The L&D accessibility coordinator playbook: building the role, managing caption compliance at scale, and what to track on a dashboard

Why the accessibility coordinator role exists now

The accessibility coordinator as a distinct role in the L&D function is not a new concept, but its scope has expanded significantly in the past two years as the caption compliance landscape has matured from guidance to enforcement. Understanding why the role has become more necessary is important context for designing it correctly — the scope of the role should match the scope of the compliance exposure, and that exposure has grown substantially.

The enforcement timeline

The compliance framework that drives caption requirements in the United States has three primary pillars, each with a different enforcement mechanism and a different organizational applicability:

Section 508 of the Rehabilitation Act applies to federal agencies and to recipients of federal financial assistance. It requires that electronic and information technology, including training video, be accessible to people with disabilities. For federal contractors and universities that receive federal funds, Section 508 caption compliance has been a legal requirement for years — but enforcement was historically light-touch for training content. The Department of Justice's increased activity on digital accessibility complaints has changed that posture, and organizations that previously treated Section 508 as an annual checkmark are now seeing it as an active compliance obligation.

ADA Title II covers state and local government entities, including all public universities, community colleges, public K-12 districts, and municipal government employers. The Department of Justice issued final rules in April 2024 establishing specific ADA Title II digital accessibility requirements under the WCAG 2.1 AA standard, with an enforcement date of April 24, 2026 for the largest covered entities. For public higher education L&D and instructional technology teams, this created a hard deadline that elevated caption compliance from a project to a program — one that needed someone to own it.

ADA Title I and Title III apply to private employers (Title I) and public accommodations (Title III). For L&D teams at private companies, ADA Title I drives caption requirements for internal training through reasonable accommodation obligations — an employee who is deaf or hard of hearing can request accessible training content as a workplace accommodation, and the employer must provide it. ADA Title III creates additional exposure when a company runs an externally accessible training academy or customer education program, since the public-facing nature of those platforms brings them into the public-accommodation framework. The US caption compliance matrix breaks down which framework applies to which type of training content in detail.

The European dimension adds to this picture for any organization with EU operations or EU-based learners. The European Accessibility Act has been enforceable since June 2025 and requires that digital services, including e-learning and training platforms, meet EN 301 549 (which references WCAG 2.1 AA). Member-state enforcement agencies are beginning to act on complaints, and organizations that assumed the EAA had a soft enforcement period are now discovering otherwise.

From compliance project to compliance program

The distinction between a project and a program is operationally significant. A project has a start date, an end date, and a deliverable. A compliance project for captions looks like: audit the library, identify non-compliant clips, caption them, done. A program has no end date — it is a permanent operational function that ensures the organization remains compliant as content is added, LMS platforms change, vendors are replaced, and compliance requirements evolve. The accessibility coordinator role is a program function, not a project function.

Most L&D organizations went through a compliance project phase in 2024–2026 as the enforcement deadlines approached. They audited their libraries, identified the highest-priority clips, captioned them, and declared compliance. The problem with project-mode thinking is that it produces a compliant snapshot — a point-in-time state — not a compliant program. Every week of new content production, every LMS migration, every vendor change, and every glossary update creates new compliance risk. Without a program function that monitors and maintains compliance continuously, the organization's compliance posture degrades from day one after the project closes.

The accessibility coordinator role is the mechanism by which a compliance project becomes a compliance program. It owns the ongoing monitoring, the new-content gate, the audit trail that demonstrates continuous compliance over time, and the escalation paths for when the program encounters a failure mode.

The specific driver: distributed production at scale

The compliance pressure is sufficient to justify the role in many organizations, but the operational driver that makes the role urgent at almost every organization above 100 employees is the distributed production model. Prior to 2020, most training video production was centralized: an instructional designer or content producer recorded and edited video, submitted it to a caption vendor or an L&D admin, and uploaded the finished file to the LMS. Caption compliance was a step in a centralized workflow, and as long as that workflow was followed, every video got a caption track.

Post-2020, that model broke. The shift to hybrid and remote work created millions of async training videos produced outside the L&D workflow: Loom recordings from sales managers, Zoom recordings from engineering leads, Microsoft Teams recordings uploaded directly to SharePoint or Stream, and Vimeo or Wistia-hosted content produced by product teams. These videos are training content — they are watched by employees as part of their onboarding, compliance, or product training — but they were never submitted to the caption workflow. They have no caption tracks. And the L&D team, in most cases, has no visibility into them.

The remote and hybrid workforce captioning guide covers the audit and remediation workflow for this distributed production model in detail. The accessibility coordinator role is the function that prevents the problem from accumulating — by establishing the new-content gate, by working with IT and HR to ensure that publication-event triggers route new videos to the caption workflow automatically, and by running the periodic audits that catch content that slipped through.

What the accessibility coordinator owns: scope definition

The most important design decision for the accessibility coordinator role is scope. What does this person own? What do they not own? The confusion between "responsible for" and "does themselves" is the most common role-design failure, and it is usually the reason the role underperforms expectations.

What the accessibility coordinator owns (program scope)

The accessibility coordinator owns five program functions:

1. The compliance dashboard and audit trail. The accessibility coordinator is responsible for maintaining a current, accurate view of the organization's caption compliance posture: what percentage of the library is compliant, what percentage is in progress, what the QA pass rate is for new content, how the vendor is performing against SLA, and how the glossary is performing against accuracy targets. This is the reporting function — the deliverable that the L&D director and compliance or legal team need to answer a DOJ inquiry, an OCR complaint, or an accessibility audit. The dashboard and audit trail are the accessibility coordinator's primary artifacts.

2. The new-content gate. The accessibility coordinator owns the process that ensures every new video added to the LMS has a compliant caption track before it is published to learners. This is a workflow gate, not a production step — the accessibility coordinator does not produce captions, but they own the rule that says captions must be present and QA'd before publication, and they have the authority to block publication when that rule is not met. The new-content gate is the mechanism that prevents the compliance posture from degrading as new content is added.

3. The caption vendor relationship. The accessibility coordinator manages the captioning vendor relationship: SLA monitoring, accuracy benchmarking, escalation for failing clips, glossary updates, contract renewal preparation, and the vendor evaluation process when performance degrades or a contract comes up for renewal. The captioning vendor RFP playbook and the comparison pages for Rev, 3Play Media, and Verbit provide the framework for vendor evaluation; the accessibility coordinator is the internal owner of that process.

4. The caption governance documentation. The accessibility coordinator writes, maintains, and communicates the three governance documents that underpin the program: the caption policy (the rule), the caption procedure (the how), and the producer training (the what-everyone-else-needs-to-know). These documents change infrequently — typically once a quarter as compliance requirements evolve or as the organization's LMS platform or vendor changes — but they are the foundation on which everything else runs. Without clear documentation, every content producer makes individual judgment calls about when captions are required and what acceptable accuracy looks like, and the program loses integrity.

