Website accessibility cost calculator

Christopher Ross

6 min read

AI and learning, kept human · Niagara, Ontario

You have been told the website needs to be accessible, and now you need a number for a budget line. The quotes you have seen range from a few hundred dollars to six figures, which is not a range, it is a shrug. The spread is real, but it is explainable, and most of it comes down to one thing nobody puts in the quote.

The calculator below estimates the audit, remediation, document work, and a retest, and it does it the way the work actually runs: the audit covers a representative sample, and the fixing concentrates in a handful of templates. Enter your page count, your page types, and how interactive the site is, and it returns a budget band you can defend in a meeting. The band assumes a conformance target of WCAG 2.1 AA, the version of the Web Content Accessibility Guidelines that most Canadian and American obligations currently point at.

Estimate what it costs to audit, remediate, and validate a website against WCAG 2.1 AA, using published market rates and a sample-based scope.

Your site

A rough count is fine. Check your sitemap if you are not sure.

Home, standard page, blog post, listing, contact form, and so on. Most sites have between 5 and 12.

Interactive components are where the expensive defects live.

An honest answer here changes the number more than anything else on this form.

Scope and delivery

Documents a member of the public is expected to read. Zero if none.

The audit needs specialist testing. Many of the fixes are ordinary front-end work.

How this is calculated

Nobody audits every page. The scope is a representative sample: every distinct template plus a slice of content pages. Remediation is then weighted toward templates, because fixing a template fixes every page built on it.

Audit sample = every distinct template + about 10 percent of content pages, with a floor of 5 and a ceiling of 40.
Audit cost = sample screens x the market rate per screen, adjusted for interactive complexity.
Remediation = template fixes at full rate + remaining pages at a reduced content-only rate.
Validation retest = about a quarter of the audit cost.

Market rates are published North American ranges: roughly 100 to 250 US dollars per screen to audit, and 350 to 550 US dollars per page to remediate. Converted at the Bank of Canada rate of 1.4068 on August 4, 2026 and rounded to planning figures.

Sample sizing and the template-leverage discount are my own planning assumptions, not published benchmarks. They reflect how remediation actually runs: most of the fix lands in a handful of templates.

This is a budget range for a business case, not a quote. A real quote follows a scoping conversation and a look at the actual site.

Why not just audit every page?

Because you would be paying to find the same eight problems two hundred times. Almost every accessibility defect lives in a template: the navigation, the header, the card component, the form. Audit one page of each type and you have found the issues that affect every page of that type. A vendor quoting per-page across your whole site is either misunderstanding the work or pricing for it.

Does an accessibility overlay or widget do this for less?

No. Overlays are a JavaScript layer that claims to fix accessibility at runtime, and they do not. They have been the subject of hundreds of lawsuits in the United States, they are opposed by essentially every disability advocacy organization, and screen reader users routinely report that they make sites harder to use rather than easier. Money spent on an overlay is money not spent on the fix, and it does not satisfy an AODA or Accessible Canada Act obligation.

Can our own developers do the remediation?

Often yes, and it is usually cheaper. The audit is the part that needs specialist skill and assistive-technology testing; many of the fixes are ordinary front-end work once someone has written down precisely what is wrong. Select in-house in the tool and the remediation portion drops to your own cost of time. What does not change is the retest: someone independent should confirm the fixes actually landed.

What about PDFs?

PDFs are priced separately because they are separate work, and for many public sector organizations they are the larger problem. A simple text document remediates cheaply. A scanned form, a multi-column report, or anything with tables costs considerably more per page than a web page does. If you have thousands of documents, the real strategy conversation is which ones still need to exist as PDFs at all.

Is this the same as an AODA or Accessible Canada Act obligation?

This tool costs the work. Whether you are legally obligated, by when, and under which statute is a different question, and the Canadian web accessibility deadline lookup on this site answers it. If the contract will be publicly tendered, the government web procurement threshold lookup tells you whether the value crosses a trade-agreement threshold.

This is a budget range for a business case, not a quote. A real quote follows a look at the actual site. I am not a lawyer and this is not legal advice about whether you are obligated; that is a separate question, and there is a separate tool for it.

The thing nobody puts in the quote

You do not audit every page. You audit one of each kind. It is closer to a building inspection than a census. The inspector checks one unit on each floor plan, because units built from the same drawing have the same problems.

