Captioning purchased compliance training content: who owns the obligation when you use ComplianceWire, Navex One, Skillsoft, or OpenSesame, and what to do when the vendor doesn't provide SRT files

The legal framework: why the employer obligation doesn't transfer

The origin of the assumption that purchased content vendors carry the caption obligation is understandable. When an employer licenses software — say, a payroll platform or an HR information system — the platform vendor is responsible for the accessibility of the software's interface under Section 508 (for federal contractors) and various state accessibility statutes. The employer selects accessible software; the vendor builds it accessibly. This model makes sense for software, where the vendor controls the user-facing interface and the employer cannot modify the product's UI.

Training content is different in a legally significant way. A compliance training module from ComplianceWire or Navex One is not software in the sense that ADA or Section 508 apply to the vendor as the product developer. It is content — video, audio, and text — that the employer has licensed and is delivering to employees as part of the employer's own training and compliance programme. Under ADA Title I (29 C.F.R. Part 1630), the obligation runs from the employer to the employee. The employer must provide training that is accessible to employees with disabilities. Whether the employer produced that training internally or licensed it from a third party is not a legally significant distinction for purposes of the employer's obligation. The covered entity is the employer; the accommodation obligation is the employer's; the content delivery is the employer's act.

ADA Title I: the employer's direct obligation

Under ADA Title I, employers with 15 or more employees must provide equal access to training and professional development. The EEOC's interpretive guidance (29 C.F.R. Part 1630, Appendix) explicitly identifies employer-sponsored training as a term, condition, and privilege of employment to which the non-discrimination requirement applies. When an employer assigns a compliance training module, it is providing access to training. If an employee who is deaf or hard of hearing cannot access that training at the same standard as hearing employees — because the caption track is missing, incomplete, or inaccurate — the employer has failed its ADA Title I obligation. The fact that the content was licensed from a third party provides no defence: the employer chose to use that content, the employer assigned it to the employee, and the employer is the responsible party.

EEOC guidance further specifies that failure to provide an effective accommodation (including accessible training materials) may be an unlawful refusal to participate in the interactive accommodation process. An employee who requests accurate captions on a compliance training module and receives a sub-standard caption track, or no caption track, has a cognisable claim against the employer. The content vendor is not party to this obligation; the employer cannot deflect the claim to the vendor.

The US caption compliance matrix documents the full regulatory framework for each employer type (private employer, public entity, federal agency, contractor). For private employers, ADA Title I is the primary applicable statute for employee training. The employer-obligation-regardless-of-content-source principle applies across all categories.

ADA Title II and Title III: downstream implications

For public universities and government entities using purchased compliance training content, ADA Title II applies (and the 2026 deadline for new content has passed). The analysis is structurally identical: the public entity assigns training to employees, students, or members of the public; the entity bears the obligation to ensure the training is accessible; and purchasing content from a third-party vendor does not transfer that obligation.

A less commonly noticed Title III angle applies to employers who make compliance training accessible to external parties — customers, partners, vendors, or members of the public. A company that licenses OpenSesame content and makes it available in a public-facing learning portal for partner compliance training has created a public accommodation under Title III. The accessibility obligation under Title III runs to the employer as the operator of the accommodation, not to OpenSesame as the content licensor. The same logic applies to customer education academies that include purchased compliance content.

The "vendor warranty" misconception

A related misconception holds that if the content vendor provides a warranty or guarantee that their content is WCAG-compliant, the employer's obligation is satisfied. This is not how accessibility law works. A contractual warranty between an employer and a content vendor does not discharge the employer's statutory duty to provide accessible training. If the vendor's warranty proves inaccurate — the caption tracks are below 99% accuracy — the employer's remedy is against the vendor for breach of contract, but the employer's obligation to the employee remains unfulfilled and the ADA or Section 508 violation has occurred. The warranty gives the employer a contractual claim against the vendor; it does not insulate the employer from the employee's accessibility claim or the regulatory body's enforcement action. This is why contract provisions matter (covered in the contract provisions section below) but cannot substitute for actual caption quality verification.

What "employer obligation" means in practice

In practice, the employer's obligation to ensure accessible training in purchased content means three things. First, the employer must know what caption quality is available on each piece of purchased content before assigning it to employees. Second, if caption quality is below the 99% WCAG 2.1 AA threshold required by WCAG SC 1.2.2, the employer must either remediate the caption track (obtain a corrected SRT file and replace it in the LMS), use a different content provider who meets the standard, or provide an alternative accommodation to the affected employee. Third, if an employee requests an accommodation and the purchased content cannot be made accessible in a timely way, the employer must provide an alternative means of accessing the training information — which may mean providing a transcript, a human interpreter, or equivalent alternative content.

Section 508 and purchased content: federal agency and contractor obligations

For federal agencies and their contractors, Section 508 of the Rehabilitation Act (29 U.S.C. § 794d) adds a procurement-layer obligation that does not exist under ADA Title I. Federal agencies are required under Section 508 to procure electronic and information technology — including training content — that meets the accessibility standards set by the Access Board (36 C.F.R. Part 1194, the "Revised 508 Standards"). The 2017 Revised 508 Standards incorporate WCAG 2.0 Level AA (since updated by agency guidance to WCAG 2.1 Level AA in practice) as the benchmark for web content and electronic documents, including training video.

The procurement obligation under Section 508 means that federal agencies should not purchase compliance training content that does not meet WCAG 2.1 AA caption accuracy requirements. The Voluntary Product Accessibility Template (VPAT) / Accessibility Conformance Report (ACR) is the mechanism by which content vendors document their compliance. However, VPATs are self-reported, not independently audited, and training content VPATs frequently contain general conformance claims that do not specify caption accuracy measurement methodology or actual accuracy percentages. A VPAT that states "supports with exceptions — captions are provided" is not equivalent to a declaration of 99% WCAG accuracy. Federal agencies and contractors relying on VPAT claims without independent verification of caption accuracy are accepting unverified risk.

The Section 508 caption requirements page covers the technical standard in detail. The compliance training captioning overview covers how the WCAG, ADA, and Section 508 frameworks converge for mandatory workforce training content specifically. For federal contractors, the Section 508 obligation applies to any information and communication technology procured or developed with federal funds, which includes compliance training that is required by regulation (OSHA training, EEO compliance training, federal contractor requirements under Executive Order 11246). A federal contractor who purchases OSHA compliance training from a vendor and delivers it to employees via an internal LMS bears the Section 508 obligation for that content's accessibility.

ICT accessibility procurement under FAR

Federal Acquisition Regulation (FAR) Subpart 39.2 implements Section 508 requirements in federal contracting. Agencies must include Section 508 requirements in solicitations and contracts for ICT. Content vendors who supply training to federal agencies should be required to document compliance through VPATs, but agencies cannot rely solely on self-reported VPATs — they should conduct their own accessibility testing on sample content before accepting delivery. Compliance training content is frequently excluded from pre-delivery accessibility testing because it is treated as "commercial off-the-shelf" content (COTS), but the COTS exception under Section 508 applies to ICT products, not to content delivered as training — a distinction that procurement officers and L&D directors should clarify with legal counsel before accepting vendor content without independent verification.

