Media Accessibility Principles for Complex Institutions
What universities, health systems, and nonprofits need to understand before they build a program
For most institutions, media accessibility starts with a trigger — a complaint, a lawsuit, a policy review, or a new federal guidance document. What follows is usually a scramble: a quick audit, a vendor search, a few new workflows, and a hope that it's enough.
It rarely is. Not because the effort wasn't genuine, but because media accessibility in complex organizations isn't primarily a technical problem. It's an operational one. The question isn't just *what* accessible media looks like — it's how an institution with years of existing content, limited staff, competing priorities, and ongoing media production builds a program that can actually sustain itself.
This article covers the foundational principles behind media accessibility for higher education, healthcare, research, and nonprofit organizations. It won't tell you exactly what to do — every institution's situation is different — but it will help you understand the landscape clearly enough to ask the right questions and make decisions that hold up over time.
---
Why media accessibility is a distinct challenge
Accessibility as a field covers an enormous range: built environment accessibility, document accessibility, web accessibility, software accessibility, and more. Media accessibility — the accessibility of video and audio content — sits within that field but has its own character.
Video content is time-based, which means accessibility can't be added by changing a setting. It has to be produced deliberately, often through separate workflows, by people who understand both the content and the access format. A caption file is a separate artifact that needs to be created, reviewed, and maintained. An audio description is an additional audio track that has to be scripted, voiced, and timed. These aren't trivial additions; they're production tasks.
At the same time, video content is everywhere in modern institutional life. Universities produce lecture recordings, training videos, promotional content, research presentations, and event recordings. Health systems produce patient education materials, staff training, telehealth recordings, and public communications. Nonprofits produce advocacy content, educational media, and donor communications. The volume is enormous, the formats are varied, and the content keeps accumulating.
This combination — high volume, ongoing production, time-based format, separate accessibility artifacts — is what makes media accessibility genuinely hard for institutions. The technical standards exist. The harder problem is building a program that works within real operational constraints.
---
The regulatory landscape
Media accessibility obligations for institutions come from several overlapping sources, and understanding them is essential to setting priorities.
**Section 508 of the Rehabilitation Act** applies to federal agencies and, through procurement requirements, to any organization that receives federal funding or develops technology for federal use. For higher education and many nonprofits and health systems, Section 508 is a meaningful compliance obligation. Its standards for video and multimedia content require synchronized captions, audio description where needed, and accessible players.
**Title II of the Americans with Disabilities Act (ADA)** applies to state and local government entities, which includes public universities. The ADA requires effective communication with people with disabilities, which courts and enforcement agencies have consistently applied to digital and media content. In April 2024, the Department of Justice published a final rule under Title II explicitly addressing web and mobile accessibility, with provisions that extend to media content.
**Title III of the ADA** applies to places of public accommodation, which includes many private institutions and healthcare providers. Its application to digital content has been less uniform but is increasingly enforced.
**WCAG (Web Content Accessibility Guidelines)**, published by the W3C, provides the technical standard most regulations reference. WCAG 2.1 and 2.2 include specific success criteria for captions (1.2.2 for prerecorded video, 1.2.4 for live), audio description (1.2.3 and 1.2.5), and transcripts (1.2.1 for audio-only content). The DOJ's 2024 rule references WCAG 2.1 Level AA as the applicable standard for Title II entities.
For most institutions, the practical implication of this regulatory landscape is that captioning of prerecorded video is a firm requirement, audio description is required for video content where the visual track carries meaningful information, and live events with audio or video components need accessible alternatives. The specifics depend on your institution type, funding sources, and content, but the general direction is clear.
---
Captioning: The foundation
Captioning is the most established and most commonly required form of media accessibility. It benefits people who are deaf or hard of hearing, people watching in sound-sensitive environments, non-native speakers, and anyone who benefits from reading alongside listening.
For institutions, captioning typically involves three categories of content:
**Prerecorded video** — recorded lectures, training materials, marketing content, patient education videos, and similar assets. This is the bulk of most institutions' content, and it represents the most significant backlog challenge (more on that below). WCAG 2.1 Level AA requires accurate captions for all prerecorded video that contains audio.
