WCAG Auditor
Scoring Methodology
How the Nimble WCAG Auditor turns an automated WCAG 2.2 AA scan into a 0–100 accessibility score — which axe-core rules map to which category, the category weights, and the pass / partial / fail thresholds.
What this page is
Nimble does not author a new accessibility standard. WCAG 2.2 AA is an external, published standard maintained by the W3C. What this page documents is Nimble’s scoring methodology and interpretation — the specific decisions that turn an automated axe-core scan into a single 0–100 score:
- The five categories the score is built from.
- The curated axe-core rule subset backing each category.
- The weight each category contributes to the total.
- The pass / partial / fail thresholds for each category.
- The score bands that label the final number.
This methodology is versioned. Every version is permanently archived and reachable. When the rubric is re-tuned, a new version is published as a new archived page — prior versions are never overwritten, so a score computed under an older version always links to the methodology it was actually measured against.
What the score can and cannot tell you
Automated testing detects only a portion of WCAG 2.2 AA success criteria — industry consensus puts automated coverage at roughly 30–40% of the standard. The remaining criteria require human judgement (does this alt text actually describe the image? is the reading order sensible?).
The score is an automated-coverage indicator, not a conformance claim. A high score means the automated checks Nimble runs found few issues. It does not certify WCAG conformance and it is not legal advice. Full conformance requires manual review.
How the score is computed
- The page is rendered in a real headless browser (Chromium) so computed styles, contrast, and the accessibility tree are all available.
- axe-core runs the curated rule subset below against the rendered page.
- Each of the five categories is evaluated to pass, partial, or fail using its own scoring mode and thresholds.
- Each category contributes a weighted share of 100 points. Pass earns the full weight, partial earns half, fail earns none.
- Categories that do not apply to a page (for example, the image category on a page with no images) are excluded from the denominator — they neither help nor hurt the score.
- The weighted total is the 0–100 accessibility score.
The five categories
The rubric is grounded in WCAG 2.2 AA’s POUR model (Perceivable, Operable, Understandable, Robust). Version v1.0 ships a curated, high-confidence axe-core rule subset; later versions grow the subset (see Versioning below).
Category weights
| # | Category | POUR principle | Weight |
|---|---|---|---|
| 1 | Text alternatives | Perceivable | 20% |
| 2 | Color & contrast | Perceivable | 25% |
| 3 | Structure & semantics | Perceivable / Robust | 20% |
| 4 | Forms & controls | Operable / Understandable | 20% |
| 5 | Keyboard & navigation | Operable | 15% |
The weights sum to 100%. Color & contrast carries the largest weight (25%) because contrast failures affect the most users and are among the most common, highest-impact violations.
1. Text alternatives — 20%
POUR principle: Perceivable. WCAG 2.2 AA criterion: 1.1.1.
Images, icons, and other non-text content carry a meaningful text alternative (or are correctly marked decorative) so screen-reader users get the same information.
axe-core rules (v1.0): image-alt, input-image-alt, area-alt, svg-img-alt, object-alt
Scoring mode: share of images with a correct text alternative.
| Result | Threshold |
|---|---|
| Pass | ≥ 95% of images have a meaningful alt or correct decorative marking |
| Partial | 70% – 95% |
| Fail | below 70% |
If a page has no images at all, this category does not apply and is excluded from the score.
2. Color & contrast — 25%
POUR principle: Perceivable. WCAG 2.2 AA criterion: 1.4.3.
Text meets the WCAG 2.2 AA contrast minimum (4.5:1 for normal text, 3:1 for large text) so low-vision users can read it.
axe-core rules (v1.0): color-contrast
Scoring mode: count of contrast violations (failing elements).
| Result | Threshold |
|---|---|
| Pass | 0 contrast violations |
| Partial | 1 – 5 violations |
| Fail | more than 5 violations |
3. Structure & semantics — 20%
POUR principle: Perceivable / Robust. WCAG 2.2 AA criteria: 1.3.1, 2.4.2, 2.4.6, 3.1.1.
The page has a title and a declared language, an ordered heading outline, landmark regions, and well-formed lists and data tables — the structure assistive technology navigates by.
axe-core rules (v1.0): document-title, html-has-lang, heading-order, landmark-one-main, region, list, td-headers-attr
Scoring mode: count of failing structure rules.
| Result | Threshold |
|---|---|
| Pass | 0 failing rules |
| Partial | 1 – 2 failing rules |
| Fail | 3 or more failing rules |
4. Forms & controls — 20%
POUR principle: Operable / Understandable. WCAG 2.2 AA criteria: 1.3.1, 2.4.4, 4.1.2.
Every form field, button, and link has an accessible name and valid ARIA so it can be operated and understood without sight.
axe-core rules (v1.0): label, select-name, button-name, link-name, aria-required-attr, aria-valid-attr-value
Scoring mode: count of unlabeled or invalid controls (failing elements).
| Result | Threshold |
|---|---|
| Pass | 0 violations |
| Partial | 1 – 3 violations |
| Fail | more than 3 violations |
5. Keyboard & navigation — 15%
POUR principle: Operable. WCAG 2.2 AA criteria: 2.1.1, 2.4.1, 2.4.3.
The page can be operated by keyboard alone: a skip link, no positive tabindex traps, focusable frame content, and a sane focus order.
axe-core rules (v1.0): bypass, tabindex, focus-order-semantics, frame-focusable-content, accesskeys
Scoring mode: count of failing keyboard rules.
| Result | Threshold |
|---|---|
| Pass | 0 failing rules |
| Partial | 1 – 2 failing rules |
| Fail | 3 or more failing rules |
Score bands
The final 0–100 number is labelled with one of three bands:
| Score range | Band |
|---|---|
| 90 – 100 | Strong accessibility coverage |
| 50 – 89 | Partial coverage — gaps to fix |
| 0 – 49 | Significant accessibility gaps |
A band describes what the automated checks measured. As with the score itself, it is not a conformance verdict.
Versioning
This methodology is published under a version label — v1.0 is current. Each published version corresponds to a registered row in Nimble’s WCAG rubric registry, and every audit lead and scan record stores the exact rubric_version it was scored against.
Why versioning matters: the rubric grows over time. From week 2 of launch, Nimble adds axe-core rules toward the full WCAG 2.2 AA ruleset on a daily-ship cadence. Re-tuning a threshold or adding a rule changes how a score is computed — so a re-tune is published as a new version, never an edit to an existing one. An old score always links to the methodology version it was actually measured against, so a re-tune can never silently re-interpret a score a merchant already received.
This page always shows the current version; every prior version is permanently archived and reachable from the version history below.
Version history
| Version | Published | Categories | Notes |
|---|---|---|---|
| v1.0 — current | 2026-05-20 | 5 | Initial WCAG 2.2 AA rubric — 5 categories on a curated, high-confidence axe-core rule subset. |
Not legal advice
This methodology and the scores it produces cover automated accessibility checks only. They are not legal advice and not a certification of ADA or WCAG conformance. The scan runs on the open-source axe-core engine. Full WCAG 2.2 AA conformance requires manual review by a qualified accessibility professional.