Accessibility Audit Before and After: What the Results Actually Show
Most organizations that commission an accessibility audit receive a report. Many of those reports sit unactioned for weeks — not because the organization does not want to fix things, but because the findings are difficult to prioritize, the severity labels are inconsistently defined, and there is no clear way to measure whether remediation is actually working.
Understanding accessibility audit before and after results requires more than counting issues. It requires a scoring framework that connects individual findings to a measurable picture of overall accessibility — one that improves in a trackable way as remediation progresses.
This article explains how accessibility audit results are structured, what before-and-after comparison actually looks like, and what the numbers mean in practice.
Why Accessibility Audit Results Are Hard to Read Without a Framework
A typical accessibility audit report lists issues against WCAG success criteria. Each issue is tagged with a criterion — 1.1.1, 2.4.7, 4.1.2 — a brief description, and often a severity or priority level.
The problem: without a consistent scoring methodology, two reports of the same product produced by two different auditors can look completely different in apparent severity. One auditor’s “major” is another auditor’s “critical.” One report lists 47 issues; another on the same product lists 12, because the second auditor grouped repeated instances.
More importantly, a flat list of issues cannot answer the most practical question product teams ask: are we getting more accessible over time, and how much more accessible?
This is where a structured scoring methodology matters. Rather than simply counting issues before and after remediation, a scoring framework weights issues by their impact on users with disabilities, then produces a number that reflects the actual accessibility risk or conformance level of the product.
The Two Metrics That Matter: Defect Score and Conformance Score
Accessibility measurement works best when it tracks two distinct dimensions simultaneously: how severe the problems are, and how aligned the product is with the standard.
Defect Score: Measuring Risk
A defect score assigns weight to each issue based on its severity level. BarrierBreak’s scoring methodology, for example, uses four severity levels:
- Blocker — the issue completely prevents a user with a disability from completing a task. A screen reader user cannot log in. A keyboard user cannot reach the checkout button.
- Critical — significant impact that creates a major barrier, even if a workaround exists with substantial effort.
- Major — a meaningful barrier that significantly degrades the experience for users with certain disabilities.
- Minor — impacts usability but does not block task completion.
Each severity level carries a different weight in the overall score calculation. Blockers and Critical issues contribute far more to the score than Minor ones — correctly reflecting their impact on real users.
The resulting defect score sits within one of four risk bands:
Risk LevelScore RangeLow0–15%Medium15–40%High40–60%Very High60% and above
A product with a defect score of 71% is at Very High Risk — it has a significant concentration of severe barriers. After a remediation cycle focused on Blockers and Critical issues, that score might fall to 38%, moving the product from Very High Risk to Medium Risk. That shift is meaningful progress, and it is measurable.
The goal over time is to bring the defect score as close to zero as possible — meaning that the remaining issues are predominantly Minor in nature and do not block users from completing core tasks.
Conformance Score: Measuring WCAG Alignment
Where the defect score measures risk, the conformance score measures how well the product meets WCAG success criteria. Each success criterion is assigned a criticality level — High, Medium, or Low — based on its importance to accessibility. A pass on a High criticality criterion contributes significantly more to the conformance score than a pass on a Low criticality one.
The conformance score is calculated as the ratio of the score actually achieved across all applicable criteria to the maximum possible score, expressed as a percentage.
RatingScore RangePoor0–20%Average20–70%Good70–90%Excellent90% and above
A product scoring 42% on conformance is rated Average — it is partially conformant but has significant gaps across multiple success criteria. After remediation, a score of 78% reflects a Good rating — most success criteria are now passing, with remaining failures concentrated in lower-criticality areas.
Get BarrierBreak’s stories in your inbox
Join Medium for free to get updates from this writer.
For a detailed breakdown of how these two scoring dimensions are structured and calculated, BarrierBreak’s explanation of their accessibility scoring methodology covers the formulas, weightages, and rating bands in full.
What Before-and-After Results Actually Look Like
In practice, a well-structured accessibility audit cycle follows this pattern:
Baseline audit: The product is tested against WCAG 2.1 or 2.2 Level AA using a combination of automated scanning and manual testing with assistive technologies. All issues are logged with severity levels and WCAG criterion mapping. The baseline defect score and conformance score are calculated.
Remediation sprint: Development and design teams work through the prioritized issue list, starting with Blockers and Critical issues. The focus is on the highest-severity items across the most-used user journeys — login, checkout, account management, core content.
Verification audit: The same pages and user journeys are re-tested. Issues marked as fixed are verified. New issues introduced during remediation are flagged. The updated defect score and conformance score are calculated.
The before-and-after comparison: The two score sets are compared. A product that moved from a defect score of 68% (Very High Risk) to 22% (Medium Risk) has eliminated most of its Blocker and Critical issues. A conformance score that moved from 45% (Average) to 82% (Good) shows that the majority of WCAG success criteria are now passing.
This is what meaningful progress looks like — not a list of “X issues fixed,” but a measurable shift across both risk and conformance dimensions.
