Accessibility testing for training provider portals

Featured image with a rose pink pastel background and a diagonal stripe pattern, showing the post title "Accessibility Testing for Training Provider Portals"

Accessibility testing for training provider portals

Accessibility testing for a training provider portal means checking, against WCAG 2.2, that learners, client managers and administrators can finish every task with a keyboard, a screen reader, zoom or voice control. Automated scans cover part of that. The rest needs a person working through your registration, payment, scheduling and attendance flows.

Training portals are harder to test than a brochure site. They have date pickers, seat-hold timers, drag-and-drop schedulers, QR check-in, recorded sessions and, for providers who serve many clients, a different brand color set per client. This guide shows what to test, what tools can and cannot do, and how often to run it.

Key takeaways

  • Test to WCAG 2.2 Level AA. It adds nine success criteria to 2.1, and several of them (target size, dragging, redundant entry, accessible login) hit portal features directly.
  • Map each portal surface to the criteria it can break, then test the five flows that carry revenue and compliance first.
  • Automated tools assist; W3C says they cannot determine accessibility. Plan manual keyboard and screen reader passes for every release that changes a key flow.
  • If clients set their own brand colors, check contrast for every tenant theme, not just your default.
  • Legal duties depend on who you serve and where. Treat the legal section below as a map for a lawyer, not as advice.

Which WCAG version should you test against?

Test against WCAG 2.2 Level AA. W3C published it as a Recommendation on 12 December 2024, adding nine success criteria to WCAG 2.1 and removing 4.1.1 Parsing. W3C describes the older criteria as essentially the same in 2.2, so a 2.2 AA test also covers most of what laws that cite 2.0 or 2.1 ask for.

The WCAG 2.2 Recommendation sorts the new criteria by level. Two are Level A (3.2.6 Consistent Help and 3.3.7 Redundant Entry) and four are Level AA (2.4.11 Focus Not Obscured (Minimum), 2.5.7 Dragging Movements, 2.5.8 Target Size (Minimum) and 3.3.8 Accessible Authentication (Minimum)). Three more sit at AAA, which most teams treat as optional.

What should accessibility testing cover in a training portal?

Cover every surface where a learner, client manager or administrator must act: search, calendar, forms, login, checkout, session delivery and the admin scheduler. The table maps each surface to the WCAG 2.2 criteria it most often breaks. It is our working map, not an official W3C list.

Portal surface What to check WCAG 2.2 criterion (level)
Session search and filters Result counts and “no sessions found” messages are announced without moving focus 4.1.3 Status Messages (AA)
Calendar and date picker Every date reachable and selectable by keyboard; day cells expose a name and role 2.1.1 Keyboard (A); 4.1.2 Name, Role, Value (A)
Sticky headers, cookie banners, chat widgets The focused control is not entirely hidden behind them 2.4.11 Focus Not Obscured (Minimum) (AA)
Registration form Visible labels, errors named in text, autofill-ready fields, no repeated typing 3.3.2 Labels or Instructions (A); 3.3.1 Error Identification (A); 1.3.5 Identify Input Purpose (AA); 3.3.7 Redundant Entry (A)
Login and account recovery No memory or puzzle test unless an alternative method or a helping mechanism is offered 3.3.8 Accessible Authentication (Minimum) (AA)
Checkout with seat-hold timer Warning before expiry and a simple way to extend the hold 2.2.1 Timing Adjustable (A)
Admin scheduler (drag a trainer onto a session) A single-pointer alternative to dragging 2.5.7 Dragging Movements (AA)
Admin tables with icon buttons Targets at least 24 by 24 CSS pixels, or enough spacing 2.5.8 Target Size (Minimum) (AA)
Recorded sessions Captions on prerecorded video with audio 1.2.2 Captions (Prerecorded) (A)
Help links across pages Repeated help or contact options appear in the same order relative to other page content 3.2.6 Consistent Help (A)

How do you test registration forms, errors and checkout?

Complete the whole registration and payment path using only a keyboard, then repeat it with a screen reader. Submit the form with mistakes. Each error must be identified and described in text, and any detail you already gave, such as a billing address, should not have to be typed again in the same process.

WCAG 3.3.7 requires information entered earlier in the same process to be auto-populated or available to select, with exceptions such as security and invalid data. In a training flow this can fail on the second and third attendee of a group booking. See our guide to the registration and payment workflow for where those steps sit.

How do you test calendars, date pickers and schedulers?

Put the mouse away and try to book a session, move a session and assign a trainer. Every date and every drag action needs a keyboard path. For 2.5.7, the standard asks that dragging functions can be done by a single pointer without dragging, unless dragging is essential.