Almost every accessibility defect on a normal website sits in a template: navigation, header, card components, forms, the footer, a modal. Audit one blog post and you have found the issues on all four hundred of them. Fix the template and you have fixed all four hundred at once. What is left after that is content-level work, alt text and heading order and link text, and that is genuinely per page but it is a fraction of the cost and most of it can be done by whoever writes the content.

This is why a per-page quote across an entire site should make you cautious. It prices your sitemap, and your sitemap is not where the difficulty lives. Ask any bidder how many screens they intend to audit and why that sample, and listen for whether the answer walks you through your templates or your page count.

Why interactivity moves the number so much

A brochure site of text and images is mostly a content problem, and content problems are cheap to fix once someone has written them down. A site with forms, filtering, menus, and pop-up dialogs is a different job, because those are the parts where someone navigating by keyboard gets stuck, or where the page changes and a screen reader never announces it. An application-like interface with dashboards, booking, or a checkout is different again: custom widgets need custom testing, and the errors are the kind that only show up when someone actually tries to complete a task with a screen reader.

That is the single choice on the form that moves your estimate the most, so answer it honestly. If you are not sure whether your site counts as interactive, the test is simple: how many things on it can a visitor do, as opposed to read.

Overlays are not a cheaper version of this

If you have been offered an accessibility widget or overlay for a few hundred dollars a year, it will not do what this budget is for. Overlays are a script that claims to repair accessibility in the browser at page load, and they do not achieve conformance. The Overlay Fact Sheet, signed by hundreds of accessibility practitioners and disabled users, sets out the case against them in detail, and screen reader users regularly report that overlays make sites harder to use. Sites running an overlay have still been named as defendants in American accessibility lawsuits, which is the opposite of what the product is sold to prevent.

Both the Accessibility for Ontarians with Disabilities Act (AODA) and the Accessible Canada Act point at WCAG conformance, and an overlay does not produce conformance. Your own counsel should tell you what that means for your organization; for budgeting, plan as though the overlay is not installed. If you already have one, I am not judging, and buying it mostly bought a year of believing the problem was handled. If the budget is tight, the honest move is to fix the templates first and phase the rest, because the script changes nothing underneath.

Which numbers here are published and which are mine

The per-screen audit rate and the per-page remediation rate are published North American market ranges, roughly 100 to 250 US dollars per screen to audit and 350 to 550 US dollars per page to remediate, converted to Canadian dollars at the Bank of Canada rate of 1.4068 on August 4, 2026 and rounded to planning figures, which lands at roughly 140 to 350 Canadian dollars per screen to audit, and 490 to 775 per page to remediate.

The rest is mine and the tool labels it as such: how the sample is sized, the reduced rate for pages that share a template that has already been fixed, the complexity adjustment, the discount when your own developers do the fixing, and the retest at about a quarter of the audit. Nobody publishes those, because they come from doing the work. I would rather tell you which half of the model is market data and which half is my judgment than present the whole thing as though it were sourced.

One line I keep separate on purpose is the retest, and it stays priced independently even when your own team does the remediation. The entire value of a retest is that the person checking is not the person who fixed it.

The two questions this tool does not answer

This one costs the work. It does not tell you whether you are legally obligated, by when, or under which statute; the Canadian web accessibility deadline lookup does that. And if you are a municipality, school board, college, or hospital, it does not tell you whether the contract value has to be openly tendered; the government web procurement threshold lookup does that.

Run in that order and you have the three things a business case needs: the obligation, the cost, and the procurement route. Everything runs in your browser and nothing you type is stored or sent anywhere, which matters when the figure you are testing is a budget nobody has approved yet.

If the band is wider than your budget, the first conversation is about sequencing: which templates carry the most risk, and what a phased year looks like. I build to WCAG 2.2 AA as a matter of course, which adds a handful of success criteria on top of the 2.1 AA this estimate assumes, so treat the band as a floor if 2.2 is your target. I work with organizations that have to satisfy an accessibility review as well as a procurement one. Book a discovery call

Related tools

Need the work behind the number rather than the estimate? Tell me what you are planning →

Leave a reply

Your email address will not be published. Required fields are marked *

Your rating (optional)

Your name and email are stored with your comment; only your display name is shown publicly. See our privacy policy.