue with. - **Courts reference WCAG; no binding holding adopts it.** "ADA compliant" is a legal conclusion, not a measurement. We're not lawyers, and this isn't legal advice. ### Sources - **Deque**, axe-core 4.10 rule `meta-viewport`, "Zooming and scaling must not be disabled," mapped to WCAG 1.4.4 (AA) — https://dequeuniversity.com/rules/axe/4.10/meta-viewport - **W3C**, Web Content Accessibility Guidelines (WCAG) 2.1, Success Criterion 1.4.4 Resize Text (Level AA) — https://www.w3.org/TR/WCAG21/#resize-text - **Deque**, axe-core 4.10 rule `link-name`, "Links must have discernible text" — https://dequeuniversity.com/rules/axe/4.10/link-name - **Deque**, axe-core 4.10 rule `label`, "Form elements must have labels" — https://dequeuniversity.com/rules/axe/4.10/label - **Portland Metro Chamber**, public member directory, category "Restaurants, Food and Drinks," captured 2026-09-08 — https://community.portlandmetrochamber.com/membership-directory/Search/restaurants-food-and-drinks-249449 - Guava scan campaign, chamber member sites, 2026-09-08 — Cloudflare Browser Run + axe-core 4.10.3, ruleset `wcag2a, wcag2aa, wcag21a, wcag21aa`. Findings, pre-registered sample and raw results held in our research repo.