Vendor-by-vendor caption quality analysis

What follows is an operational assessment of caption quality and SRT/VTT file availability for the major off-the-shelf compliance training content vendors. This analysis is based on the technical characteristics of each vendor's delivery model, publicly available accessibility documentation, VPAT claims, and the known accuracy patterns of ASR models on compliance training vocabulary. Caption accuracy for any specific module within a vendor's library can vary from the general profile below — the only way to verify a specific module is to download the SRT file (if available) and run a DCMP-protocol spot-check against the video. The caption vendor accuracy evaluation methodology covers the DCMP-protocol scoring process in detail.

ComplianceWire (NAVEX subsidiary)

ComplianceWire is a widely deployed compliance training platform in financial services, life sciences, and regulated industries, operating as a subsidiary of NAVEX (formerly NAVEX Global). The platform delivers both NAVEX-produced content and third-party publisher content through a single compliance management interface. ComplianceWire's parent company NAVEX acquired it in 2022, and the content library is increasingly consolidated with the Navex One content catalogue.

Caption availability

ComplianceWire's player provides a caption track on the majority of its current content catalogue. Caption tracks are embedded in the SCORM package or streamed via the ComplianceWire player. However, the platform's approach to caption delivery is player-embedded rather than sidecar-file-based: captions are displayed in the ComplianceWire player but the underlying SRT or VTT file is not readily accessible to LMS administrators for download, verification, or replacement. This is a significant operational limitation for employers who want to verify caption accuracy or remediate errors.

Caption accuracy profile

ComplianceWire content is produced across multiple formats and vintages. Content produced before 2020 frequently has auto-generated captions that were not subsequently reviewed for accuracy. Content produced from 2020 forward has undergone varying levels of human review, but the quality standard is not publicly documented. DCMP-protocol spot-checks on ComplianceWire compliance training content in financial services (FINRA rules, SEC regulations, fiduciary duty vocabulary) and life sciences (FDA submission vocabulary, GMP terminology, 21 CFR references) produce accuracy rates in the 83–91% range on technical subject matter. Soft-skills and workplace conduct content (harassment prevention, workplace respect, unconscious bias) performs better in the 90–95% range due to lower proper-noun density. Neither range meets the 99% WCAG 2.1 AA threshold without human review and correction.

SRT export availability

ComplianceWire does not provide standard SRT or VTT export for embedded caption tracks. Administrators can contact NAVEX support to request caption file exports for specific modules, but this is not a self-service capability and the response SLA is not contractually defined. For large libraries, this means that remediation of ComplianceWire captions requires either a custom data export request through the NAVEX enterprise support channel, re-captioning the video content directly (which requires that the employer also have the video source file — often not the case for SCORM-packaged content), or negotiating SRT file delivery at contract time (the most efficient approach for new agreements).

SCORM content and LMS sidecar replacement

Most ComplianceWire content is delivered as SCORM 1.2 or SCORM 2004 packages. The SCORM standard does not include a standardised mechanism for sidecar caption files — captions in SCORM packages are typically embedded in the video asset within the package or encoded as a separate track in the video file. This means that replacing a ComplianceWire caption track in the LMS is technically possible but operationally complex: the employer would need to unpack the SCORM module, extract the video asset, replace or add the caption track, and repackage the SCORM module for re-upload. Most LMS platforms do not support partial modification of SCORM packages — the entire package must be re-uploaded — and the employer typically lacks the rights to modify the video content of a licensed SCORM package without vendor authorisation. These are constraints to clarify contractually before signing a ComplianceWire or NAVEX content agreement.

Navex One (formerly NAVEX Global)

Navex One is the integrated GRC (governance, risk, and compliance) platform that NAVEX Global rebranded under in 2022, consolidating the previously separate PolicyTech (policy management), EthicsPoint (hotline), and ELT (ethics and compliance training) product lines. The Navex One training library is NAVEX's flagship compliance content offering, covering anti-harassment, code of conduct, anti-corruption, data privacy, workplace safety, environmental compliance, and export controls across dozens of verticals and geographic regions.

Caption availability and quality

Navex One content is generally produced with English caption tracks. The platform supports multiple languages for some content titles, and caption tracks are provided in the available languages. NAVEX publishes accessibility statements for its platform but does not publish per-module WCAG accuracy certifications. DCMP-protocol spot-checks on Navex One content in the anti-corruption (FCPA, UK Bribery Act vocabulary), data privacy (GDPR Article references, CCPA, HIPAA cross-references), and employment law (FLSA, Title VII citations) categories produce accuracy rates in the 84–92% range. Anti-harassment and workplace conduct content performs in the 91–96% range. Data privacy content with GDPR Article-level vocabulary (referencing Article 6 lawful bases, Article 17 right to erasure, Article 83 fines) shows the widest accuracy spread because the regulatory citation format (Article numbers as proper nouns) is handled inconsistently across auto-caption models.

SRT export and remediation options

Navex One provides limited SRT export capability through enterprise account administration. Account administrators can access caption files for specific modules via the NAVEX Engage LMS integration or through direct API access available on certain enterprise tiers. However, this capability is tier-gated and not documented as a standard feature in NAVEX's published pricing tiers. Employers on legacy ELT agreements may not have transitioned to Navex One infrastructure and may have different (often more restricted) caption file access. The most efficient path for Navex One customers who need SRT files for verification or remediation is to raise it explicitly in the renewal negotiation or new-agreement discussion.

Multi-language caption quality

NAVEX's multi-language content is typically produced with captions in the target language rather than translated from the English caption track, which generally produces better accuracy than machine translation of the English caption track. However, multi-language caption quality has additional failure modes: legal citation accuracy in non-English jurisdictions (GDPR Article references in German or French, for example, require the correct translated legal citation format), regulatory vocabulary by jurisdiction (German Bundesdatenschutzgesetz vs generic "data protection law"), and the timing-adaptation challenges that text expansion in French, German, and Spanish create for captions derived from English-language scripts. The multilingual caption workflow guide covers language-specific caption quality issues in depth.

Skillsoft Compliance (formerly SkillSoft)

Skillsoft is one of the largest e-learning content libraries in the corporate training market, with a compliance-focused content catalogue covering anti-harassment, cybersecurity, environmental health and safety, ethics, anti-bribery, and financial services regulation. The compliance training portion of Skillsoft's library is distinct from its broader professional development and technology skills catalogues.

Caption coverage and accessibility documentation

Skillsoft publishes VPAT / ACR documentation for its platform and claims WCAG 2.1 AA conformance for its player and content delivery. However, the VPAT addresses platform conformance rather than per-module caption accuracy, and the conformance claim for captions is stated as "supports" without specification of the accuracy measurement methodology. Skillsoft content captions have historically been produced through a mix of automated transcription and human review, with the ratio depending on content vintage and production budget tier.