5. The periodic library audit. The accessibility coordinator runs a periodic audit of the existing caption library — typically annually for a mature program, quarterly during the initial program-build phase — to identify content that has fallen below the compliance threshold, content that was added outside the new-content gate, and content whose accuracy has degraded due to glossary updates that were not retroactively applied. The LMS caption audit methodology covers the five-dimension framework for prioritizing a large library audit in detail.

What the accessibility coordinator does not own

The accessibility coordinator does not produce captions. They do not operate the captioning software. They do not upload caption files to the LMS. They do not edit caption files. In small organizations where the accessibility coordinator is a hat worn by an L&D manager or instructional designer, these production tasks may be done by the same person who owns the program function — but the distinction matters because it defines what the role is accountable for. If the accessibility coordinator is accountable for caption production, they become a bottleneck in the content production workflow. If they are accountable for caption compliance, they are a program function that enables everyone else to produce content within a compliant framework.

The accessibility coordinator also does not own the LMS platform. LMS administration, including caption track configuration, format compatibility settings, and player settings, is typically owned by the LMS admin. The accessibility coordinator works with the LMS admin to ensure the platform is configured correctly for WCAG-compliant caption delivery, but they are not the LMS admin's manager and do not have platform admin rights in most configurations.

Similarly, the accessibility coordinator does not own the organization's broader accessibility program — the ADA coordinator function, the Section 504/508 coordinator at a university, or the digital accessibility team at a larger organization. Caption compliance for training video is a subset of the broader accessibility program. The accessibility coordinator should have a reporting relationship or a regular touchpoint with whoever owns the broader program, but the two functions are distinct.

The scope in practice: a one-page charter

The most useful artifact for defining the accessibility coordinator scope is a one-page charter that specifies: the role's five program functions, the authority the role has to block content publication, the reporting structure (who the accessibility coordinator reports to and who they report compliance status to), the resources available (vendor budget, staff hours, tooling), and the success metrics (library compliance percentage, QA pass rate, audit trail freshness, vendor SLA compliance rate). A charter of this kind takes 2–3 hours to write, requires sign-off from the L&D director and, ideally, from the compliance or legal function, and prevents the majority of scope disputes that undermine early-stage programs.

Building the role at different org sizes

The accessibility coordinator role does not look the same at a 50-person engineering startup and a 500-person healthcare organization, even if the compliance obligations are similar. Designing the role without calibrating to org size leads to either a role that is too lightweight to manage the compliance risk or a role that is too heavy for the organization to staff. Here is how the role typically scales across three size bands.

Small org (50–150 employees): the accessibility function

At a small organization, the accessibility coordinator is almost always a hat worn by an existing L&D team member — most commonly the L&D manager, a senior instructional designer, or a learning operations specialist. The time commitment is typically 2–5 hours per week, depending on the volume of new content and the maturity of the vendor relationship. The function is primarily administrative: maintaining the compliance dashboard, running the new-content gate checklist, managing the vendor relationship (submitting content for captioning and reviewing QA outputs), and updating the governance documents annually.

At this size, the full RACI is usually collapsed into two people: the accessibility coordinator hat (L&D manager or instructional designer) and the LMS admin. The content producer and the accessibility coordinator are often the same person. The vendor relationship is a recurring task rather than a strategic management function.

The biggest risk at small orgs is under-resourcing: the accessibility coordinator hat is added to an already full role without any time being removed, and the compliance function gets squeezed out by content production demands. The solution is to define the minimum viable accessibility function — typically: maintain the dashboard, run the new-content gate, manage the vendor relationship — and treat any time beyond that as program investment rather than baseline obligation.

Mid-size org (150–500 employees): the 20% allocation

At a mid-size organization with a dedicated L&D team, the accessibility coordinator function typically warrants a 20–30% allocation for one person, rather than being absorbed entirely into an existing full-time role. This allocation is enough to run the program functions described above, manage a vendor relationship that involves significant weekly volume, and begin building out the governance documentation and producer training that support a distributed production model.

At this size, the role often sits with a learning operations specialist, an instructional design manager, or a content quality specialist. The person in the role typically does not produce captions themselves but does manage the workflow end-to-end: submitting batches, reviewing QA outputs, escalating failures to the vendor, updating the glossary, and running the monthly dashboard review. The LMS admin is a separate person who handles the technical configuration and upload workflow.

The biggest risk at mid-size orgs is role clarity: without a written charter, the accessibility coordinator function bleeds into the LMS admin role (the admin starts doing QA work that belongs to the coordinator) or into the content production workflow (the coordinator starts doing caption editing that the vendor should own). The charter is more important at this size than at small orgs, where everyone knows everyone and informal coordination fills the gaps.

Large org (500+ employees): the dedicated role

At a large organization with a mature L&D function, distributed content production across multiple departments, and a significant existing caption library, the accessibility coordinator warrants a dedicated full-time role. At organizations with 1,000+ employees in compliance-intensive verticals (healthcare, financial services, government, higher education), the function may require two people — one focused on the new-content gate and vendor management, one focused on the library audit and remediation backlog.

At this size, the role has a strategic dimension that is absent in smaller configurations: the accessibility coordinator is managing vendor relationships across multiple platforms (separate captioning vendors for different content types or regions), influencing LMS selection decisions to ensure accessibility capabilities are in scope, working with legal and compliance on accessibility statements and audit responses, and building the producer training program that prevents the distributed production problem from creating a growing backlog of uncaptioned content.

The biggest risk at large orgs is the backlog problem: an existing library with tens of thousands of hours of uncaptioned legacy content, combined with a new-content gate that has not been running, creates a remediation backlog that is too large for the accessibility coordinator to manage alone. Prioritization frameworks (by view count, compliance classification, and content age) and batch vendor workflows are essential tools at this scale. The audit methodology post covers the triage framework in detail.

RACI for caption operations

Caption operations in most L&D environments involve four roles: the content producer, the accessibility coordinator, the LMS admin, and the caption vendor or caption specialist. The RACI for caption operations defines who is Responsible (does the work), Accountable (owns the outcome), Consulted (provides input), and Informed (receives updates) for each of the seven key caption operations activities.

RACI table: caption operations activities

Key RACI decisions and rationale

Why the accessibility coordinator is Accountable for caption file production but the vendor is Responsible. The vendor produces the file, but the accessibility coordinator owns the outcome — including whether the vendor performed within SLA and whether the output meets the WCAG threshold. If the vendor delivers a non-compliant file, the accountability sits with the coordinator to escalate the failure and ensure remediation. Assigning Accountability to the vendor for their own work output creates a conflict of interest in the QA process: the vendor has an incentive to over-report compliance.

Why the content producer submits but does not upload. Separating submission (producer's responsibility) from upload (LMS admin's responsibility) creates a natural QA gate: the caption file must pass QA review before the LMS admin has anything to upload. This prevents the common failure mode where a content producer uploads both the video and a caption file simultaneously, skipping QA entirely because the two steps are combined into one publish action. At small organizations where the same person plays both roles, the gate should be explicit in the checklist: QA must be documented before upload, even if the same person does both steps.