Then check focus. With a sticky header or filter bar open, tab through the page and confirm the focused control is never entirely hidden, which is the test for 2.4.11. Also confirm that date fields accept typed input, which is good practice rather than a named criterion.

How do you test live sessions, recordings and QR check-in?

Check recordings for captions, check join and help links for consistent placement, and test check-in by keyboard and screen reader as well as by scanning. A QR code that is the only way in locks out anyone who cannot aim a phone, so confirm that a roster or code-entry alternative exists.

Our guide to QR code attendance tracking covers those alternatives from the operations side. For captions, SC 1.2.2 requires captions for prerecorded audio in synchronized media, so review them for accuracy rather than only for presence.

Pro tip: Test with the timer running. Shorten the seat hold to 60 seconds in a staging environment, then complete checkout with a screen reader. If the expiry warning is not announced, or extending takes more than one simple action, you have a 2.2.1 defect that a static scan will never show.

How much can automated accessibility testing find?

Less than most teams expect. W3C’s guidance on evaluation tools says they “can not determine accessibility, they can only assist in doing so,” and that human judgment is required. Use scans to catch mechanical problems early, and reserve people for anything involving meaning, order or interaction.

The split below is our practitioner view of where each method earns its time. Confirm it against the documentation of the tool you use, because rule sets differ and change.

Check Automated scan Needs a person
Missing form labels, names and roles Usually flags them Confirm the name makes sense in context
Text contrast on a plain background Usually measures it Text over images, gradients and hover states
Keyboard access to a custom date picker Cannot operate it Tab, arrow and Escape behavior, no keyboard trap
Error message quality Can detect missing association Whether the text tells the learner how to fix it
Status messages Can detect a missing live region Whether a screen reader announces it at the right moment
Caption accuracy Cannot judge Read along with the audio
2.2-only criteria (dragging, help, login) Coverage depends on the tool Manual steps from the table above

One tool detail matters for 2.2 work. Microsoft’s Accessibility Insights for Web pairs about 50 automated checks with guided manual tests, but, as of our November 2026 read, its overview describes the full Assessment against WCAG 2.1 Level AA. Check whether your tool covers the six new AA and A criteria, and test the gaps by hand. NVDA, a free Windows screen reader, is a practical first assistive technology to add.

How do you test per-client branding in a multi-tenant portal?

Check every tenant theme, not only your default one. When a client picks its own button and link colors, nothing stops it choosing a pair that fails the 4.5:1 minimum text contrast in WCAG 1.4.3 Contrast (Minimum). Run a contrast check as part of onboarding each new client.

The script below applies the formula W3C publishes: the contrast ratio is (L1 + 0.05) / (L2 + 0.05), where L1 and L2 are the relative luminance of the lighter and darker colors. The three themes are invented for illustration. We ran it, and it checks only the 4.5:1 threshold for normal-size text.

def channel(c8):
    s = c8 / 255
    return s / 12.92 if s <= 0.04045 else ((s + 0.055) / 1.055) ** 2.4

def luminance(hex_color):
    h = hex_color.lstrip("#")
    r, g, b = (int(h[i:i + 2], 16) for i in (0, 2, 4))
    return 0.2126 * channel(r) + 0.7152 * channel(g) + 0.0722 * channel(b)

def contrast(fg, bg):
    l1, l2 = sorted((luminance(fg), luminance(bg)), reverse=True)
    return (l1 + 0.05) / (l2 + 0.05)

# Illustrative tenant themes: (button text, button background, link on white)
tenants = {
    "client-a": ("#FFFFFF", "#0B5CAD", "#0B5CAD"),
    "client-b": ("#FFFFFF", "#F28C28", "#F28C28"),
    "client-c": ("#FFFFFF", "#767676", "#999999"),
}
for name, (btn_text, btn_bg, link) in tenants.items():
    button = contrast(btn_text, btn_bg)
    link_ratio = contrast(link, "#FFFFFF")
    ok = button >= 4.5 and link_ratio >= 4.5
    print(f"{name}: button {button:.2f}:1, link {link_ratio:.2f}:1 -> {'PASS' if ok else 'FAIL'}")

It printed PASS for client-a (6.67:1), FAIL for client-b (2.45:1) and FAIL for client-c, whose button passes at 4.54:1 but whose link text reaches only 2.85:1. Large text, icons and focus outlines have their own thresholds, so use this as a gate, then verify the rendered pages with a browser tool.

How often should you run accessibility testing?

Run it at three speeds: automated checks on every code change, a short manual pass on every release that touches a key flow, and a full pass on a schedule. The cadence below is our suggestion; raise it for flows with legal or revenue exposure.