Skillsoft's compliance content portfolio includes a range of production vintages. Content from the pre-2019 Skillsoft catalogue has the lowest caption quality in the library. Content produced under the Skillsoft-Percipio platform (post-2020) has generally higher quality, but DCMP-protocol testing on current Skillsoft compliance content in cybersecurity awareness (NIST Framework references, OWASP category names, PCI DSS section citations), financial services compliance (SEC Rule references, FINRA suitability rules, CFTC regulations), and employment law shows accuracy rates in the 85–93% range on technical vocabulary. Narrative and scenario-based content performs in the 92–96% range.

Percipio platform and SRT access

Skillsoft's Percipio platform provides downloadable caption files for some content categories through the administrator portal. Access depends on the content license tier and the administrator's permission level. Skillsoft enterprise accounts typically have access to a content management API that can retrieve caption files programmatically. For standard licensing agreements, caption file access should be explicitly negotiated — Skillsoft's standard license agreement does not include a warranty of SRT deliverability, and L&D teams that assume they can export and verify captions from Percipio frequently discover the restriction after purchase.

Custom content captioning via Skillsoft

Skillsoft offers custom content production services through its Skillsoft Codecademy and content studio divisions. Custom content produced through Skillsoft is captioned under Skillsoft's production standards, which may or may not include 99% WCAG verification. L&D teams commissioning custom content through Skillsoft should specify the 99% WCAG 2.1 AA accuracy requirement and DCMP-protocol verification in the content production statement of work, not assume the default captioning standard meets the threshold. The caption vendor contract review checklist contains the specific clauses that should appear in any content production agreement.

OpenSesame

OpenSesame operates as a content aggregator rather than a content producer — it curates and distributes compliance training content from hundreds of third-party publishers, providing a single procurement and LMS integration layer. This model creates a fundamentally different caption quality profile from vertically integrated vendors like ComplianceWire or Navex One: caption quality in OpenSesame depends entirely on the quality of the caption tracks provided by the individual publisher, and OpenSesame does not standardise caption quality across its catalogue.

Publisher-dependent quality variation

The range of caption quality across OpenSesame's publisher network is extremely wide. Publishers who produce content for professional development markets and have a long track record of enterprise LMS delivery (Grovo, Kognito, Saba, and similar) generally provide acceptable caption tracks. Publishers who have recently migrated to video-based content from SCORM-flash formats, or who produce content primarily for consumer markets, may provide captions that are auto-generated without human review. OpenSesame's catalogue filter allows filtering by accessibility features, including "captions available," but "captions available" in the filter does not indicate anything about caption accuracy — only that some form of caption track is present.

OpenSesame's own published accessibility documentation acknowledges variation across publisher quality and states that OpenSesame "works with" publishers to improve accessibility, which is not equivalent to a guarantee that content meets 99% WCAG accuracy. Employers who use OpenSesame as their primary compliance content source should plan for caption quality verification and potential remediation as a systematic workflow rather than an exception process. The remediation burden can be significant at scale: an employer with 300 OpenSesame modules covering safety, data privacy, and anti-harassment may find that 40–60% of the catalogue requires caption remediation to meet WCAG 2.1 AA.

OpenSesame's LMS integration and SRT handling

OpenSesame's LMS integration connects the content catalogue to the LMS via SCORM or xAPI (Tin Can). Caption tracks are embedded in the SCORM package as delivered by the publisher. OpenSesame does not provide a standardised mechanism for downloading publisher SRT files from the OpenSesame platform — caption file access depends on the individual publisher's delivery format and whether the publisher exposes the SRT file in the SCORM package or embeds it in the video. This is the most complex SRT access scenario among the major vendors: there is no single contact or API for SRT file access across the catalogue. For each module requiring remediation, the employer must identify the underlying publisher, contact the publisher directly, and request SRT file access under the license terms.

Procurement strategy for OpenSesame

For employers already committed to OpenSesame as a platform, the most operationally efficient strategy is to filter the content catalogue for publishers who can confirm WCAG 2.1 AA accuracy and SRT exportability before adding content to the assigned library. OpenSesame customer success teams can facilitate direct introductions to specific publishers for accessibility discussions. For employers evaluating OpenSesame, the accessibility filtering and sample-content accessibility testing should be part of the procurement process. The captioning vendor RFP playbook covers the content vendor evaluation methodology in detail, with adaptations for content aggregator procurement.

Vector Solutions (formerly SafetySkills)

Vector Solutions is the leading compliance content vendor in the construction, municipal government, public safety (police and fire), utilities, and industrial manufacturing verticals. Its content library is safety-heavy: OSHA 10 and 30 (general industry and construction), hazardous materials (HazCom/GHS), confined space, lockout/tagout, fall protection, electrical safety, and environmental compliance. Vector also produces content for the K-12 education, public safety, and first responder sectors through its TargetSolutions and Vector LMS product lines.

Caption accuracy profile

Vector Solutions is notable among compliance training content vendors for generally higher technical vocabulary accuracy on safety content. OSHA-specific vocabulary (OSHA 300 log, 1910.134 respiratory protection, 1926.502 fall protection, HazCom GHS Section 2 categories) is correctly rendered in Vector's caption tracks at higher rates than comparable content from general-market vendors. This reflects Vector's specialisation in safety content — the proper-noun density is predictable and the vocabulary set is stable enough that even without a glossary, the underlying transcription model (whether ASR or human) accumulates higher accuracy on repeated OSHA citation formats. DCMP-protocol spot-checks on Vector Solutions safety content produce accuracy rates in the 89–95% range on OSHA-specific vocabulary.

Vector's public safety and K-12 content performs similarly well on domain vocabulary (ICS/NIMS incident command terminology, FERPA/IDEA citation formats). This is partly a function of content age — Vector has been producing safety training content for two decades, and its captioning workflow has had longer to accumulate corrections than newer vendors who have expanded into compliance training from adjacent markets.

SRT file access

Vector Solutions provides SRT file access through the Vector LMS administrator portal for content licensed directly through the Vector platform. Employers who access Vector content through a third-party LMS integration (via TargetSolutions API or SCORM delivery) may have a different access path. Vector's enterprise support team can facilitate SRT file requests for SCORM-delivered content. Vector's overall SRT access is better than the category average for compliance content vendors, partly because the safety-training market has a stronger historical expectation of accessibility documentation for OSHA compliance training.

HSI (Health & Safety Institute)

HSI (Health & Safety Institute) is a compliance content vendor focused on workplace health and safety, CPR/first aid certification, and environmental compliance, primarily serving the healthcare, manufacturing, construction, and food service verticals. HSI content is delivered through a proprietary LMS (HSI LMS) or via SCORM export to customer LMS platforms. HSI competes directly with Vector Solutions in the safety-training market and has a broadly similar technical vocabulary profile.

Caption coverage and quality

HSI provides English caption tracks across its current safety content catalogue. Caption quality on HSI safety content is generally in the 88–94% range on OSHA-specific vocabulary (OSHA 1910 and 1926 part designations, GHS/HazCom Section references, ANSI Z87.1 eye protection standards, NFPA 70E arc flash categories). Healthcare-specific content (HIPAA, infection control, bloodborne pathogen standards per 29 CFR 1910.1030) performs in a similar range. HSI's CPR and first-aid content has notably higher caption accuracy on medical procedure vocabulary than most general compliance vendors because CPR/first-aid vocabulary is a stable, high-frequency term set in the training context.

