The AODA multi-year accessibility plan template for Ontario 50+ orgs, annotated section by section

Who AODA's multi-year accessibility plan obligation covers

IASR Section 4 imposes the multi-year accessibility plan requirement on two categories of organizations. The first is designated public-sector organizations — a defined term under the AODA that covers the Government of Ontario and its agencies, the Ontario Legislature, municipalities, school boards, district school boards, universities and colleges of applied arts and technology, public hospitals, transit commissions, hydro utilities, and any other body designated by regulation. Size does not matter for this category: a three-person municipality with a single employee qualifies. The second is private-sector and non-profit organizations with 50 or more employees in Ontario — including cross-border Canadian companies (British Columbia-based company with 60 Ontario employees: fully in scope), US-headquartered multinationals with Ontario operations (California-headquartered SaaS company with 80 Ontario employees: fully in scope for the Ontario workforce), and international organizations with Ontario-registered employees.

The 50-employee threshold is headcount across Ontario, not just Ontario-based employees of an Ontario legal entity. An organization that employs 30 people in Toronto and 25 people in Ottawa through a Canadian subsidiary has 55 Ontario employees and is in scope. Seasonal, part-time, and contract workers typically count, though the Ministry guidance is that it is "all employees" without specifying FTE. If your organization is near the threshold, apply the plan obligation — the cost of a false positive (doing a plan you didn't have to) is roughly one day of documentation work. The cost of a false negative (not doing a plan for an organization in scope) is a compliance order, a public posting on the ACA's non-compliance registry, and potentially a fine of up to $100,000 per day for corporations.

One-to-49-employee private organizations have a much lighter AODA footprint. They still need to file a compliance report (stating they have completed their obligations) and must comply with the Customer Service Standard, but they do not need a multi-year accessibility plan and are not subject to the WCAG 2.0 AA web-content obligation under Section 17. The training-video caption obligation under Section 14 (accessible formats on request) still applies at all sizes, however — if an employee or customer requests an accessible format, you must provide it.

Organizations often ask whether AODA applies to their Canadian operations if they have no Ontario-registered legal entity. The answer is yes if they employ workers in Ontario, because the employment relationship triggers AODA compliance obligations independently of the corporate structure. Cross-border orgs operating in Ontario and the US frequently hold simultaneous obligations under AODA, ADA Title II (if they are public entities), ADA Title III (if they are places of public accommodation), and the post-2026 ADA Title II training-video mandate. The cross-walk section below addresses how one captioning policy satisfies all three.

The legal framework: AODA, IASR, and the compliance schedule

The Accessibility for Ontarians with Disabilities Act, 2005 creates the framework and empowers the Lieutenant Governor in Council to make regulations specifying standards. The IASR (Ontario Regulation 191/11) is the primary implementing regulation and consolidates five standards that were originally proposed and phased in separately: the Customer Service Standard (AODA s. 6), the Employment Standard (IASR Part III), the Information and Communications Standard (IASR Part II), the Transportation Standard (IASR Part IV), and the Design of Public Spaces Standard (O. Reg. 413/12, later integrated). Your multi-year accessibility plan must address all five — even the standards whose requirements your organization has fully met — because the plan is a forward-looking strategy document, not just a checklist of open items.

The compliance schedule is staggered by organization type and size. Designated public-sector organizations have been subject to WCAG 2.0 AA web-content obligations since 2016-01-01. Large organizations (50–249 Ontario employees) have been subject since 2021-01-01. Small organizations (1–49 employees) are not subject to the WCAG web-content obligation but are subject to the Section 14 accessible-formats obligation on request. These deadlines have all passed. If your organization is subject to the WCAG obligation and has not yet brought its training-video library into conformance, the obligation is overdue — the plan needs to reflect both the fact of non-conformance and the specific timeline for remediation.