Why the accessibility coordinator owns glossary updates. Glossary governance is a compliance function, not a production function. A term that is added to the glossary incorrectly — misspelled, wrong phonetic encoding, wrong context tag — will produce substitution errors across all future content on that topic. The accessibility coordinator owns the accuracy of the glossary because they are accountable for the QA outcomes that the glossary drives. In practice, the coordinator typically receives term suggestions from content producers and subject-matter experts, validates them against the phonetic and contextual standards, and approves additions to the production glossary. The glossary architecture guide covers the term sourcing, validation, and ingestion process in detail.

Why "Publish to learners" has a gate-approval step for the accessibility coordinator. The gate-approval step is the mechanism by which the new-content gate functions. The accessibility coordinator's approval is the signal that the QA step has been completed and the clip meets the WCAG threshold. Without an explicit gate-approval step in the RACI, the new-content gate exists in policy but not in workflow — producers publish directly, QA is bypassed, and the gate exists only on paper. The approval does not need to be a formal sign-off document; it can be a Slack message, a status column in a shared tracking spreadsheet, or a field in the LMS workflow. What matters is that it is an explicit step that cannot be skipped.

LMS-specific caption management for the accessibility coordinator

One of the most practically demanding aspects of the accessibility coordinator role is managing caption compliance across multiple LMS platforms with different caption management interfaces, different format support, and different administrative workflows. This section covers the accessibility coordinator's view of caption management on eleven of the most common L&D platforms: what the admin interface looks like, what the coordinator needs to monitor, where compliance gaps typically appear, and what configuration decisions the coordinator should influence.

TalentLMS

TalentLMS caption management is handled through the Unit editor for each lesson. Caption files (SRT or VTT) are attached at the unit level via the "Captions" tab in the video unit editor. The accessibility coordinator's compliance check on TalentLMS has three components: (1) verify that the Captions tab for each video unit shows an attached caption file rather than the default empty state, (2) verify that the caption file is in the correct format (SRT is preferred over VTT for TalentLMS due to historical rendering quirks in some browser contexts), and (3) verify that the caption track is enabled by default rather than requiring the learner to manually activate it.

The most common compliance gap on TalentLMS is the auto-caption toggle: TalentLMS supports YouTube embeds natively, and YouTube-embedded videos display YouTube's auto-generated captions by default. These auto-captions are not WCAG 2.1 AA compliant — they consistently score 82–91% on training content with technical vocabulary, well below the 99% WCAG threshold. The accessibility coordinator should configure all YouTube embeds to disable YouTube's auto-captions and attach a WCAG-compliant caption file instead, or replace YouTube embeds with direct video uploads where the caption track is fully controllable. The branch-level compliance check for TalentLMS should include a filter for embedded-video units specifically.

Docebo

Docebo's caption management is handled at the asset level in the Central Repository. Caption files are attached to video assets, and the caption track propagates to all course usages of that asset. This asset-level attachment model has an important implication for the accessibility coordinator: a single caption update to an asset in the Central Repository corrects the caption track in every course that uses that asset, which is more efficient than per-course caption management. But it also means that a glossary update applied to a new version of a caption file must be pushed through the Central Repository, not applied at the course level.

The accessibility coordinator's compliance dashboard for Docebo should pull from the Central Repository's asset library, not from individual courses. Docebo's reporting module can generate a list of video assets without caption files attached — the accessibility coordinator should run this report monthly and treat the output as the remediation backlog. Docebo also supports subtitle tracks in multiple languages for the same asset, which the coordinator needs to track separately if the organization has multilingual compliance obligations under the multilingual caption workflow.

Absorb LMS

Absorb LMS caption management operates through the File Manager for each lesson. Caption tracks are attached in the SCORM/video lesson editor, and Absorb supports both VTT and SRT format uploads. The accessibility coordinator's key configuration check for Absorb is the "Show Captions by Default" toggle in the video player settings — Absorb defaults to captions off, requiring learners to activate the caption track manually, which is a WCAG compliance issue because WCAG SC 1.2.2 requires synchronized captions without requiring user action to activate them.

Absorb's compliance reporting is less robust than Docebo's — there is no native report that lists all video lessons without attached caption tracks. The accessibility coordinator typically manages this gap with a manually maintained tracking spreadsheet that maps Absorb course IDs to caption track status, updated during the monthly audit cycle. Some Absorb customers use the Absorb API to build a custom compliance report that can be run on demand; this is worth the investment for organizations with libraries above 500 courses.

Cornerstone OnDemand

Cornerstone OnDemand caption management is handled through the Learning Object editor in the course catalog. Cornerstone supports caption file upload at the content object level for native video content. For SCORM-packaged content, caption files must be embedded in the SCORM package at the authoring stage (Articulate Storyline, Rise, Lectora), which requires the accessibility coordinator to influence the content production workflow upstream — ensuring that instructional designers include caption tracks in their SCORM export settings, not just in preview mode.

Cornerstone's compliance gap for the accessibility coordinator is SCORM-packaged content: unlike native video uploads where the caption track is a separate controlled attachment, SCORM packages are opaque at the platform level. The platform cannot report on whether a SCORM package has an accessible caption track without opening the package and inspecting it. The accessibility coordinator's control over SCORM-packaged content compliance is therefore at the authoring stage, not the LMS admin stage. The governance procedure document should specify that all SCORM packages must be pre-screened for caption track presence before upload to the catalog.

Workday Learning

Workday Learning caption management is handled through the Media Library, where video assets are managed separately from course enrollments. Caption tracks (SRT or VTT) are attached to media assets in the Media Library, and the accessibility coordinator's compliance check runs against the Media Library asset catalog rather than against individual learning content objects. Workday Learning's integration with Microsoft Viva Learning and the Workday Skills Cloud means that videos pushed from external sources (Microsoft Stream, LinkedIn Learning, Coursera) may not have caption tracks attached in the Workday environment even when they have captions in their native platform — the accessibility coordinator needs to maintain a separate compliance check for externally sourced content.

Workday Learning's administrative interface for caption management is less intuitive than dedicated LMS platforms, and changes are typically managed through HCM administration credentials that the L&D team may not have direct access to. The accessibility coordinator should confirm the access level available to the L&D function for Media Library management at the start of the program design, and escalate to HR/IT if Media Library admin access is not available to the L&D team.

Kaltura

Kaltura's caption management is the most technically flexible of any major LMS platform, and also the most complex to administer compliantly. Kaltura supports multiple caption tracks per video asset (including machine-generated captions from Cielo24 or NOVA AI, human-edited tracks, and uploaded SRT/VTT files), each with a separate accuracy status, language designation, and publication state. The accessibility coordinator's compliance task on Kaltura is ensuring that the default caption track — the one served to learners who do not select a specific track — is the WCAG-compliant human-edited or glossary-enhanced track, not the machine-generated draft track that Kaltura creates automatically for all ingested videos.