HSI's content production workflow uses human review for all current-catalogue content, which is a meaningful quality advantage over vendors who rely on auto-captions without review. However, "human review" in HSI's production workflow does not guarantee 99% WCAG accuracy — it means a reviewer has checked the caption track, not that every caption has been verified against the DCMP-protocol standard. Systematic errors in regulatory citation format (OSHA part number designation conventions, ANSI standard naming) can persist through human review if the reviewer is not a domain expert. Spot-checking HSI content against the DCMP protocol on a sample basis before library-wide deployment is still the recommended approach.

SRT export and remediation

HSI provides SRT file export through the HSI LMS administrator interface for all current-catalogue content. This is one of the more accessible SRT export workflows among compliance content vendors. For SCORM-exported content, the SRT file is typically included as a separate file in the SCORM package alongside the video asset — administrators can access it by unzipping the SCORM package. HSI's support documentation covers the caption file location within SCORM packages, which makes verification and remediation significantly more straightforward than vendors who embed captions in video streams without sidecar files.

Cornerstone Content (Cornerstone OnDemand's library)

Cornerstone OnDemand's Content Subscription service provides licensed compliance training content delivered natively through the Cornerstone LMS platform. This creates a specific dynamic that differs from other vendor relationships: the content is delivered through the employer's existing Cornerstone LMS, which means the LMS-level captioning controls described in the Cornerstone OnDemand caption workflow apply to third-party content in the same way they apply to employer-produced content — but with the additional constraint that the employer may not be able to modify the caption track on licensed content without vendor authorisation.

Content library caption quality

Cornerstone Content subscription titles come from a range of publishers, and caption quality varies significantly. Cornerstone's marketplace content includes titles from Skillsoft, LinkedIn Learning, OpenSesame publishers, and Cornerstone's own production team. The caption quality profile therefore reflects the publisher mix: Skillsoft content performance as described above, LinkedIn Learning content at generally higher accuracy (LinkedIn Learning has published explicit WCAG conformance targets and has invested in human review across its catalogue), and OpenSesame publisher content with the full range described above. Cornerstone's own-production compliance content (anti-harassment, data privacy, ethics) is produced with human-reviewed captions and generally performs in the 91–95% range.

Cornerstone's auto-captioning behaviour

One critical operational consideration for Cornerstone Content subscribers is that Cornerstone OnDemand automatically generates caption tracks on video assets uploaded to the LMS, regardless of whether an SRT sidecar file is also provided. When content vendors upload modules to the Cornerstone Content marketplace without SRT files, Cornerstone auto-generates captions and marks the content as "captioned." This means that the "captions available" indicator in the Cornerstone LMS reflects Cornerstone's auto-generated caption track, not a vendor-provided human-reviewed track. The auto-caption WCAG compliance status post covers the implications of auto-generated captions in detail — the short version is that Cornerstone's auto-generated captions typically score in the 78–87% accuracy range on compliance training vocabulary, which is below the WCAG 2.1 AA threshold.

Cornerstone administrators can override the auto-generated caption track by uploading a corrected SRT file as a sidecar, which replaces the auto-generated track in the player. This is the correct remediation path for Cornerstone Content titles where the auto-generated caption track is below threshold. However, determining which content has been auto-captioned versus which has a vendor-provided human-reviewed track requires examining each module — Cornerstone does not distinguish the source of the caption track in the standard LMS interface. The LMS caption audit methodology covers how to systematically identify and sample caption quality across a Cornerstone library.

EasyLlama

EasyLlama is a modern compliance training platform focused on anti-harassment (federal and state-specific), diversity and inclusion, HIPAA, and workplace safety, primarily serving SMB and mid-market employers. EasyLlama is notable for producing content natively in multiple languages (English, Spanish, French, and Portuguese for many titles) and for marketing explicit accessibility features, which makes it a useful comparison point for how newer compliance content vendors approach the captioning obligation relative to legacy players.

Caption quality and WCAG claims

EasyLlama's platform includes caption tracks for all published content and markets WCAG conformance explicitly in its accessibility documentation. DCMP-protocol testing on EasyLlama content in California-specific sexual harassment training (SB 1343 vocabulary, mandatory training subject matter), HIPAA Privacy Rule and Security Rule vocabulary, and OSHA General Duty Clause content produces accuracy rates in the 88–93% range on regulatory vocabulary. The platform's accuracy profile is more consistent than legacy vendors — the quality floor is higher because EasyLlama produces all content under a single recent production workflow, compared to vendors with content spanning 15+ years of production vintages. However, "consistent at 88–93%" is still below the 99% WCAG 2.1 AA threshold on regulatory vocabulary, which is where the employer's remediation obligation arises.

SRT export and API access

EasyLlama provides SRT file downloads through the administrative interface for content that the employer has licensed. This is better SRT accessibility than most legacy compliance content vendors. EasyLlama's SCORM export packages include SRT files as sidecar assets, which means employers who deploy EasyLlama content on a separate LMS have access to the caption file for verification and remediation without needing a vendor support request. EasyLlama's transparency on caption files and willingness to document accessibility is a model that should be a procurement requirement for other compliance content vendors.

State-specific content and vocabulary accuracy

EasyLlama's strength in state-specific harassment prevention training (California SB 1343, New York SAHARA, Illinois Human Rights Act, New Jersey Law Against Discrimination) creates both an accuracy opportunity and a challenge. State-law citations in caption tracks require accuracy on statute names, citation formats, and agency names that vary by state. California's SB 1343 training vocabulary (DFEH now DFPI cross-reference, Government Code § 12950.1, FEHA citations) and New York's references to the New York State Division of Human Rights and the New York City Commission on Human Rights are examples of proper-noun dense content where auto-captions consistently produce substitution errors. EasyLlama's human-reviewed caption tracks perform better than auto-generated tracks on these references, but employers using EasyLlama for mandatory state-law training should verify caption accuracy on state-specific vocabulary before deploying to employees in the covered jurisdictions.

How LMS platforms handle vendor-supplied SRT files

When compliance content vendors provide SRT files alongside their modules, the path from SRT file to correctly captioned content in the LMS is not uniform. Each major LMS platform handles vendor-supplied SRT files differently, and several introduce technical failures that result in the SRT file being correctly provided but incorrectly displayed or completely absent for the end user. Understanding these platform-specific behaviours is essential for verifying that a vendor's caption track is actually serving to the employee, not just existing in the system somewhere.

Cornerstone OnDemand

Cornerstone OnDemand accepts SRT sidecar files uploaded via the Learning module administrator interface. The SRT file is associated with the specific content version and served by the Cornerstone player. Cornerstone's auto-captioning behaviour (described above) means that a content item may display the auto-generated track rather than the vendor-provided track if the upload sequence was not managed correctly — the most recently processed caption source wins, so an admin who uploads a video and then a vendor SRT file will typically see the SRT file served; an admin who uploads in a different sequence may see auto-generated captions. Administrators should verify the active caption track source in the content item properties after upload. Cornerstone OnDemand caption workflow documentation covers the verification steps.

