Platform reference · Cornerstone OnDemand

Cornerstone OnDemand captions: enterprise LMS captioning that survives audit

Cornerstone OnDemand (now Cornerstone, post-Saba acquisition and the rebrand) is the enterprise LMS that underpins L&D at the bulk of the Fortune 1000 — manufacturing, financial services, federal contractors, large healthcare systems, multinational SaaS, the Big-4 consultancies. Where TalentLMS, Docebo, and Absorb live in the SMB-and-mid-market range, Cornerstone is what L&D operations leads at 5,000-to-100,000-employee organisations are running. The captioning question on Cornerstone is not whether the platform supports captions — it does, with a content-asset-level caption-track-upload model, SCORM 1.2 / SCORM 2004 / xAPI / AICC publish-target compatibility, and multi-language support — but whether the captions survive the audit lens that comes with enterprise scale: OFCCP for federal contractors, the Section 508 procurement-evidence review, the European Accessibility Act mandatory-training inspection, the OCR HIPAA workforce-training file review for healthcare tenants. At enterprise scale, the captioning provenance is part of the audit evidence, and the upstream glossary-biased workflow is what makes the provenance log defensible.

TL;DR

Cornerstone supports caption-track upload at the content-asset level via SRT and WebVTT for direct video assets in the Content Library, plus full SCORM 1.2 / SCORM 2004 / xAPI / AICC ingestion for content authored in Articulate Storyline, Articulate Rise 360, Adobe Captivate, Lectora, and iSpring with embedded captions in the published package. Multi-language caption tracks are supported on a single video asset via the language-tagged caption-track upload pattern. The technical caption upload is straightforward; the difficulty is upstream — at enterprise scale, the caption-vocabulary surface is dense (drug names, SDK terms, regulatory citations, product names, customer-confidential identifiers, internal acronym registers, multi-jurisdictional regulatory references) and ASR-generated captions mangle systematically. Glossary-biased captioning with the enterprise's controlled vocabulary as the project glossary is what produces caption files clean enough to satisfy the audit lens. The provenance log per asset is the audit-evidence shape.

What Cornerstone is, and where in the workflow captioning lands

Cornerstone OnDemand is a multi-product enterprise talent-management platform; the Learning module is what L&D operations leads care about in the captioning conversation. Distinguishing characteristics:

Content Library asset model. Training content is organised as Learning Objects (LOs): online courses, video assets, materials, tests, evaluations, sessions. Video assets and online courses are the captioning surface.
Multi-tenant deployment with per-OU configuration. Cornerstone deployments are organisationally scoped — Organisational Units (OUs) for division, country, business unit. Caption configuration can vary per OU; multi-region deployments need per-OU caption-language defaults.
Native authoring + third-party ingestion. Cornerstone Content Studio (the native authoring tool, post-acquisition rebrands) handles in-platform authoring; the dominant authoring path remains third-party (Articulate, Adobe Captivate, iSpring, Lectora) with SCORM/xAPI ingestion. The captioning surface differs by path.
Cornerstone Content Anytime / xPLOR / Skillsoft partnership content. Pre-built content from Cornerstone's content partnerships (or the Cornerstone Content Anytime catalogue) generally ships with captions; the captioning workflow concerns are the customer-authored content that wraps or supplements the catalogue.
Compliance and reporting layer. Cornerstone's reporting layer (Reporting 2.0, custom reports, the Compliance Toolkit) supports per-asset metadata that lets the captioning provenance log live in the LMS rather than in a separate spreadsheet.

Captioning lands at three points: (1) direct video upload into the Content Library, with a sidecar caption file uploaded alongside; (2) SCORM/xAPI/AICC content packaging where captions are embedded in the publish artefact; (3) external-hosted video referenced from a Content Library entry, where captions live with the host (Vimeo, Wistia, Kaltura, Panopto).