**Live video** — webinars, virtual events, live lectures, public meetings, and broadcasts. Live captioning is a different technical challenge than prerecorded captioning. Automated live captions (available through platforms like Zoom and Teams) are widely used but vary in accuracy. Human-provided live captions — typically through CART (Communication Access Realtime Translation) — provide higher accuracy and are appropriate for formal events, high-stakes communications, and situations where accuracy matters significantly.
**Automatically generated captions** — most video platforms now offer auto-captioning, and the quality has improved substantially. Auto-captions can be a useful starting point, but they are not a compliance solution without review and correction. Technical vocabulary, proper nouns, accents, and audio quality all affect accuracy. Institutions that rely on uncorrected auto-captions are exposed to both compliance risk and real harm to users who depend on them.
Quality matters as much as presence. A caption file that is 70% accurate is not accessible. Institutions should have a clear standard for caption accuracy — WCAG's standard is "accurate" — and a process for reviewing and correcting captions before they're published.
---
Audio description: The underestimated requirement
Audio description (sometimes called video description or described video) is an additional narration track that describes the visual elements of a video — action, text on screen, speaker identification, and other visual information — for people who are blind or have low vision.
It is also, at many institutions, the least-addressed accessibility requirement.
This is partly because audio description is harder to produce than captions. It requires identifying what visual information is meaningful, writing description that fits within natural pauses in the audio, and producing an additional audio track. For complex or information-dense videos, it may require extended description — a technique where the video is paused to allow time for more detailed narration. This is a real production challenge, and it's one reason many institutions have captioning programs but no audio description program.
The other reason audio description gets less attention is that the affected population — people who are blind or have low vision — is smaller and often less visible in institutional settings. This is not a good reason to deprioritize it. The ADA and Section 508 do not allow institutions to avoid requirements because fewer people are affected.
WCAG 2.1 Level AA requires audio description for all prerecorded video where the visual track contains information not available in the audio. Level AAA extends this to all prerecorded video. Most institutions should be working toward Level AA compliance as a minimum.
For institutions beginning to address audio description, a few principles matter:
**Not all videos need audio description.** If the visual track contains no information beyond what's already in the audio — a talking-head video of a speaker with no slides, text, or action — audio description may not be required. Institutions should have a decision framework for evaluating which content needs description.
**New production is the right starting point.** Retrofitting audio description onto existing video is expensive and time-consuming. Building audio description into new production workflows — as a standard part of the production process — is far more sustainable.
**Script-based workflows help.** When a video is produced from a script, description can be written alongside that script and recorded in the same session. This is far easier than returning to a finished video and trying to fit description into existing pauses.
---
Live event accessibility
Live events — in-person conferences, public lectures, institutional ceremonies, board meetings, community events — present accessibility challenges that are distinct from recorded media.
**Live captioning** at in-person events is provided through CART (Communication Access Realtime Translation), where a trained stenographer captions speech in real time. CART output can be displayed on a screen visible to the audience, provided as a personal feed to individual attendees on their devices, or both. For formal events and events where any attendee has indicated a need for captioning, CART is typically the appropriate solution.
**Audio description at live events** is provided through assistive listening and description systems. Audio description for live theater, presentations, or events with significant visual components can be provided through dedicated description headsets or apps. This is less common in institutional settings than captioning but is increasingly expected at public-facing events.
**Hybrid events** — events with both in-person and remote attendance — multiply the accessibility requirements. Remote participants need accessible video streams, including captions and, where relevant, audio description. Technical failures that affect accessibility for remote participants are more common than institutions anticipate.
Planning for live event accessibility requires building it into event production workflows, not treating it as a last-minute accommodation request. Institutions that handle accessibility as an add-on — procured the week before an event — consistently have worse outcomes than those that build it into standard event planning.