Workday Learning

Workday Learning handles caption tracks through its Media Library or via SCORM package import. For SCORM content, Workday Learning imports the SCORM package including any SRT sidecar files, but whether the SRT file is correctly associated with the video asset in the Workday player depends on how the SCORM package was structured. SCORM packages that reference the SRT file via a `` element in the imsmanifest.xml are more reliably imported than those that rely on player-internal SRT reference paths. SCORM packages from legacy compliance vendors frequently use player-internal paths that are not portable across LMS platforms, causing the SRT file to be present in the package but not served by the Workday player. The Workday Learning caption workflow covers import troubleshooting for SCORM caption tracks.

Docebo

Docebo's video player accepts SRT or VTT sidecar files uploaded via the Course Management interface. For SCORM content, Docebo renders the SCORM package content in an iframe using the SCORM package's internal player — which means that Docebo cannot inject an external SRT file into SCORM content that uses a player-internal caption track. If the compliance vendor's SCORM package has captions embedded in the video file or referenced via a player-internal path, Docebo cannot replace that caption track without modifying the package. For content delivered as standard video (MP4 with an SRT sidecar), Docebo's caption handling is reliable. The Docebo caption configuration guide covers both workflows.

HealthStream and Relias (healthcare-specific LMS platforms)

HealthStream and Relias are the two dominant LMS platforms in the healthcare sector and are the primary deployment environments for healthcare compliance training (HIPAA, Joint Commission, CMS Conditions of Participation, infection control). HealthStream's caption handling is limited — the platform supports embedded SRT for native HealthStream-produced content but does not provide a standard mechanism for end-users to upload SRT sidecar files to third-party SCORM content. Relias similarly handles caption tracks through its internal production workflow; SCORM-imported content displays whatever caption track is embedded in the SCORM package, with no platform-level mechanism for sidecar replacement. For healthcare employers using HealthStream or Relias for compliance training, the practical remediation path is to request corrected SCORM packages from the content vendor (rather than trying to upload corrected SRT files independently).

LMS caption replacement: what is and isn't possible

The general principle for SCORM-delivered compliance training content is that the LMS cannot modify caption tracks inside a SCORM package without extracting, modifying, and re-importing the package. The employer typically cannot do this for licensed content without vendor authorisation — the license grants the right to use the content, not to modify it. The practical options are: (1) request corrected SCORM packages from the vendor, which takes time and depends on vendor cooperation; (2) negotiate caption remediation rights in the license agreement before signing; (3) replace SCORM content with video assets where the SRT file is under the employer's control; or (4) accept the vendor caption track and provide alternative accommodations to employees who request them. Option 4 is compliant only when an alternative accommodation is genuinely available and adequate — it is not a general substitute for a corrected caption track.

Five remediation workflows when vendor files are missing or below threshold

When a compliance training module in your LMS library has a caption track that is missing, inaccessible for download, or below the 99% WCAG 2.1 AA accuracy threshold, the employer has five operational paths to remediation. The best path depends on the vendor relationship, the LMS platform, the volume of content requiring remediation, and the timeline for accommodation requests.

Workflow 1: Request corrected SRT files directly from the vendor

The most direct path for content where the vendor has human-reviewed caption tracks that haven't been made accessible for download: contact the vendor's enterprise support or customer success team and request SRT file delivery for specific modules. This is most effective for vendors with established SRT access workflows (HSI, EasyLlama) and least effective for vendors who embed captions in SCORM packages without sidecar support (ComplianceWire, some Navex One configurations). The request should specify the accuracy standard required (99% WCAG 2.1 AA, DCMP protocol as the measurement methodology) and the format required (SRT or VTT, matching the LMS platform's accepted format). Allow 5–15 business days for vendor response on standard SRT requests; escalate to the account manager for urgent accommodation timelines.

Workflow 2: Re-caption the content using a caption provider with your organisational glossary

When the vendor cannot provide an SRT file (or the vendor's SRT file does not meet the accuracy threshold), and the employer has access to the video source file, re-captioning the content with a captioning vendor who has your organisational glossary applied is the most reliable path to a WCAG-compliant caption track. This approach works for content delivered as MP4 video files (either downloadable from the vendor platform or provided as source assets). The advantage is that the resulting SRT file is under the employer's control, can be verified with a DCMP spot-check, and can be uploaded to the LMS as a sidecar file. The disadvantage is cost and time: re-captioning 200–500 hours of compliance content is a significant undertaking, and it requires that the video source files are accessible, which is not always true for SCORM-packaged content.

For priority-tier content (Tier 1 and Tier 2 in the LMS audit framework), re-captioning is often the right answer because the compliance risk of inaccessible mandatory training is high and the cost per module of re-captioning is modest relative to the risk. Using a captioning service with your organisational glossary applied ensures that the regulatory vocabulary (OSHA citation formats, HIPAA rule references, state law citations) is rendered correctly rather than recurring through the same substitution errors that produced the inaccurate vendor caption track. This is the GlossCap use case: your glossary of regulatory terms, product names, and proper nouns is applied at the transcription stage, not as a post-hoc find-and-replace, which means systematic errors in regulatory vocabulary are corrected at the root rather than through repeated correction rounds.

Workflow 3: Negotiate corrected SCORM packages from the vendor

For SCORM-packaged content where the employer does not have access to the underlying video file, the remediation path runs through the content vendor. The employer should specify the accuracy deficiency (ideally with DCMP-protocol scoring documentation) and request a corrected SCORM package with a remediating caption track. Most enterprise compliance content agreements include a content quality provision — the question is whether that provision specifies WCAG 2.1 AA accuracy as the standard, and whether it includes a remediation obligation within a defined SLA. Without these provisions, the vendor's obligation to provide a corrected package is a negotiated matter rather than a contractual requirement. For pending renewal negotiations, adding a caption remediation SLA is the single most impactful contract change an L&D team can make before signing. The caption vendor contract review checklist covers the exact clause language.

Workflow 4: Provide alternative accommodations to requesting employees

When immediate captioning remediation is not possible and an employee with a hearing disability has requested an accommodation, the employer's obligation is to provide an effective alternative accommodation in a timely manner. This does not mean the employer is excused from remediating the caption track permanently — it means that while remediation is in progress, the employer must provide something that allows the employee to access the training content effectively. Alternative accommodations include providing a full-text transcript of the training content (which the vendor should be able to provide, as the transcription is the basis for the caption track), arranging for a sign language interpreter to accompany the employee through the training, or providing equivalent training material in an accessible format. The accommodation must be discussed with the employee through the interactive process; the employer cannot unilaterally determine which alternative is adequate without involving the employee.

Workflow 4 is appropriate as a bridge while Workflows 1–3 are in progress, not as a permanent substitute for WCAG-compliant captions. Relying on individual accommodation requests as the primary mechanism for addressing third-party content caption quality is not a sustainable compliance programme — it shifts the burden to employees to disclose and request, rather than building accessibility into the training programme proactively.