Kaltura's machine-generated captions are created automatically on ingest and set as the default track if no higher-quality track is attached. This means that for every new video added to a Kaltura-managed library, the default state is a non-compliant caption track unless the accessibility coordinator's new-content gate explicitly includes a "set compliant caption track as default" step in the Kaltura Caption Management interface. The accessibility coordinator should configure a Kaltura notification rule that sends an alert when a new video is ingested without a manually-set caption default, to prevent the automatic machine-generated caption default from silently propagating to learners.

Panopto

Panopto's caption management is primarily used in higher education environments for lecture capture. Panopto generates automatic Speech-to-Text (STT) captions for all uploaded recordings, which creates a similar compliance challenge to Kaltura: the automatic captions are displayed by default, they are not WCAG-compliant (typical accuracy 84–91% for lecture content with technical vocabulary), and they create a false sense of compliance — the caption track is present, the caption track is active, but the caption track fails the WCAG 2.1 AA accuracy threshold.

For a university's accessibility coordinator, the Panopto STT default is the single largest source of caption compliance risk in the library. The recommended configuration for a compliant Panopto deployment is: disable the STT caption display toggle by default (preventing auto-captions from being shown to learners), enable STT as a draft only (making the STT output available as an editing starting point but not serving it as a caption track), and require all published recordings to have a manually edited or vendor-produced caption track marked as the primary caption track before the video is shared with students. This configuration requires working with Panopto's admin settings and, potentially, with Panopto's customer success team to implement site-wide defaults rather than per-recording settings.

LearnUpon

LearnUpon caption management handles caption tracks as attachments to video modules within course definitions. The accessibility coordinator's compliance check for LearnUpon runs through the Courses section of the admin dashboard, verifying that each video module has a caption file attached. LearnUpon supports SRT and VTT uploads natively. LearnUpon does not have a native compliance report for caption track presence, so the accessibility coordinator typically manages this through a combination of the course audit checklist and a manually maintained tracking spreadsheet.

One LearnUpon-specific consideration for the accessibility coordinator is the SCORM content integration: LearnUpon is frequently used with SCORM packages from Articulate Storyline and Rise, and the same SCORM compliance gap that exists in Cornerstone applies here — the platform cannot verify caption track presence inside a SCORM package without opening and inspecting it. The procedure document for LearnUpon should specify pre-upload SCORM inspection as a required step.

360Learning

360Learning's caption management is integrated into the Collaborative Learning platform's module editor. Caption tracks are attached to video modules as subtitle files, and 360Learning supports SRT format uploads with language selection. 360Learning's collaborative authoring model — where subject-matter experts and peers co-create content — creates a distributed production compliance challenge: any team member with author access can add video content to a learning path without going through the L&D team's caption workflow. The accessibility coordinator needs to configure 360Learning's author permissions to require L&D review before content with video modules is published to learners, and to establish a submission channel for authors to request caption file production.

Bridge LMS

Bridge LMS caption management for video content is handled through the Course Library, where caption files are attached to video assets as SRT or VTT uploads. Bridge's accessibility compliance features include an automatic accessibility check during course publishing that flags courses missing caption tracks, which provides a useful built-in gate for the accessibility coordinator. However, the Bridge accessibility check does not verify caption accuracy — it only verifies the presence of a caption file. The accessibility coordinator's QA process must still run separately from the Bridge built-in check.

Rippling Learning Management

Rippling's LMS, as part of the broader Rippling HR platform, manages video content with caption track support at the module level. Caption files (SRT) are attached to video modules through the module editor. The accessibility coordinator's consideration for Rippling is the HR system integration: Rippling's compliance training completions and training assignments are tied to the HR system's employee records, which means that compliance training content — ADA, OSHA, harassment prevention — is tracked directly in the HR record. A compliance gap in caption tracks on mandatory training content in Rippling creates a documented, auditable record of employees having been exposed to non-compliant training content, which is a higher legal risk than a gap in optional or development-focused content.

The accessibility coordinator dashboard: what to track and where the data comes from

The compliance dashboard is the primary artifact that the accessibility coordinator produces for the L&D director and the compliance or legal function. A well-designed dashboard answers the questions that matter in an enforcement scenario without requiring the accessibility coordinator to manually generate a report: are we compliant now? are we improving? and what is the risk exposure?

The seven metrics that belong on the dashboard

Metric 1: Library compliance percentage. The percentage of video content in the LMS library that has a WCAG-compliant caption track (defined as: a caption file present, attached as the default track, and last QA'd within the audit cycle). This is the headline metric. A compliant caption track means it has been QA'd at the DCMP protocol level and achieved a score at or above the 99% threshold — not merely that a caption file exists. A library compliance percentage of 97% means 3% of the library is potentially non-compliant, and the accessibility coordinator should know which clips are in that 3%.

Data source: The compliance tracking spreadsheet or, for organizations that have automated tracking, a query against the LMS API that returns video assets with and without attached caption tracks, cross-referenced against the QA log. The LMS admin typically provides the raw asset list; the accessibility coordinator maintains the QA status layer.

Metric 2: New content gate compliance rate. The percentage of new video content published in the current period (week or month) that went through the new-content gate and was published with a compliant caption track. A 100% gate compliance rate means no new content slipped through without captions. A rate below 100% means the gate has gaps — either the workflow is not being followed or the gate process has a loophole that allows publication before caption QA is complete.

Data source: The content production workflow log. For organizations using a project management tool (Asana, Jira, Monday.com, Notion) for content production, the accessibility coordinator should add a caption gate approval step to the production workflow and track completion of that step. For organizations without a project management system, the LMS admin can pull a list of new video assets published in the period and the accessibility coordinator cross-references against the QA log.

Metric 3: QA pass rate for new content. Of the new content that went through the QA process in the current period, what percentage passed the DCMP-protocol accuracy threshold on first submission from the vendor? A high first-pass QA rate (95%+) indicates the vendor is performing within SLA and the glossary is well-seeded for current content topics. A low first-pass rate (below 90%) signals either a vendor performance problem, a vocabulary gap in the glossary, or a content production quality issue (audio problems). This metric is the earliest warning signal for a vendor relationship that is degrading.

Data source: The QA log, which the accessibility coordinator maintains. Each QA review generates one row: clip identifier, submission date, QA date, score, pass/fail, error type breakdown, and whether the clip required resubmission. The first-pass rate is the percentage of QA rows that show "pass" on the first review.

Metric 4: Glossary term count and update cadence. The current number of terms in the production glossary and the date of the last update. A glossary that has not been updated in 60+ days in a content-active environment is probably missing new product names, personnel changes, or vocabulary shifts. The accessibility coordinator should review the glossary update cadence monthly and flag any domain where the content team has produced new content but no new glossary terms have been submitted for that domain in the same period.

Data source: The glossary management system — the source of truth for the production glossary, whether that is GlossCap's glossary interface, a shared Notion database, a Confluence page, or a structured spreadsheet. The accessibility coordinator owns the glossary log that records every addition, correction, and deprecation.