What Drives the Biggest Before-and-After Improvements
Not all issues contribute equally to score improvement. Targeting the right categories produces the fastest measurable gains.
Form and input accessibility — unlabeled form fields, missing error associations, and inaccessible validation messages often account for a large share of Blocker and Critical issues, particularly on e-commerce, banking, and government platforms. Fixing labeling and error handling across a form library can dramatically reduce the defect score in a single sprint.
Keyboard and focus management — elements that are unreachable by keyboard, interactive components without visible focus indicators, and modal dialogs that trap or lose focus are consistently high-severity findings. These are often systemic — fixing the component once fixes every instance of it across the product.
Image alt text — missing or generic alternative text on informational images is one of the most common and easiest-to-verify fixes. It is a WCAG 1.1.1 failure (High criticality in most scoring frameworks) and directly affects the conformance score.
Screen reader compatibility for custom components — custom tabs, accordions, dropdowns, and data tables that do not expose role, state, and name to assistive technologies are Blocker-level issues for screen reader users. Remediating these typically requires developer work, but the conformance impact is significant.
Color contrast — contrast failures on body text and interactive elements affect the conformance score for WCAG 1.4.3 and 1.4.11 but are design system fixes, not page-by-page repairs. Fixing the design token once propagates the fix across the entire product.
Common Mistakes in Interpreting Before-and-After Results
Celebrating issue count reduction rather than score improvement. A team that closes 40 Minor issues and leaves 5 Blockers unfixed has reduced their issue count significantly but may have barely moved their defect score. Score-based measurement prevents this kind of misleading progress tracking.
Treating a Good conformance score as full accessibility. A conformance score of 82% reflects strong WCAG alignment — but it does not guarantee that every user with every disability can complete every task. Testing with real users who have disabilities remains essential alongside conformance scoring.
Not re-testing after remediation. Teams sometimes mark issues as fixed in the audit report without independent verification. A verification pass by the audit team — not the development team — confirms that fixes have been implemented correctly and have not introduced new issues.
Scoping the audit too narrowly. A before-and-after comparison is only meaningful if it covers the same scope. If the baseline audit covered five pages and the verification covers a different five, the scores are not comparable.
Organizations that want to understand what a structured accessibility audit and remediation cycle looks like in practice — including how scoring is applied across a product and used to track improvement over time — can find more about the process at BarrierBreak, which has developed and refined its accessibility scoring methodology over two decades of audit and remediation work.
Conclusion
Accessibility audit before and after results are most meaningful when they are expressed through a consistent, weighted scoring framework — not just issue counts. A defect score that tracks severity-weighted risk, and a conformance score that tracks WCAG alignment, together give product teams a clear, comparable picture of where they started and how far they have come.
The goal is not a perfect score at a single point in time. It is a steadily improving score maintained through ongoing testing, remediation, and integration of accessibility into the development cycle — so that each release starts from a stronger baseline than the last.
FAQ
What is the difference between an accessibility defect score and a conformance score?
A defect score measures the risk level of a product based on the severity of accessibility issues found — it reflects how many barriers exist and how serious they are. A conformance score measures how well the product meets WCAG success criteria — it reflects how aligned the implementation is with the standard. Both are useful together: a product can have a low defect score (few severe issues) while still having a moderate conformance score if many minor gaps exist across multiple criteria.
Why does the defect score go down when accessibility improves, while the conformance score goes up?
They are intentionally inverse. The defect score measures problems — more and more severe problems produce a higher score, which represents higher risk. The goal is to reduce it toward zero. The conformance score measures alignment with the standard — more passing criteria produce a higher score. The goal is to increase it toward 100%.
How many issues need to be fixed for a meaningful improvement in the scores?
It depends on severity, not count. Fixing one Blocker issue can produce a larger improvement in the defect score than fixing ten Minor issues, because Blockers carry higher weight. Similarly, fixing a Critical-level WCAG success criterion (like 2.1.1 Keyboard or 4.1.2 Name, Role, Value) produces a larger conformance score improvement than fixing a Low criticality criterion.
Does a high conformance score mean a product is fully accessible?
Not necessarily. A conformance score reflects WCAG alignment across testable success criteria. WCAG is a technical standard, and even full WCAG conformance does not guarantee that every user with every disability can use every part of a product effectively. Testing with disabled users — not just technical auditing — remains essential alongside conformance measurement.
How often should accessibility audits be conducted to track before-and-after improvement?
At minimum, a full audit should occur at major product releases, with verification audits after remediation sprints. For products under active development, a quarterly audit cycle provides the most useful before-and-after tracking. Some organizations conduct lightweight monthly spot-checks on critical user journeys between full audits to catch regressions early.
Is there a universal WCAG score that all accessibility tools use?
No. WCAG itself does not define a scoring methodology — WCAG measures pass/fail conformance per success criterion, not an aggregate score. Note that WCAG 3 (currently in development) is working toward a measured, scored approach. In the meantime, individual accessibility firms and platforms use their own scoring frameworks — which is why comparing scores across different tools or auditors requires understanding each framework’s methodology.