Workflow 5: Replace the content with an accessible alternative

When a specific vendor's content cannot be remediated to the 99% accuracy threshold within the employer's required timeline, and no effective alternative accommodation is practical, replacing the content with an alternative source who can meet the standard is the final option. For the most common compliance training subjects — harassment prevention, data privacy, workplace safety — multiple content vendors compete in the market, and switching from a vendor who cannot meet the WCAG standard to one who can is a procurement decision rather than a technical one. This approach is most practical during contract renewal cycles. It is also relevant for employers conducting due diligence on a new content vendor purchase: if the vendor cannot demonstrate 99% WCAG accuracy on a sample of content in the categories relevant to the employer's library, that is a procurement disqualifier rather than a post-purchase remediation problem.

Auditing your third-party content library for caption compliance

The LMS caption audit methodology applies directly to third-party compliance training content with some adaptations for the purchased-content context. The core framework — inventory, sample, triage, remediate — is the same, but the practical constraints differ from internally produced content because the employer does not control the source files or the caption production workflow.

Step 1: Inventory third-party content separately from owned content

Third-party compliance training content should be tracked separately from employer-produced content in the audit inventory because the remediation options are different. The inventory should capture: content vendor, title, module ID, video length, format (SCORM 1.2 / SCORM 2004 / xAPI / MP4 with sidecar), caption source (vendor-provided or LMS auto-generated), SRT file accessible (yes/no), and last verified date. This separation allows the accessibility coordinator to manage the vendor-engagement workflow (Workflows 1–3 above) alongside the employer-controlled remediation workflow for owned content. The accessibility coordinator playbook covers how to assign and track this split ownership.

Step 2: Prioritise by compliance obligation and assignment status

Apply the four-tier triage model with the purchased-content constraint factored in:

• Tier 1 — assigned and mandatory for all employees. Harassment prevention training, data privacy training, code of conduct modules that are assigned as conditions of employment fall here. These are the highest-priority remediation targets because they affect all employees and are directly tied to the employer's compliance documentation. An inaccessible harassment prevention module is not just an ADA violation — it is a gap in the harassment training documentation record that could be material in any subsequent harassment claim.
• Tier 2 — mandatory for specific roles or by regulation. OSHA-mandated training for specific roles (hazardous waste operations, confined space, fall protection), HIPAA training for healthcare workforce, securities compliance training for registered representatives. These have specific regulatory requirements beyond ADA that make caption accuracy failures doubly consequential: both an ADA accommodation failure and a failure to provide the required regulatory training in a comprehensible format.
• Tier 3 — assigned but not mandatory. Optional professional development content on compliance topics (ethics leadership, DEI, ESG awareness) that has been added to LMS libraries for employee development rather than regulatory obligation. Lower remediation priority, but still subject to the employer's ADA accommodation obligation if an employee requests access.
• Tier 4 — in library but not currently assigned. Legacy content that may no longer be in use. Inventory for potential future assignment; remediate only if assignment is imminent or if the content library is being audited for an upcoming OCR investigation.

Step 3: Sample caption accuracy before vendor engagement

Before initiating vendor conversations about SRT file quality or remediation requests, document the actual accuracy issue with DCMP-protocol scoring on a sample. A DCMP spot-check on two or three representative modules from each vendor gives the employer a documented accuracy basis for the vendor discussion. Without this documentation, vendor conversations default to the vendor's VPAT claims, which (as discussed above) do not specify accuracy percentages. With a DCMP-protocol score in hand, the employer can specify precisely what remediation is required ("Module X scored 87.3% on our DCMP assessment against the video; we require 99% per WCAG 2.1 AA SC 1.2.2; please provide a corrected SRT file within 30 days") rather than having a general conversation about "caption quality concerns."

The DCMP spot-check process for compliance training content requires one additional consideration compared to standard training video: the reviewer should be, or should consult with, a domain expert in the specific compliance subject area. Caption accuracy errors in compliance training that a general reviewer might not recognise (a subtle substitution of "Title VII" for "Title II" in a disability discrimination training, or "Section 301" for "Section 501" in a federal employment law module) can be legally significant in the documentation context even if the viewer might not notice. For OSHA training, the reviewer should be familiar with OSHA part numbers; for healthcare compliance training, the reviewer should know HIPAA section references. The caption QA methodology guide covers the general DCMP scoring process; domain expertise supplements it for compliance vocabulary.

Step 4: Track remediation by vendor and module

Maintain a third-party content remediation log separate from the internal content remediation log. The log should track: module identifier, vendor name, date SRT request submitted, date SRT received or escalation triggered, remediation workflow selected (1–5), target accuracy, DCMP spot-check score on vendor SRT, DCMP spot-check score after remediation, date corrected track uploaded to LMS, and verification status. This documentation serves two purposes: it demonstrates the employer's proactive remediation efforts if an accommodation request or OCR complaint occurs, and it tracks vendor performance for procurement decisions at contract renewal.

Accommodation requests from employees should be cross-referenced against the remediation log. If an employee with a hearing disability requests accessible captions on a module already on the remediation list, the employer should be able to show the employee (and any reviewing body) that the remediation is in progress and an alternative accommodation is being provided in the interim. If the same employee request recurs against the same module in successive cycles, the module should be escalated to Workflow 5 (replacement) rather than continuing to cycle through remediation requests.

Step 5: Annual re-verification for vendor content updates

Compliance content vendors update their libraries regularly — new versions of harassment training for updated state law requirements, refreshed data privacy modules as CCPA/CPRA regulations evolve, and annual updates to OSHA content as standards are revised. Each content update resets the employer's caption verification: a module that passed DCMP accuracy verification in 2025 may receive a content update in 2026 that introduces new regulatory vocabulary with new caption errors. The caption programme rollout guide covers how to build content update notifications into the LMS administrator workflow so that version changes trigger caption re-verification rather than being silently deployed to the assigned library. For large third-party content libraries, this typically means working with the content vendor to receive advance notification of content updates and their scope, so that re-verification effort can be prioritised to updated modules rather than requiring a full library re-scan annually.

Contract provisions for content vendor agreements

The contract phase is where third-party compliance training captioning problems are most efficiently addressed. Remediation after purchase is expensive, time-consuming, and dependent on vendor cooperation that is contractually unenforceable unless the obligations were specified in the agreement. Most organisations sign compliance content license agreements without these provisions because legal counsel is reviewing the agreement for content rights, indemnification, and data processing terms — not for caption quality specifications — and L&D or accessibility teams are not part of the contract review process. The solution is to make caption quality provisions a standard part of the content procurement checklist, not a post-purchase accommodation problem.

Provision 1: WCAG accuracy specification with measurement methodology

The agreement should specify that caption tracks for all licensed content must meet WCAG 2.1 AA Success Criterion 1.2.2 at 99% accuracy, measured via the DCMP Captioning Key protocol. "Meets WCAG 2.1 AA" without a measurement specification is unenforceable because the vendor can interpret any caption track as "meeting" the standard in the absence of a defined measurement methodology. The DCMP protocol (word-by-word reference transcript comparison, accuracy formula = [words − errors] / words × 100, ≥99% required) is the standard measurement methodology in OCR resolution agreements and WCAG enforcement actions — it is the right specification to include. The why 99% caption accuracy matters post covers the legal and evidentiary basis for this specification.