  • Every pull request: automated scan of the changed pages and components, failing the build on new violations.
  • Every release that changes a key flow: keyboard-only run, 200 percent zoom check and one screen reader pass of that flow.
  • Every new client theme: the contrast script, then a visual check of focus indicators and error text.
  • Each quarter: a full manual pass of registration, checkout, login, scheduler and check-in, with results logged by criterion.
  • After a redesign or yearly: an expert audit, plus sessions with disabled users who use assistive technology daily.

Pro tip: Tag every defect with its criterion. Record the WCAG number and level (for example, 2.5.8, AA) on each bug. Your backlog then doubles as a conformance report, and a client or auditor can see which criteria are open without a separate spreadsheet.

What do ADA, Section 508 and the European Accessibility Act mean for a training provider?

The duties depend on who your customers are and where they live, and some are not settled for private companies. This is a map for a legal review, not advice. As of November 2026, the main regimes use different WCAG versions, so a single 2.2 AA test is a practical baseline.

Regime Who it reaches Technical benchmark Status as of November 2026
ADA Title II web rule (US) State and local governments WCAG 2.1 Level AA Dates moved by an April 2026 interim final rule: April 26, 2027 for entities of 50,000 or more people, April 26, 2028 for smaller ones and special districts
ADA Title III (US) Businesses open to the public No detailed DOJ web standard DOJ’s 2022 guidance says it has no regulation setting detailed standards and that businesses can choose how to comply
Section 508 (US) Federal agencies, through what they develop, procure and use WCAG 2.0 Level A and AA Access Board ICT standards, section E205.4
European Accessibility Act Services provided to consumers after 28 June 2025, including e-commerce services; microenterprises providing services are exempt Essential requirements in the Directive; EN 301 549 is one voluntary means of conforming Measures apply from 28 June 2025; check each member state’s national law

The Department of Justice’s web rule page covers the Title II dates above. Most private training companies are not Title II entities, but a public-sector client may ask for conformance evidence, so ask counsel whether any contract language reaches you.

Under Directive (EU) 2019/882, e-commerce services are services provided at a distance through websites and mobile apps, at a consumer’s request, to conclude a consumer contract. A checkout that sells courses to individuals may fit. Whether a business-only portal does is a question for counsel and for each country’s law.

The standard is moving. EN 301 549 V3.2.1 draws on WCAG 2.1. EN 301 549 V4.1.1, adopted on 24 August 2026, aligns with WCAG 2.2, but its presumption of conformity applies only once it is cited in the Official Journal of the EU. Ireland’s National Disability Authority expects that citation on 16 December 2026, so verify the date. Also ask counsel how to treat 4.1.1 Parsing where a law still cites WCAG 2.0 or 2.1.

What should you do next?

Pick the five flows that carry the most revenue or legal exposure, such as search, registration, checkout, login and check-in. Build a one-page test sheet from the surface table, with the criterion number beside each step. Run a keyboard pass and one screen reader pass this week, and log defects by criterion.

Then run the contrast script on every client theme you host. If you are evaluating a training management system with per-client portals, note that SimpliTrain’s page for professional training companies describes white-labelled portals with each client’s own logo, colors and domain, but its pages do not state accessibility conformance. Ask any vendor for evidence, and test the themes yourself.

Frequently asked questions

What are the main types of accessibility testing?

The common types are automated scanning, manual testing and user testing. Manual testing means working through the interface with a keyboard, a screen reader or magnification. User testing brings in people with disabilities who use assistive technology daily. Most teams combine all three, because each finds problems the others miss.

Is automated accessibility testing enough on its own?

No. W3C states that tools cannot check all accessibility aspects automatically, that human judgment is required, and that tools can sometimes give false or misleading results. A scan is a fast first filter that belongs in your build pipeline. It cannot confirm that a date picker works by keyboard or that an error message helps.

Which screen reader should I test with?

Start with one you can run today. NVDA is free and runs on Windows, which makes it easy to add to a QA machine. If your learners book from phones, also test with a mobile screen reader on the devices they use. Record which browser and screen reader pairing you tested.

Do we need to test with people who have disabilities?

It is worth doing for your highest-risk flows. Microsoft’s training guidance lists user testing with people who represent a range of needs and abilities alongside manual and automated testing. Checklist testing shows whether criteria are met. Sessions with real users show whether a task is actually workable, and which fixes matter most.

Does WCAG 2.2 replace WCAG 2.1?

W3C published WCAG 2.2 as a Recommendation on 12 December 2024. It adds nine success criteria, removes 4.1.1 Parsing and keeps the earlier criteria essentially the same. Some laws still cite 2.0 or 2.1, so check which version your contracts and local rules name before you make a formal conformance claim.

Alert: mundrisoftcareers.com is not affiliated with Mundrisoft Solutions. Beware of fake job offers and trust only our official recruitment channels.