Dispensary Digital Accessibility Data 2026

95.9% of the one million popular home pages scanned by WebAIM in 2024 had automatically detectable WCAG failures [1]. The same study found 56.8 detected errors per page [1]. These are not dispensary-specific failure rates, but they show why Dispensary Digital Accessibility Data 2026 must go beyond a menu widget or a one-time scan: customers need to find products, understand restrictions, reserve an order, fix errors, and request help across the complete journey.

The direct 2026 accessibility answer

Public evidence does not provide a representative national accessibility score for dispensary websites. The strongest approach combines population need, technical standards, enforcement context, broad web scans, and direct testing of the dispensary’s priority tasks.

The CDC estimated that 28.7% of US adults, or more than one in four, had some type of disability based on Behavioral Risk Factor Surveillance System data [2]. Disability categories and survey methods matter, so that estimate should not be converted into a forecast of one store’s customers.

WCAG 2.2 defines 86 success criteria distributed across three conformance levels: A, AA, and AAA [3]. Counting criteria is useful for scope, but conformance applies to complete pages and processes under WCAG’s requirements; it is not calculated as a simple percentage of checks passed.

The Department of Justice’s 2024 rule for state and local government web content generally uses WCAG 2.1 Level AA as its technical standard, with compliance dates that depend on population [4]. That Title II rule does not itself create a new universal deadline for private dispensaries, which are generally considered under the separate Title III framework [5].

Screenshot-ready accessibility data table

WebAIM reported that low contrast, missing image alternative text, missing form labels, empty links, empty buttons, and missing document language accounted for 81% of detected errors [1]. This concentration helps triage automated findings, but keyboard traps, confusing instructions, screen-reader announcements, and many workflow failures require human evaluation.

Why dispensary journeys have distinctive barriers

A dispensary menu is a dense application, not just a brochure. It can combine age gates, geolocation, product filters, promotional banners, potency and ingredient information, stock status, pickup slots, account login, cart controls, and third-party checkout.

Age gates often fail when focus is not moved into the dialog, background content remains operable, or the controls have poor names. A person using a keyboard or screen reader may be blocked before reaching the menu even when the page behind the gate is well coded.

Product cards can rely on color, hover, or images to communicate category and availability. WCAG requires information not to depend on color alone and includes contrast requirements for text and meaningful graphics [3]. Alternative text should convey the image’s purpose without duplicating adjacent product text.

Filters need programmatic names, states, and predictable focus. When results update dynamically, assistive technology users need an appropriate status message rather than an unexplained page change.

Pickup flows add timing and error-recovery risks. A timeout should be communicated, form errors should identify the affected field in text, and users should be able to review important information before submission under the applicable WCAG criteria [3].

Keyboard and focus data

A keyboard test should record tasks, not merely count focusable elements. Test opening and dismissing the age gate, navigating categories, applying and removing filters, opening product details, modifying the cart, selecting pickup information, correcting errors, and reaching support.

WCAG 2.2 added Focus Not Obscured criteria and a Focus Appearance criterion at different levels, reflecting the importance of visible, usable focus [3]. Sticky banners, chat launchers, cookie notices, and cart drawers can obscure the focused control even when tab order technically moves.

Use “task completed without keyboard trap” as a local binary observation for each tested task. The completion rate is completed tasks divided by eligible tasks tested, labeled as a local calculation, not a conformance percentage.

Record browser, operating system, viewport, zoom, input method, page version, tester, and evidence. A fixed defect may return after a menu vendor update, so the result needs a release or date.

Screen-reader and accessible-name checks

Screen-reader testing asks whether controls expose the correct role, name, state, and value and whether changes are announced. An icon that looks like a cart to a sighted customer may be announced as an unlabeled “button” if it lacks an accessible name.

Names should remain consistent with visible labels. WCAG’s Label in Name criterion helps speech-input users activate controls using words they can see [3]. Adding invisible text that conflicts with the visible label can make a nominal fix create another barrier.

Use at least one supported desktop and one supported mobile assistive-technology path as an editorial testing recommendation, not a universal legal minimum. Pair test results with the site’s actual traffic and support commitments, and do not claim coverage of every assistive technology.

Dynamic stock and cart messages should be understandable without moving focus unexpectedly. Excessive live-region announcements can be as disruptive as silence, so manual task observation matters.

Zoom, reflow, and mobile access

WCAG includes requirements related to text resizing and reflow [3]. Test priority content at the criterion’s specified conditions, preserving the exact browser and viewport evidence rather than describing a page as simply “mobile friendly.”

Cannabis product names and laboratory fields can be long. Fixed-height cards, clipped badges, and side-by-side potency columns may overlap when text grows. Users must not lose ingredients, warnings, prices, or transaction controls because a component was designed around one line length.

Mobile orientation and touch-target behavior also affect access. WCAG 2.2 introduced a minimum target-size criterion at Level AA with defined exceptions [3]. Do not replace the criterion’s wording with a homegrown pixel rule.

Captions, documents, and customer communications

Prerecorded synchronized media generally needs captions at Level A under WCAG, while audio description and other media criteria vary by level and context [3]. Auto-generated captions should be reviewed for cannabis product names, dosage language, and speaker identification because transcription errors can change meaning.