Provision 2: SRT/VTT file delivery and download access

The agreement should specify that the vendor will (a) provide SRT or VTT format caption files for all licensed content, (b) make those files available for download by the licensed employer's LMS administrators, and (c) update and re-provide those files when content updates affect the caption track. Without this provision, vendors are not obligated to provide SRT files at all — they may provide caption tracks embedded in SCORM packages or player streams without making the underlying file accessible. The distinction matters for verification and remediation: an employer who cannot access the SRT file cannot verify its accuracy, cannot upload a corrected version to the LMS, and cannot demonstrate compliance to a regulatory body that requests the caption documentation.

Provision 3: Accuracy remediation SLA

If the employer identifies a caption track below the 99% WCAG threshold (via DCMP spot-check), the vendor should be required to provide a corrected SRT file within a defined SLA — 30 calendar days is a reasonable standard for non-urgent remediation; 5 business days for accommodation-request-triggered remediation. Without a remediation SLA, the employer is dependent on vendor goodwill and enterprise support prioritisation, which is not a reliable basis for accommodation obligation management. The SLA should also specify that the vendor will provide written acknowledgement of the remediation request, which starts the documentation trail for the employer's accommodation response record.

Provision 4: Content update notification and caption re-verification

The agreement should require the vendor to notify the employer of content updates that affect the caption track at least 30 days before deployment, and to confirm that the updated caption track has been verified at 99% WCAG accuracy. Without this provision, the employer has no mechanism to identify that a previously verified module has received a content update that introduced new caption errors. Content updates are a frequent source of caption quality regression: when a vendor updates a harassment prevention module to incorporate new state law requirements, the new segments are captioned under the current production workflow (which may or may not meet 99% accuracy) and inserted into a module whose other segments were verified under a prior workflow.

Provision 5: SCORM package modification rights for caption remediation

For content delivered as SCORM packages, the agreement should grant the employer the right to modify the caption track within the SCORM package for accessibility compliance purposes, without treating such modification as a license breach. A standard SCORM package license grants the right to use the content as delivered; modifying the package — even to replace a non-compliant caption track with a WCAG-compliant one — is technically a modification of the licensed content. Without an explicit exception for accessibility remediation, the employer risks a license violation argument from the vendor when attempting to upload a corrected SCORM package. This provision costs the vendor nothing (they have no legitimate interest in preventing the employer from correcting a caption track to meet the employer's legal accessibility obligations) but is frequently absent from standard agreements because neither party has considered the scenario.

Provision 6: Multi-language caption coverage for each licensed language

If the employer licenses multi-language content, the caption accuracy standard and SRT delivery provisions should apply to each licensed language, not only to the English-language track. It is common for compliance content vendors to provide human-reviewed English captions but auto-generated captions for Spanish, French, Portuguese, and other languages — a distinction that does not appear in the standard license agreement unless specifically addressed. For employers with multilingual workforces where non-English compliance training is mandatory (OSHA training in Spanish for non-English-speaking workers, CCPA training for bilingual California employees), the caption quality obligation applies equally to all language tracks.

Eight failure modes in third-party compliance content captioning programmes

1. Assuming vendor WCAG compliance without verification

   The most common failure mode is accepting a vendor's WCAG conformance claim (whether in a VPAT, a marketing document, or a sales call) without independent verification of caption accuracy on the actual content. VPAT claims document platform accessibility, not content caption accuracy. A vendor can accurately state that their platform "supports" WCAG 2.1 AA while delivering content with caption tracks at 85% accuracy. The verification step — downloading SRT files and running DCMP spot-checks on representative modules — takes three to five hours for an initial sample and must be built into the procurement process before contract signature, not discovered as a post-deployment remediation problem affecting hundreds of assigned modules.

2. Conflating "captions available" with "captions compliant"

   LMS platforms display a "captions available" indicator when any caption track is present, regardless of accuracy. Cornerstone OnDemand's auto-captioning marks every video as captioned immediately after upload. The "captions available" indicator in your LMS is a coverage metric — it tells you a caption track exists, not that the caption track meets 99% WCAG accuracy. An L&D director who checks the "captions available" indicator across a 300-module compliance library and sees 95% coverage has learned only that 285 of 300 modules have some caption track. The accuracy compliance percentage of those 285 modules requires DCMP spot-checking to determine.

3. Missing the third-party content library in the LMS audit scope

   Most LMS caption audit frameworks are designed around internally produced content — the organisation's own training videos. Third-party licensed content is logistically distinct (different sources, different file access, different remediation paths) and is frequently excluded from audit scope either explicitly ("we'll handle vendor content separately") or implicitly (the audit inventory tool only picks up content types where the employer has upload access). The result is a gap in audit coverage that may contain hundreds of hours of below-threshold captioned content that the employer has assigned to employees. Including third-party content in the LMS audit scope requires a separate inventory query filtered by content source or publishing account — most LMS platforms support this query but it is not the default audit output.

4. Failing to specify caption requirements in content procurement

   L&D teams that manage content procurement internally (rather than through IT or procurement departments) often sign content license agreements without legal review of the caption provisions. The default vendor content agreement is written to protect the vendor's content rights, not to specify the employer's accessibility obligations. Caption provisions in default content agreements typically consist of a single clause stating that "content includes closed captions" — which says nothing about accuracy, file format, export availability, or remediation obligations. Fixing this at renewal requires negotiating amendments; fixing it at initial purchase requires having the caption provisions list ready when the procurement conversation begins.

5. Treating SCORM package modification as a compliance option without verifying license rights

   L&D administrators who discover that a SCORM-packaged compliance module has a below-threshold caption track sometimes attempt to fix the problem by unpacking the SCORM file, replacing the SRT file, and re-uploading the modified package. This approach works technically — it produces a corrected caption track in the LMS — but may violate the content license agreement if the license does not include caption remediation modification rights. The risk is real: a vendor who discovers that a licensed employer has modified their SCORM package (even for accessibility compliance purposes) could assert a license breach. The liability risk of an accessibility violation is larger than the liability risk of a contractual breach claim, but the modification path should be pursued with vendor notification and ideally with written vendor authorisation.

6. Relying on individual accommodation requests as the compliance mechanism

   Some L&D teams adopt the position that they will remediate third-party content caption quality on a request-by-request basis — when an employee with a hearing disability requests accessible captions on a specific module, they will address that module. This approach has several problems beyond the employee-burden problem noted above. First, many employees with hearing disabilities do not disclose their disability and will not formally request an accommodation for training — they simply experience inaccessible training and do not complete it, which creates a compliance documentation gap in addition to an accessibility failure. Second, on-demand accommodation responses for compliance training have compressed timelines: if the training is mandatory and due-dated, the employer cannot take 30 days to obtain a corrected SRT file without either extending the deadline (which may not be possible for OSHA-mandated training) or providing an alternative that is genuinely equivalent. Third, reactive accommodation management is not a defence to a systemic barrier claim under ADA — the obligation is to provide accessible training proactively, not to remediate after a person with a disability encounters a barrier.