Metric 5: Vendor SLA compliance rate. The percentage of caption jobs submitted to the vendor in the current period that were returned within the contracted turnaround time. A 98%+ SLA compliance rate indicates a functional vendor relationship. A rate that drops below 90% in a given period is an escalation trigger — the accessibility coordinator should open a formal vendor inquiry and document the response as part of the SLA monitoring record.

Data source: The vendor submission log, which records the submission timestamp and the delivery timestamp for each job. Many captioning vendors provide a job status dashboard or an API that can feed this tracking automatically; the accessibility coordinator should confirm the data export method with the vendor at the start of the relationship and build it into the monthly reporting cycle.

Metric 6: Audit trail freshness. The date of the most recent completed library audit entry in the compliance documentation. The audit trail serves as the evidence that the program is running continuously, not just as a point-in-time exercise. A stale audit trail — where the most recent entry is more than 90 days old for a large library, or more than 12 months old for a mature, low-volume library — is a compliance documentation gap that the accessibility coordinator needs to address before an audit or complaint.

Data source: The compliance documentation, which the accessibility coordinator maintains. The audit trail typically consists of the periodic audit report (date, scope, clips reviewed, findings, remediation actions taken) plus the QA log (which serves as a continuous audit trail for new content). The dashboard should surface the date of the most recent audit entry so the coordinator can see at a glance whether the documentation is current.

Metric 7: Remediation backlog count. The number of video content items currently in the compliance library that have been identified as non-compliant but not yet remediated. This is the gap between the current library compliance percentage and 100% — quantified as a count of clips with outstanding remediation actions. A shrinking remediation backlog indicates the program is making progress against the historical gap. A growing backlog indicates that new non-compliant content is being identified faster than remediations are being completed, which is a resource allocation signal for the L&D director.

Data source: The remediation tracking log, which the accessibility coordinator maintains as part of the audit cycle. Each identified non-compliant item should have a record that includes: clip identifier, non-compliance reason (no caption track, caption track below QA threshold, wrong format, not set as default), date identified, assigned to (who is responsible for the remediation action), and target remediation date.

Dashboard format and reporting cadence

The dashboard does not need to be a sophisticated tool. A well-structured spreadsheet with one row per metric, current value, prior-period value, and a trend indicator (arrow or color coding) is sufficient for most organizations. The metrics should update monthly for a mature program (quarterly library compliance, weekly gate compliance and QA pass rate). The accessibility coordinator should plan a 30-minute monthly dashboard review with the L&D director, with a quarterly review that includes the compliance or legal function.

For organizations that have compliance reporting requirements — universities responding to OCR requests, federal contractors under Section 508 reporting obligations, organizations subject to EAA reporting in EU member states — the dashboard should be structured so that the relevant metrics can be exported directly into a compliance report format without manual reformatting. A compliance report typically requires: the library compliance percentage at the current date, the methodology used to determine compliance (DCMP protocol, 99% threshold), the audit trail showing when compliance was measured and how, and the remediation plan for any known gaps.

Caption governance: policies, procedures, and producer training

The three governance documents that the accessibility coordinator owns — the caption policy, the caption procedure, and the producer training — are the foundation of the compliance program. Without them, caption compliance depends entirely on individual knowledge and good intentions. With them, the program runs consistently even when the accessibility coordinator is on leave, the L&D team changes, or the organization expands into new content production patterns.

The caption policy: the rule document

The caption policy is a short (one to two page) document that answers four questions: what does it cover, what does it require, who is accountable, and what happens when it is not followed. The policy is written for a leadership audience — it should be approvable by the L&D director and the compliance or legal team without deep technical knowledge of captioning.

A functional caption policy for a mid-size organization includes:

• Scope: The policy covers all video content hosted in or linked from the organization's LMS that is viewable by employees as part of their onboarding, compliance, or professional development training. It does not cover video content hosted on external platforms and accessed via deep link without integration into the LMS (though the procedure document may address this separately).
• Requirement: All in-scope video content must have a synchronized caption track that meets WCAG 2.1 AA SC 1.2.2, defined as: a caption file attached as the default track, covering all speech and significant non-speech audio, with a word-level accuracy score of 99% or above measured using the DCMP Captioning Key protocol.
• Timeline for new content: New video content must have a compliant caption track attached before it is published to learners. Content may be in draft state without captions; it may not be accessible to learners without captions.
• Timeline for existing content: Existing content that does not have a compliant caption track must be remediated within 12 months of the policy adoption date (or a shorter period for compliance-classified content).
• Accountability: The accessibility coordinator is accountable for monitoring compliance and maintaining the audit trail. Content producers are responsible for submitting content for captioning before publication. LMS admins are responsible for ensuring the caption track is attached as the default before content is published.
• Exceptions: Video content for which captioning is technically infeasible (e.g., content with audio quality too degraded for accurate transcription) must be documented with an accessibility statement that discloses the limitation and provides an alternative means of access.

The caption policy should be reviewed annually and updated when compliance requirements change (e.g., new enforcement guidance from DOJ/OCR, new requirements under EAA member-state implementation), when the LMS platform changes, or when the vendor relationship changes in a way that affects the capability definition.

The caption procedure: the how document

The caption procedure is the operational document that translates the policy into a step-by-step workflow. Unlike the policy (which is for leadership), the procedure is for the people who do the work: content producers, LMS admins, and the accessibility coordinator themselves. It should be specific enough that a new team member can follow it without additional training.

A functional caption procedure includes the following sections:

The new-content captioning workflow: Step-by-step instructions for how a new video moves from production completion to caption file attached and published. This includes: where to submit the video for captioning (vendor submission form URL, email address, or API), what metadata to include in the submission (clip title, duration, content category, vocabulary domain, any known proper nouns not in the current glossary), the turnaround time to expect (per the vendor SLA), how to verify that the caption file meets the WCAG threshold (QA checklist or protocol reference), how to attach the caption file to the LMS platform (platform-specific steps for each LMS in use), and how to confirm that the caption track is set as the default for learners.

The glossary update workflow: Instructions for how a content producer or subject-matter expert submits a new term for addition to the production glossary. This includes: where to submit term suggestions (shared form, email to the accessibility coordinator, or directly in the glossary management system), what information is required for a valid submission (term, correct spelling, phonetic pronunciation if non-standard, context tags for domain routing, and an example sentence), the review and approval process (accessibility coordinator validates and approves within N business days), and how the term is propagated to the production captioning system.

The QA review workflow: Step-by-step instructions for how the accessibility coordinator runs a DCMP-protocol QA review on a caption file returned by the vendor. This should reference the caption QA methodology guide for the full protocol description and focus on the specific steps required for the organization's implementation: which clips to sample, how to document the review, how to record the score and error breakdown, and how to trigger a vendor resubmission if the clip fails.

The vendor resubmission and escalation workflow: Instructions for what to do when a caption file fails QA. This includes: how to format the error report to the vendor (clip ID, timestamps, error list, corrected text, accuracy score, failure reason), the expected vendor turnaround time for a correction, and the escalation path if the corrected file still fails (second resubmission, account manager escalation, contract remediation clause activation).