The Cornerstone caption-upload mechanic

Direct video asset + caption sidecar. The Content Library's Video / Video LO type accepts MP4 video plus a sidecar caption file in SRT or WebVTT. The Manage Captions interface on the asset's edit page exposes upload, replace, and language-tag operations on each caption track. Multiple language tracks can attach to a single video asset.
SCORM / xAPI / AICC online course. The dominant ingestion path for full courses. Authored in Articulate Storyline, Rise 360, Adobe Captivate, Lectora, or iSpring; published with embedded captions; uploaded as a SCORM 1.2 / SCORM 2004 / xAPI / AICC zip. The captions ride inside the package; Cornerstone's player respects them.
Content Studio in-platform authoring. The native authoring path supports caption-track configuration on slide-level video. The captioning workflow upstream is the same as for third-party authoring tools — the difference is the publish target is internal rather than SCORM/xAPI ingestion.
External hosted video. Where the Content Library entry references a video hosted on Kaltura, Panopto, Vimeo, or Wistia, the captioning lives at the host. See our Kaltura, Panopto, Vimeo, and Wistia references.
Live training session recordings. Cornerstone Live Sessions (or recorded sessions ingested from Zoom, MS Teams, WebEx) are video assets at rest after the session ends; the caption-track-upload mechanic on the recorded asset is the same as for any other video asset. ASR-generated session transcripts from the conferencing tool typically need a glossary-biased re-pass before promotion to the Content Library.

The vocabulary surface at enterprise scale

Cornerstone tenants concentrate every vocabulary surface we measure. Common patterns at the 5,000-to-100,000-employee scale:

Multi-jurisdictional regulatory citations. US (HIPAA, SOX, OSHA, EEOC, ADA, FDA, FTC, FCPA), EU (GDPR, EAA, DORA, NIS2, EU AI Act), UK (UK GDPR, Equality Act 2010, EHRC), Canada (CASL, AODA, PIPEDA), APAC (PDPA, Privacy Act 1988, PIPL). Citation density per minute of compliance training is the highest of any vertical. See compliance training captions.
Product, programme, and SKU names. Multinational organisations carry hundreds of internal product names per division, with the additional complexity of brand-name overlap across geographies. Generic ASR mangles these uniformly.
SDK and engineering vocabulary. For engineering-heavy organisations (manufacturing, automotive, aerospace, software), the engineering onboarding catalogue carries dense SDK / standard / specification vocabulary. See engineering onboarding captions.
Healthcare-specific surface. Drug names, procedure codes, EHR-specific terminology, accreditation acronyms — every healthcare Cornerstone tenant carries this. See medical training captions and HIPAA training captions.
Safety-training surface. OSHA / MSHA / EU Council Directive 89/391/EEC framework directives, chemical names from SDS, equipment and PPE proper nouns. See safety training captions.
Customer / partner / vendor identifiers. For sales-enablement and customer-success training, customer names appear in the audio. The privacy posture is the same as for Loom-recorded async content — a controlled-access glossary tier with documented data flow.
Internal acronym register. Every enterprise has an OKR / KPI / programme / division / system acronym register. Generic ASR mangles every entry.
Multilingual content. Cornerstone tenants in multinational deployments need language-tagged caption tracks on the same video asset; the glossary-biased workflow runs once per language, with a per-language reviewer pass.

The audit-evidence shape at enterprise scale

Cornerstone tenants face a denser audit calendar than mid-market LMS tenants:

OFCCP audits for federal contractors. The Office of Federal Contract Compliance Programs reviews affirmative-action plans and accessibility evidence; training accessibility is part of the package. Section 508 flow-down through procurement contracts is the relevant technical bar.
OCR HIPAA workforce-training file review. For healthcare tenants, the Office for Civil Rights resolution-agreement pattern includes workforce-training-documentation findings. The captioning provenance per training asset is part of the documentation packet.
Joint Commission triennial survey for healthcare tenants. The HR file review and tracer methodology reach training records; the captions surface during the surveyor's view of the training. See Healthstream captions for the Joint Commission lens (Cornerstone is the alternative LMS for hospital systems not on Healthstream).
EU mandatory-training inspection. Member-state regulators inspect mandatory-training programmes (occupational-safety, GDPR, AML, financial-services compliance). The EU regime adds the EAA / EN 301 549 accessibility lens on top.
SOX 404 ICFR control-evidence review. Internal-control-over-financial-reporting walkthroughs include management-asserted training controls — for many SOX-relevant controls, training completion plus training-content adequacy is the evidence.
SOC 2 / ISO 27001 surveillance audits. Continuous-monitoring audits reach security-awareness training. Captions on security-awareness training that mangle the relevant vocabulary (data-classification names, system names, role names) are an audit gap.

The captioning provenance log per asset — caption source (vendor + glossary version), reviewer name and role, review date, glossary term count — is the audit-evidence shape. Cornerstone's per-asset metadata fields permit this log to live in the LMS rather than in a parallel spreadsheet, which is the operational reason to land the caption-provenance metadata at the LMS layer.