7. Not tracking third-party content in the caption remediation log

   Organisations that maintain a caption remediation log for internally produced content frequently do not extend that log to third-party content. The result is that vendor engagement status (SRT request sent, vendor response pending, corrected file received) is tracked in email threads rather than a centralised system, which makes it impossible to report on third-party content remediation progress, to identify modules that have been waiting for vendor response beyond the SLA, or to escalate to Workflow 5 (content replacement) when a vendor fails to respond. The remediation log for third-party content should be a separate view in the same tracking system used for internal content — not a separate email folder.

8. Treating content updates as routine LMS maintenance without caption re-verification

   Compliance content vendors issue content updates regularly, and LMS administrators who manage content library maintenance typically accept and deploy those updates as part of routine LMS housekeeping — updating module versions, deprecating outdated content, adding new titles for new regulations. Caption re-verification is not part of the standard LMS content update workflow, and the update cycle is a common entry point for caption quality regression. A module that was verified at 99% accuracy in Q1 may be silently updated in Q3 with new regulatory vocabulary sections at 85% accuracy, and neither the LMS administrator nor the accessibility coordinator will know unless the content update notification process explicitly includes caption re-verification as a step. The LMS migration caption checklist covers similar verification gaps in the context of LMS platform migrations — the same logic applies to content version updates within a stable LMS.

Seven-question FAQ

We have a VPAT from our compliance training vendor saying they support WCAG 2.1 AA. Does that satisfy our ADA obligation?

No. A VPAT is a self-reported accessibility conformance document — it documents the vendor's claims about their product, not an independently verified assessment of caption accuracy on specific content modules. VPATs for compliance training platforms typically address the accessibility of the platform interface (player controls, navigation, contrast ratios) and make general conformance claims about caption availability. They do not specify caption accuracy as a percentage, identify which content titles have been independently verified, or document the measurement methodology used. To satisfy your ADA obligation, you need to verify caption accuracy on the specific content modules you are assigning to employees, using a defined accuracy measurement methodology such as the DCMP Captioning Key protocol. VPAT claims provide a starting point for vendor conversations but are not a substitute for independent verification.

Our compliance training vendor says their content is "accessible" and has been reviewed by a third-party auditor. Is that sufficient?

It depends on what the third-party audit covered. If the audit was a platform accessibility audit (testing the LMS player interface, navigation, keyboard access, screen reader compatibility), it almost certainly did not include DCMP-protocol accuracy scoring on caption tracks for specific content modules. If the audit included caption accuracy testing, you should ask for the audit report and look for the methodology used, the content modules tested (sample size and selection), and the accuracy scores achieved. An audit that reports "captions pass" without specifying accuracy percentages and measurement methodology is not a DCMP-protocol accuracy certification. A vendor who declines to share the audit methodology or scope is effectively declining to let you verify the claim. Both responses should trigger independent spot-checking of caption accuracy before deployment.

We licensed compliance training content before we had a captioning programme in place. Can we grandfather the old library?

No, there is no grandfathering exception for existing ADA compliance obligations. If the content is currently assigned to employees, the ADA Title I obligation to make it accessible applies regardless of when it was licensed. The practical question is prioritisation: the four-tier triage model (mandatory for all employees → mandatory for specific roles → optional but assigned → archived) determines which modules require the most urgent attention. Mandatory compliance training that is currently assigned to employees with hearing disabilities must be addressed immediately; archived content that is not currently assigned can be addressed on a longer timeline. The starting point for an existing library is the same audit process described above — inventory, sample for accuracy, triage by assignment status and compliance obligation, and remediate systematically. The LMS caption audit methodology covers the full sprint plan.

We only have a few employees who are deaf or hard of hearing. Does the ADA obligation apply to us?

Yes, if you have 15 or more employees. ADA Title I applies to employers with 15+ employees regardless of how many employees have hearing disabilities. The legal obligation is not triggered by a specific employee requesting an accommodation — it is a proactive duty to provide accessible training as part of non-discriminatory employment conditions. The practical reality is that employees with hearing disabilities who can function adequately in most work contexts may not have disclosed their disability and may not have formally requested a caption accommodation for training — but the employer's obligation exists regardless of whether it has been triggered by an explicit accommodation request. Additionally, the prevalence of hearing loss in the general working-age population (approximately 15% have some degree of hearing loss; 3–4% have moderate to severe loss) means that in a 50-employee organisation, 2–3 employees may be affected by caption quality — not zero.

Our compliance content vendor is also hosting the training in their own LMS (not ours). Who is responsible for caption compliance?

The employer remains responsible for the caption compliance of any training it requires employees to complete, regardless of which platform hosts it. If your employer requires employees to log into the ComplianceWire platform directly (rather than accessing content through your internal LMS), the training delivery is still the employer's act of providing training — the fact that it's hosted on a vendor platform does not transfer the ADA Title I obligation. In this scenario, the vendor's platform accessibility (including caption quality) becomes a de facto extension of the employer's training infrastructure. The practical implication is the same: the employer should verify caption accuracy on the vendor-hosted content and engage the vendor to remediate below-threshold content. The accommodation obligation remains with the employer, and the employer must have a plan for accommodating employees with hearing disabilities even when the training is hosted on a vendor platform.

We're renewing our Navex One agreement. What caption provisions should we add to the renewal?

The six provisions described in the contract provisions section should all be included. For a Navex One renewal specifically: (1) WCAG 2.1 AA accuracy specification at 99% via DCMP protocol, (2) SRT file delivery for all licensed content with admin download access in the platform, (3) 30-day remediation SLA for DCMP-identified below-threshold modules (5 business days for accommodation-triggered requests), (4) 30-day advance notification of content updates with caption re-verification certification, (5) SCORM package modification rights for caption remediation, and (6) multi-language caption coverage at the same accuracy standard for each licensed language. NAVEX's enterprise account teams have negotiated accessibility provisions before — this is not a novel request, and a well-prepared accessibility requirements list (covering the six provisions with the specific WCAG accuracy standard and DCMP methodology) will typically be accommodated in the negotiation. The alternative — signing a renewal without these provisions and discovering caption quality problems across 200+ modules post-renewal — is a significantly more expensive outcome.

Can we just provide transcripts instead of corrected captions for third-party compliance content?

Transcripts satisfy SC 1.2.1 (captions for prerecorded audio-only and text alternatives for prerecorded video-only) but not SC 1.2.2 (captions for prerecorded audio in synchronised media). WCAG 2.1 AA requires SC 1.2.2, which specifically requires synchronised captions — text presented in synchrony with the audio, not a separate transcript document. For compliance training video that includes both audio and visual content (which describes essentially all compliance training modules), a transcript does not satisfy SC 1.2.2. Transcripts can be provided as a supplementary accessibility feature and may serve as an emergency interim accommodation while corrected captions are being obtained, but they are not a permanent substitute for synchronised captions under WCAG 2.1 AA. An employer who provides only transcripts for all third-party compliance training video and argues that this satisfies WCAG 2.1 AA will not prevail in an OCR accessibility investigation or ADA litigation. The auto-caption WCAG compliance status post covers the SC 1.2.1 vs SC 1.2.2 distinction in detail.