Producer training: the what-everyone-needs-to-know document

The producer training is not a formal training course — it is a one-page or two-page reference document that every content producer should receive at onboarding and that should be linked from the content production project template. It answers the questions a content producer most often has about caption compliance without requiring them to read the policy or the procedure in full:

• Do I need captions on this video? (Answer: yes, if it is in the LMS and viewable by employees.)
• How do I get captions? (Answer: submit the video here [link to submission form/contact]. Turnaround is N business days.)
• Can I use YouTube auto-captions? (Answer: no — YouTube auto-captions do not meet the WCAG accuracy threshold for training content with technical vocabulary.)
• Can I use Zoom's auto-transcript? (Answer: no — Zoom transcripts are not synchronized caption files and do not meet the WCAG threshold.)
• I recorded a Loom. What do I do? (Answer: [link to Loom captioning workflow in the procedure document].)
• I added a new product name or technical term that the captions keep getting wrong. What do I do? (Answer: submit the term here [link to glossary submission form]. The coordinator will add it within N business days.)
• My video is already recorded and I need to publish it urgently. Can I publish without captions? (Answer: no — content may not be published to learners without a compliant caption track. Contact [accessibility coordinator name/email] to request expedited captioning.)

The producer training document should be updated each time the submission process, the vendor, or the platform changes. It should be distributed to all new content producers at onboarding and re-sent to the full content production team when significant changes occur.

Managing caption compliance at scale: batch workflows, prioritization, and the new-content gate

As the library grows and the content production team expands, the accessibility coordinator's operational challenge shifts from "how do I build the program" to "how do I keep the program running at scale without it consuming all of my time." Three mechanisms drive scale efficiency: the batch workflow, the prioritization framework, and the new-content gate as a self-sustaining process rather than a manual review step.

The batch captioning workflow

Rather than submitting videos to the caption vendor one at a time as they are produced, the batch workflow aggregates all new content submissions over a fixed period (typically one week) and submits them as a batch on a fixed schedule (e.g., every Monday morning). This approach has several operational advantages: it reduces the administrative overhead of individual submissions, it allows the vendor to optimize their workflow around predictable batch volumes, and it allows the accessibility coordinator to run QA on a batch of returned files in a single review session rather than checking files sporadically throughout the week.

A batch workflow works best when the content production cadence is predictable. For organizations with a sprint-based or project-based content production model, a weekly batch aligns naturally with the production cycle. For organizations with highly variable production cadences — where some weeks produce no new content and others produce 20 clips — the batch workflow may need to be triggered by volume threshold rather than fixed schedule: submit a batch when 10 or more clips are queued, regardless of which day of the week that threshold is reached.

The batch workflow requires a submission queue — a shared location where content producers drop finished videos (and their metadata) when they are ready for captioning. This can be as simple as a dedicated folder in SharePoint or Google Drive with a metadata form, or as sophisticated as an integration between the content production project management tool and the vendor's submission API. The submission queue is the input to the batch; the vendor's delivery of completed caption files is the output; and the QA review is the step that clears each batch for LMS upload.

The prioritization framework for the remediation backlog

The remediation backlog — the set of existing library content that does not yet have a compliant caption track — is always larger than the resources available to remediate it. The accessibility coordinator's prioritization framework determines which clips get remediated first. The framework typically uses three dimensions:

Compliance classification: Content that is legally required to be accessible under a specific compliance framework (OSHA safety training, HIPAA training, ADA-mandated content, compliance-required harassment prevention or EEO training) has higher remediation priority than content that is compliance-adjacent or voluntary. The rationale: a DOJ inquiry or OCR complaint is most likely to target content that is required by law to be accessible, and the legal exposure for non-compliant mandatory training content is higher than for optional development content.

View count: Content that is viewed more frequently has higher remediation priority than rarely-viewed content. A 5-year-old compliance video that 500 employees have watched in the past 12 months is a higher priority than a one-year-old product training video that 12 employees have watched. The view count dimension can be pulled from LMS analytics for most platforms.

Content age: Very old content (3+ years) in rapidly-evolving domains (technical product training, clinical protocols) may have accuracy problems that go beyond caption compliance — the content itself may be outdated. For these clips, the accessibility coordinator's recommendation may be "retire or update" rather than "remediate captions," which removes the clip from the remediation backlog rather than adding a caption file to non-current content. Flagging these cases to the L&D director is an important part of the audit function.

A practical remediation prioritization grid assigns each backlog item a score based on these three dimensions (1–3 for each) and sorts the backlog by total score. The top quartile gets remediated in the current quarter; the remaining quartiles are scheduled into future quarters. The caption compliance program guide includes a 90-day sprint template that maps this prioritization framework to an initial program build.

The new-content gate as a self-sustaining process

In a mature program, the new-content gate should operate with minimal accessibility coordinator involvement for the majority of content. The goal is for the gate to run through the content production workflow — so that the caption submission step is a standard part of how content producers work, not a special approval process that requires the coordinator's attention for every clip. Three mechanisms support this transition:

LMS publication controls: Some LMS platforms (Cornerstone, Docebo, Kaltura) support publication workflow controls that can require an approval step before content is published to learners. The accessibility coordinator should work with the LMS admin to implement a publication gate that requires caption track attachment and coordinator (or designee) approval before a course with video content is published. When the gate is built into the platform, it cannot be bypassed without an explicit override — which means the accessibility coordinator is only in the workflow for exceptions, not for every clip.

Producer self-service caption management: When content producers can submit their own content for captioning through a well-documented process and receive the caption file back directly (without the accessibility coordinator as an intermediary), the gate operates at the producer level. The coordinator's role shifts from reviewing every caption file to reviewing QA reports and exception flags. This transition requires a vendor with a self-service submission interface that is accessible to non-specialist users, which is a selection criterion worth prioritizing in a vendor RFP.

Automated compliance checks: For organizations using GlossCap or similar tools that provide automated QA scoring, the QA step can be partially automated — the tool scores the returned caption file against the accuracy threshold and flags failures automatically, without requiring the coordinator to manually run the DCMP protocol on every clip. The coordinator reviews flagged failures; non-flagged clips are cleared through the gate automatically. This significantly reduces the QA overhead for high-volume programs.

Vendor management from the accessibility coordinator seat

The accessibility coordinator's relationship with the caption vendor is a program management relationship, not a purchasing relationship. The accessibility coordinator does not typically sign the contract (that is typically the L&D director or a procurement function), but they are the primary operational contact: they submit content, review outputs, escalate failures, manage glossary updates, monitor SLA performance, and provide the data for contract renewal decisions. Getting this relationship right is one of the highest-leverage activities the accessibility coordinator performs.

What the accessibility coordinator needs from the vendor at program start

When the vendor relationship is established (or inherited), the accessibility coordinator should confirm the following operational elements within the first 30 days:

The submission workflow: Exactly how to submit a batch of videos for captioning — the URL, credentials, metadata requirements, and file format requirements. The accessibility coordinator should run a test batch of three to five clips to validate the workflow end-to-end before submitting production content.