---
The backlog problem
Perhaps the most daunting challenge for institutions building media accessibility programs is what's already there: years, sometimes decades, of video content that predates current accessibility standards, that exists across multiple platforms and storage systems, and that no one has a clear plan for.
This backlog problem is real, and there's no easy answer. But there are principles that help institutions manage it more effectively.
**Not everything needs to be remediated equally.** A lecture recording from 2009 that gets three views a year is not the same accessibility priority as a patient education video that is actively recommended to patients every week. Institutions should prioritize based on use, not based on the order content was created.
**Prioritization frameworks matter more than remediation sprints.** An institution that remediates 500 videos in a sprint — chosen without a framework — may have done less meaningful work than an institution that carefully identified its 50 highest-impact pieces of content and addressed those. High volume of remediation does not equal high impact.
**Criteria for prioritization typically include:** active use and traffic; regulatory or legal exposure (content subject to specific complaints or scrutiny); population served (content directed at students or patients is typically higher priority than internal content); content longevity (content that will remain relevant for years is a better investment than content that will be replaced); and legal or institutional importance.
**Some content should be retired, not remediated.** Institutions with large content libraries often discover, during accessibility reviews, that a significant portion of their archived content is outdated, redundant, or no longer relevant. Retiring that content is both a legitimate accessibility decision and a sound content management decision.
**Automated captioning at scale** — using AI captioning tools to caption large libraries quickly — can accelerate backlog work, but only if institutions also have a process for reviewing and correcting output. A library of auto-captioned videos with no quality review is not a remediated library; it's a library of unreviewed captions with variable accuracy.
---
Building a sustainable program
The institutions that make meaningful progress on media accessibility share a few characteristics that have less to do with budget and more to do with how they approach the problem.
**They treat accessibility as a production standard, not a post-production fix.** The most efficient accessibility work happens during content creation, not after. Institutions that build captioning, description, and accessibility review into their production workflows — as standard steps, not optional additions — spend less time and money on accessibility than those that treat it as remediation.
**They have clear decision frameworks.** When does a video need audio description? What caption accuracy standard do we hold ourselves to? How do we prioritize backlog work? What platforms are approved for video hosting, and what are the accessibility requirements for each? Institutions with documented answers to these questions make better decisions faster and with less internal friction.
**They assign ownership.** Media accessibility work that belongs to everyone belongs to no one. Institutions need to identify who is responsible for accessibility review in video production workflows, who handles accommodation requests related to media, and who maintains standards and oversight. This doesn't require a large team, but it requires clarity.
**They plan for ongoing production.** Addressing the backlog is important, but new content is being produced continuously. A program that focuses only on existing content and has no plan for new production is perpetually behind. Workflow integration — making accessibility a standard part of how content is produced — is how institutions stop accumulating new backlog.
**They don't conflate platform features with program outcomes.** Turning on auto-captioning in a video platform is a technical step, not an accessibility program. Institutions sometimes confuse the availability of accessibility features with the presence of accessible content. A platform with auto-captioning enabled and no quality review process is not running an accessible media program.
---
Where to start
For institutions that are earlier in this work, a reasonable starting point is a honest assessment of current state: what content exists, where it lives, what accessibility features are in place, and what the gap looks like between current state and a defensible standard.
That assessment typically surfaces a few things: the scale of the backlog, the absence (or inconsistency) of production workflows, the gap between what platforms offer and what's actually being used, and the absence of clear ownership or standards.
From that baseline, most institutions benefit from addressing new production workflows first — stopping the accumulation of new inaccessible content — while developing a prioritization framework for existing content. Trying to remediate everything at once is rarely feasible; trying to address everything with equal priority is rarely effective.
Media accessibility is not a project with an end date. It's an ongoing operational commitment. Institutions that understand that early build better programs than those that treat it as a one-time compliance exercise.
---
*CairnWise Consulting helps higher education, healthcare, research, and nonprofit organizations navigate media accessibility decisions — from prioritizing backlog work to building production workflows and accessibility programs that last. [Request a consultation](#contact) to talk through where your organization is and what would help most.*