The glossary-biased workflow at enterprise scale

Pull the enterprise's controlled vocabulary across business units. The federation problem at enterprise scale is real: a multinational SaaS has different feature catalogues per division; a multinational healthcare system has different drug formularies per country; a multinational manufacturer has different equipment catalogues per region. The glossary-biased workflow needs OU-aware vocabulary management — the right glossary for the right asset.
Establish per-OU caption-language defaults. Cornerstone's OU-scoped configuration handles this; the captioning workflow needs a corresponding per-OU language and glossary mapping.
Process the back-catalogue first. At enterprise scale, the back-catalogue is typically thousands of training assets. The retrofit pattern: enumerate via the Cornerstone Reporting layer, prioritise by training-frequency-times-audit-risk, batch through the captioning workflow with the correct OU-scoped glossary, replace caption tracks via the asset-management API.
SME / clinical / engineering reviewer pass. The reviewer step is non-optional at enterprise audit-relevance. The amber-highlight UI shows every glossary-applied term in context with source-line provenance. The reviewer is OU-scoped — the engineering reviewer for engineering content, the clinical reviewer for clinical content, the legal-compliance reviewer for compliance content.
Upload caption tracks via the asset-management API. For sub-thousand-asset deployments, the Cornerstone admin UI is fine. For larger retrofits, the asset-management API path is the operational answer; the SCORM/xAPI re-ingestion path handles authored content.
Document captioning provenance per asset in Cornerstone's per-asset metadata. Custom asset-fields capture caption source, glossary version, reviewer, review date, glossary term count. The Reporting layer can produce the audit-evidence packet on demand.

See pricing

Cornerstone-specific notes for the captioning RFP

Enterprise procurement teams running a captioning RFP for a Cornerstone-hosted training catalogue will expect the responses to address several Cornerstone-specific surfaces. From our captioning RFP template, the questions that have a Cornerstone-specific shading:

Bulk caption upload via Cornerstone's asset-management API. Vendors should be able to integrate at the API rather than the UI level for back-catalogue retrofits at scale.
SCORM / xAPI re-publish path. For authored content (Storyline, Rise, Captivate, Lectora) where captions are embedded in the publish artefact, the vendor's role is upstream — clean SRT / VTT delivery into the authoring source — rather than direct upload to Cornerstone. Vendor's RFP response should walk both paths.
OU-scoped glossary management. The federated vocabulary model needs to be supported. Vendors that treat the customer as a single organisation will not handle multinational deployments well.
Multi-language support across the caption-track-per-language pattern. Single-language vendors are not a fit for multinational Cornerstone deployments.
Reporting integration. The captioning provenance log should land in Cornerstone's per-asset metadata via the asset-management API, not in a parallel spreadsheet.
Data-handling posture for customer-confidential and PHI-relevant content. DPA, BAA where applicable, sub-processor list, deletion timeline. Enterprise InfoSec review will reject vendors that haven't thought this through.

How Cornerstone captions intersect ADA Title II / III, Section 508, EAA, and OFCCP flow-down

Cornerstone tenants typically face several accessibility regimes simultaneously:

Section 508 — federal contractors hold accessibility flow-down obligations on the technology used to deliver employee training. The Cornerstone deployment plus the captioned training assets are part of the 508 procurement-evidence chain.
Section 504 — recipients of federal financial assistance (large healthcare systems, university systems, certain government contractors) carry the 504 functional-access obligation on training programmes.
ADA Title II — state and local government Cornerstone tenants (state-employee training systems, large county-government L&D, public-university HR training) carry the 2026-04-24 WCAG 2.1 AA bar.
ADA Title III — private-sector Cornerstone tenants face the indirect WCAG technical bar through case-law evolution; the safer baseline is WCAG 2.1 AA.
European Accessibility Act — EU-operating tenants in scope of EAA (B2C surfaces) face the EN 301 549 / WCAG 2.1 AA technical bar with member-state enforcement. See our Q3 2026 EAA inflection-point post for the enforcement landscape.
AODA — Ontario-operating tenants face the IASR § 14 WCAG 2.0 AA bar with three-year compliance reporting.

The technical caption requirement at WCAG SC 1.2.2 is consistent across regimes; the audit mechanism differs. The captioning provenance log per asset, plus the embedded-or-uploaded caption track, satisfies the technical bar in all of them simultaneously.