The turnaround commitment: The contracted turnaround time for standard submissions and for expedited submissions, the hours of operation for the support contact (relevant if the organization is in a different time zone from the vendor's operations center), and the process for requesting same-day turnaround for urgent content.

The per-customer glossary: How the vendor manages the organization's term glossary — where it lives in their system, how the accessibility coordinator submits additions, how quickly new terms propagate to production, and whether corrections to existing terms retroactively affect clips that are already in the queue. This is especially important for organizations in technical, medical, or specialized verticals where glossary accuracy is the primary driver of caption quality. The caption feedback loop guide explains why the glossary management model is a stronger predictor of long-term accuracy than the vendor's baseline accuracy rate.

The QA failure resolution process: What happens when the accessibility coordinator flags a clip as failing QA. Specifically: the format for submitting a correction request (clip ID, error list, corrected text, accuracy score), the turnaround time for corrections, whether the correction is applied to the original file or produces a new file, and the escalation path for repeated QA failures on the same content type.

The accuracy reporting data: Whether the vendor provides accuracy data on delivered files (some vendors include a confidence score or an internal accuracy estimate), and whether that data is available via API or requires manual export. The accessibility coordinator should not rely on vendor-reported accuracy as a substitute for independent DCMP-protocol QA — vendor confidence scores are calibrated against the vendor's own benchmark, not the WCAG threshold — but vendor-reported data can be a useful early signal for content that warrants closer QA review.

Monitoring vendor performance and managing SLA failures

The vendor SLA compliance rate is one of the accessibility coordinator's seven dashboard metrics because it is the earliest operational signal of a vendor relationship that is degrading. SLA failures typically cluster around predictable patterns: end-of-quarter backlogs at vendor operations centers, staffing shortages, and periods of high industry-wide captioning demand (post-compliance-deadline filing periods, large conference season). An accessibility coordinator who monitors the SLA compliance rate in real time can identify these patterns early and proactively communicate with the vendor before the backlog becomes a program risk.

When SLA failures occur, the accessibility coordinator's escalation path should be clearly defined in the procedure document. A first-instance failure (one batch with a delay of less than 48 hours) warrants a vendor notification and a logged entry in the SLA monitoring record. A pattern of failures (three or more batches with delays in a 30-day period) warrants a formal vendor inquiry to the account manager with a documented response request. A persistent pattern (SLA compliance below 85% for two consecutive months) warrants a contract review conversation that references the SLA remediation clause — most captioning vendor contracts include a clause allowing the customer to request a service credit or contract renegotiation when SLA commitments are persistently missed.

The accessibility coordinator should document every SLA failure, every formal inquiry, and every vendor response in the vendor management log. This documentation is the evidence base for contract renewal decisions: it demonstrates whether the vendor has been meeting commitments, whether failures have been addressed, and whether the relationship warrants renewal or replacement.

Vendor evaluation and RFP from the accessibility coordinator seat

When the vendor contract comes up for renewal or when persistent performance problems warrant a vendor replacement review, the accessibility coordinator is typically the primary input to the evaluation process. The captioning vendor RFP playbook covers the full procurement process in detail. From the accessibility coordinator's perspective, the most critical evaluation criteria are:

Per-customer glossary architecture: Does the vendor maintain a per-customer glossary that persists across all submitted content, or do they require term resubmission for each batch? A vendor without persistent per-customer glossary support cannot deliver the compounding accuracy improvement that is the primary value proposition of specialized captioning for training content. This is a pass/fail evaluation criterion, not a ranking criterion.

QA failure resolution turnaround: What is the vendor's committed turnaround time for a resubmission following a QA failure? A 24-hour correction turnaround allows the new-content gate to process failures without blocking the content publication schedule by more than one business day. A 5-day correction turnaround creates a content production bottleneck that pressures the accessibility coordinator to accept failing clips rather than hold them at the gate.

Batch submission capability: Does the vendor support automated batch submission via API, or does every submission require a manual form fill? For organizations submitting 10+ clips per week, manual form-fill submission creates an administrative overhead that scales poorly. API submission capability is a strong preference criterion for high-volume programs.

HIPAA BAA availability: For healthcare organizations, the caption vendor must be willing to sign a HIPAA Business Associate Agreement covering any protected health information that may appear in the audio of training videos. A vendor that cannot provide a BAA is not a viable option for healthcare training content, regardless of their accuracy performance. As documented in the RFP playbook, this is a pre-qualification criterion that can eliminate vendors early in the evaluation process.

Eight failure modes that stall caption compliance programs

Most caption compliance programs that are well-designed at launch end up stalling at 70–80% library compliance rather than driving to 99% and holding. The stall is almost always caused by one of eight identifiable failure modes. Understanding these failure modes allows the accessibility coordinator to build mitigation into the program design before they occur.

Failure mode 1: The new-content gate is not actually blocking publication

The most common failure mode is a new-content gate that exists in the policy and procedure documents but does not actually prevent non-compliant content from reaching learners. This happens when the gate is implemented as a soft check rather than a hard stop — the procedure says "content producers should request captions before publishing," but the LMS platform allows publication without a caption track and no one is monitoring for uncaptioned new content until the quarterly audit.

Diagnosis: Pull a list of all video content published in the LMS in the past 30 days and check each item for a compliant caption track. If more than 5% are missing a caption track or have a non-QA'd caption track, the gate is not functioning as a hard stop. Compare this with the gate compliance rate on the dashboard — if the dashboard shows 100% gate compliance but the audit finds gaps, the data source for the gate metric is wrong.

Mitigation: Implement a platform-level publication control on the LMS that prevents publication of video content without a caption track attached. For platforms that do not support this natively, implement a process control: the LMS admin requires the accessibility coordinator's gate-approval signal before publishing any new video content. The soft check procedure becomes a hard stop workflow.

Failure mode 2: Captions exist but fail the accuracy threshold

A library compliance percentage based on caption track presence rather than caption track quality will overstate compliance. Many organizations measure library compliance as "percentage of clips with a caption file attached" — which counts a Zoom auto-transcript at 84% word accuracy the same as a DCMP-protocol-verified 99% caption file. When the compliance metric measures presence and the compliance standard requires accuracy, the dashboard is reporting a number that does not reflect the legal standard.

Diagnosis: Sample 20 clips from the library that show as "compliant" in the presence-based tracking. Run each one through the DCMP spot-check protocol. If more than 20% of the "compliant" clips score below 99%, the compliance percentage on the dashboard is materially overstated.

Mitigation: Redefine the library compliance metric to require both presence and QA-verified accuracy. Re-run the audit against the new definition. The compliance percentage will drop, but it will reflect actual compliance rather than caption-file presence. The remediation backlog will grow, but it will represent actual work required to reach compliance rather than an optimistic understatement.