PDF menus and certificates can create a parallel inaccessible channel. Test reading order, headings, table structure, link text, form fields, and document language. A tagged PDF is not automatically understandable.

Email confirmations should retain meaningful heading order, link purpose, text alternatives, contrast, and readable transaction details. A customer should not need to inspect an image to find pickup status or support information.

For service metrics that include accessible communication, see dispensary customer service benchmarks. Teams managing compliant campaign and consent records can review our age-gated marketing operations service, while the sibling dispensary cybersecurity statistics guide addresses security controls around these same systems.

Overlays and automated tools

Automated testing is valuable for repeated detection of machine-testable patterns. W3C’s evaluation guidance recommends combining tools with human judgment because no tool can determine accessibility by itself [6].

An overlay or toolbar does not repair inaccessible source components by default. The FTC finalized an order against accessiBe in 2025 over claims about the ability of its automated product to make websites compliant, requiring the company to pay one million dollars [7]. That enforcement action concerns specific advertising claims and should not be generalized into a legal judgment about every product.

Maintain tool findings as evidence with rule identifier, URL, component, first seen, release, severity rationale, owner, status, and retest. Deduplicate repeated component failures carefully: one template defect can appear on many pages, but affected journeys and instances still matter.

A defensible measurement program

Start with a journey inventory: store information, eligibility information, menu search, product comparison, account access, cart, pickup, payment information, and support. This list is an editorial recommendation, and each operator should add journeys created by its platform and legal framework.

For each release, run automated checks against representative templates and manually test priority tasks. Include disabled testers in research and acceptance when feasible, compensate them appropriately, and avoid treating one tester’s experience as universal.

Measure open barriers by user impact and journey, not only raw issue count. A single keyboard trap at the age gate can block the entire site, while many repeated low-impact markup findings may affect only secondary text.

Track time to verified remediation from issue acceptance to successful retest. “Developer says fixed” is an intermediate state, not closure.

Publish denominators. If a report says that most journeys passed, show which journeys, versions, technologies, and criteria were evaluated. Mark not tested, not applicable, and blocked as distinct outcomes.

Methodology and limitations

This review uses 14 sources published or updated from 2020 through 2026 and verified on July 23, 2026. Primary authorities and standards establish legal or technical context; WebAIM supplies broad scan data; CDC supplies population context.

WebAIM’s sample covers popular home pages and automated rules, not dispensary sites or complete WCAG conformance [1]. CDC disability estimates are survey-based and do not measure digital barriers [2].

Legal obligations depend on the entity, jurisdiction, technology, and facts. The DOJ Title II rule covers state and local government entities, while private businesses require a separate Title III and jurisdiction-specific analysis [4][5]. This article is research, not legal advice.

Frequently asked questions

What percentage of dispensary websites are inaccessible?

No representative national percentage was identified. WebAIM found detectable WCAG failures on 95.9% of one million popular home pages, but that sample is not dispensary-specific and automated testing is incomplete [1].

Is WCAG 2.2 Level AA legally required for every dispensary?

WCAG is the leading technical standard, but legal duties depend on jurisdiction and entity. DOJ’s 2024 web rule applies WCAG 2.1 Level AA to covered state and local governments, not automatically to every private dispensary [4][5].

Can an automated scan prove compliance?

No. W3C states that evaluation tools assist the process but cannot determine accessibility on their own [6]. Manual task and assistive-technology testing are necessary.

How many WCAG 2.2 criteria are there?

WCAG 2.2 contains 86 success criteria across three levels [3]. Conformance is not an arithmetic score based on the share passed.

What should be fixed first?

Prioritize barriers that block essential journeys, expose wrong information, or prevent support, then address repeated component defects. Severity should reflect user impact and reach, not just the scanner label.

Sources

1. WebAIM, The WebAIM Million 2024, published March 2024.
2. CDC, Disability Impacts All of Us, updated July 15, 2024.
3. W3C, Web Content Accessibility Guidelines 2.2, published October 5, 2023.
4. US Department of Justice, Fact Sheet: Accessibility of Web Content and Mobile Apps Provided by State and Local Government Entities, published April 24, 2024.
5. US Department of Justice, ADA Title III, updated March 8, 2024.
6. W3C Web Accessibility Initiative, Selecting Web Accessibility Evaluation Tools, updated May 2021.
7. FTC, FTC Approves Final Order Requiring accessiBe to Pay $1 Million, published April 22, 2025.
8. US Access Board, Information and Communication Technology Standards and Guidelines, updated 2024.
9. Section508.gov, Laws and Policies, updated May 30, 2024.
10. W3C, Accessible Rich Internet Applications 1.2, published June 6, 2023.
11. W3C, Understanding Conformance, published October 5, 2023.
12. NIST Computer Security Resource Center, Human-Centered Cybersecurity, reviewed May 26, 2026.
13. Virginia Cannabis Control Authority, Laws and Regulations, accessed July 23, 2026.
14. Virginia Information Technologies Agency, Accessibility Standard GOV103-00, updated 2024.

Conclusion

Accessibility data is credible only when it connects standards and scan results to real customer tasks and verified remediation. If you need light help maintaining that evidence, book a free consultation call.