Failure mode 3: The glossary is not being maintained

Caption programs that launch with a well-seeded glossary and strong initial accuracy often experience gradual accuracy degradation over 6–12 months as new product releases, personnel changes, and domain vocabulary shifts go untracked in the glossary. The caption accuracy at month 12 looks worse than it did at month 1, and the accessibility coordinator attributes it to vendor performance rather than glossary staleness.

Diagnosis: Compare the first-pass QA pass rate for content produced in the first three months of the program (when the glossary was most current) with the first-pass rate for content produced in the most recent three months. If the rate has dropped and the error type breakdown shows predominantly substitution errors rather than formatting or deletion errors, the glossary is the primary cause. Confirm by checking whether the substitution errors cluster around recently-changed vocabulary (new product names, new team names, new technical terms introduced in the past 6 months).

Mitigation: Establish a glossary update cadence that is triggered by content production, not by calendar. Every time a content producer submits a new video for captioning, the submission form should include a field for "new terms in this video" — this captures vocabulary at the point of production, before the accuracy failure occurs. The accessibility coordinator should also run a quarterly glossary audit against the product changelog, the org chart, and any domain-specific vocabulary updates (new drug approvals, new regulatory terminology, new product SKUs) to catch vocabulary that was not submitted through the producer workflow.

Failure mode 4: SCORM-packaged content is invisible to the compliance dashboard

SCORM packages are opaque to the LMS platform — the platform receives a ZIP file and can report that the SCORM package is published, but it cannot inspect the package contents to verify whether the video inside has a caption track. Organizations that measure compliance at the LMS asset level miss all SCORM-packaged content, which in many L&D environments represents 30–60% of the training library (all content produced in Articulate Storyline, Rise, Lectora, or Adobe Captivate).

Diagnosis: Count the SCORM packages in the LMS catalog and the percentage of the library they represent. If more than 20% of the library is SCORM-packaged content and the compliance dashboard is not separately tracking SCORM compliance, the dashboard has a structural blind spot.

Mitigation: Implement a SCORM pre-upload inspection step in the content production procedure. Before any SCORM package is uploaded to the LMS, the accessibility coordinator (or a designee with the relevant authoring tool access) opens the package and confirms that the video assets inside have a caption track embedded. The SCORM inspection is logged in the compliance tracking spreadsheet alongside the LMS-level compliance data. This adds an inspection step to the new-content gate workflow but closes the SCORM blind spot.

Failure mode 5: The distributed production model creates an invisible backlog

When content producers outside the L&D team — sales managers, engineering leads, product managers, compliance officers — create training video and upload it to the LMS (or to SharePoint, Stream, or a shared drive that is functionally used for training), that content is invisible to the accessibility coordinator's new-content gate unless the gate extends to those production channels. The result is a growing backlog of non-compliant content that the accessibility coordinator does not know exists.

Diagnosis: Survey the organization's video storage locations beyond the primary LMS: Microsoft Stream, SharePoint video libraries, Vimeo channels, YouTube playlists, Google Drive folders shared with teams. For each location, assess whether content in that location is viewed by employees as training content. If yes, that content is in scope for caption compliance and is probably not being managed by the accessibility coordinator's workflow.

Mitigation: Extend the new-content gate to non-L&D production channels by working with IT and HR to establish publication-event triggers: LMS webhook → caption job submission, Stream upload → Zapier/Make connector → caption submission form. The goal is a trigger-based workflow that routes every new training video to the caption queue automatically, regardless of who produced it or where it was uploaded. The remote and hybrid workforce captioning guide covers the implementation of publication-event triggers for each major async video platform in detail.

Failure mode 6: LMS migrations reset caption compliance without warning

LMS migrations — moving from one platform to another, or upgrading to a new major version — are the single highest-risk event for caption compliance. Caption data is not video data: it is metadata associated with the video asset, and metadata migration is not guaranteed by standard LMS migration procedures. Organizations that migrate their LMS without explicitly planning caption data migration consistently discover that the post-migration library has lost caption tracks for a significant percentage of their content — sometimes 20–40% of the library.

Diagnosis: If the organization is planning or has recently completed an LMS migration, the accessibility coordinator should run a full caption track presence audit immediately after migration and compare the result against the pre-migration compliance baseline. A material reduction in caption track presence (more than 5%) indicates that caption data was not fully migrated and requires immediate remediation.

Mitigation: Implement the pre-migration caption checklist from the LMS migration caption checklist guide as a required pre-migration activity. The checklist covers the five caption data types that need to be migrated, the migration capability matrix for major source and destination LMS combinations, and the caption freeze window recommendation. The accessibility coordinator should be involved in every LMS migration planning process from the point of vendor selection, not brought in after the migration is complete.

Failure mode 7: The compliance dashboard is not updated between audits

A compliance dashboard that is updated quarterly reflects the compliance posture as of the last quarterly audit, not the current posture. For an organization that is adding 50 new videos per month and processing 40 remediations per month, a quarterly dashboard update means the compliance percentage displayed on the dashboard is potentially 3 months out of date. When the L&D director or a compliance function needs a current compliance number — for an accessibility audit response, a DOJ inquiry, or a board report — a 3-month-old dashboard number is not the answer.

Diagnosis: Check the last-updated date on each metric in the compliance dashboard. For metrics that are driven by the weekly batch workflow (new content gate compliance rate, QA pass rate), the last-updated date should be no more than 7 days old. For metrics that require a library-level calculation (library compliance percentage, remediation backlog count), the last-updated date should reflect the last audit run plus the ongoing additions and remediations since the last audit.

Mitigation: Implement incremental dashboard updates as part of the weekly batch workflow. When the accessibility coordinator runs the weekly QA review, they update the new-content gate compliance rate and QA pass rate metrics in the same session. The library compliance percentage is updated by adding new compliant clips and removing remediated clips from the running count, rather than recalculating from a full library audit. This incremental model keeps the dashboard current without requiring a full audit each week.

Failure mode 8: The role is dissolved after the compliance project completes

The most structural failure mode is the treatment of the accessibility coordinator role as a project-phase role rather than a permanent program function. In organizations where caption compliance was framed as a compliance project with a deadline (the ADA Title II enforcement date, the EAA compliance date, a specific OCR complaint deadline), the accessibility coordinator role may be staffed for the project phase and then dissolved or reassigned when the project deliverables are complete. The result is a compliance posture that is defensible at the project close date and degrades from that point forward.

Diagnosis: If the accessibility coordinator role is described in the organization's headcount as temporary, as a project allocation, or as ending on a specific date, the program is designed as a project, not as a program. The risk of the role being dissolved increases after the compliance deadline passes and the visible urgency decreases.

Mitigation: The accessibility coordinator role should be designed and staffed as a permanent program function from the start. The business case for the permanent function is that the compliance obligation is continuous: every week of new content production, every LMS migration, every vendor change, and every compliance framework update creates new compliance risk. The cost of maintaining a 20% allocation for the accessibility coordinator function is far lower than the cost of a DOJ complaint response or an OCR investigation triggered by a compliance gap that a functioning program would have caught. Frame the role to leadership as an insurance function: a small, predictable cost that prevents a large, unpredictable one.