The compliance-reporting cycle runs separately from the plan. Organizations with 20 or more Ontario employees must file a compliance report with the Ministry for Seniors and Accessibility (now through the Accessibility Compliance Authority's online portal) every three years. The reporting windows have been: 2012, 2014 (for public sector — they report every two years), 2017, 2020, 2023 for private/NPO large organizations. The next private/NPO large-organization window is 2026-12-31. This is not the same as updating your multi-year accessibility plan — the plan is a standalone document, the compliance report is a separate filing that says you have one. Many organizations are unaware that an out-of-date plan (last updated 2021, not reviewed since) will be flagged as non-compliant in a 2026 audit even if the filing says the plan exists.

The Ontario Disability Support Program Act, the Human Rights Code (R.S.O. 1990, c. H.19), and the Employment Standards Act sit alongside AODA rather than inside it. An AODA compliance finding does not automatically constitute a Human Rights Code violation, but many facts that establish an AODA compliance failure also establish a Human Rights Code claim, particularly in employment contexts. The Human Rights Tribunal of Ontario hears both streams. When a training video lacks captions and an employee with a documented hearing disability cannot access the training, both streams open simultaneously: AODA s. 17(1) via the ICS, and Human Rights Code s. 5(1) (discrimination in employment on the basis of disability). The captioning project resolves both.

Why training video sits inside the Information and Communications Standard

The Information and Communications Standard (IASR Part II, ss. 9–22) is the AODA standard most directly relevant to training-video captioning. It operates on two parallel tracks. The first is accessible formats on request (IASR s. 14): any organization with any number of employees must, upon request from a person with a disability, provide information and communications in accessible formats or with communication supports. Training video is information. If an employee with a hearing disability requests that a training course they are required to complete be captioned, and it is not, you have a Section 14 exposure regardless of your organization's size or whether WCAG 2.0 AA applies to you. The obligation is triggered by the request, not by a proactive audit.

The second track is the WCAG 2.0 AA web-content obligation (IASR s. 17): designated public-sector organizations and large private/NPO organizations must make their websites and web-based content conform to WCAG 2.0 Level AA. The "internet and intranets" language in Section 17 extends to internal systems — LMS-hosted content, intranet-embedded training video, and Learning Record Store (LRS) or xAPI-reported content accessed through a browser all qualify as "web content" for AODA purposes. WCAG 2.0 AA's SC 1.2.2 requires synchronized prerecorded captions on any video with an audio track. If a training video is posted on your LMS after 2021-01-01 (the large-org WCAG obligation effective date) without captions, it is non-conformant under IASR s. 17.

The WCAG 2.0 vs 2.1 distinction matters in practice. AODA references WCAG 2.0; ADA Title II's 2024 final rule (28 CFR §§ 35.200–35.205) and the EAA's EN 301 549 V3.2.1 both reference WCAG 2.1 AA. The SC 1.2.2 captioning obligation is identical across 2.0 and 2.1 — the upgrade from 2.0 to 2.1 AA added 17 new success criteria (none directly related to prerecorded captioning), so for the captioning component of your plan, the 2.0/2.1 distinction is not practically meaningful. Where it matters is in colour contrast, text spacing, focus appearance, and motion — areas where organizations with a WCAG 2.0 AA plan may discover gaps if they apply 2.1 criteria. The ACA has, in audit guidance documents, indicated that conformance to a current published version of WCAG is the appropriate standard, which the Web Accessibility Initiative updates; treating 2.0 as a ceiling when 2.1 has been a W3C recommendation since 2018 is a documentation risk. For the captioning section of your plan, aligning to WCAG 2.1 AA SC 1.2.2 satisfies both AODA and any cross-border US or EU obligations simultaneously.

Section 18 (training to educators) is separate from Section 17 and covers the obligation of school boards and post-secondary institutions to provide training on AODA and the Ontario Human Rights Code to educators and support staff. It is not the main caption obligation, but it creates a direct back-loop: if the training your educators receive about AODA and accessibility is itself not captioned, you have an internal inconsistency that an ACA auditor will note. Ontario universities and colleges subject to both Section 17 and Section 18 face a compound obligation: the online course that trains instructors about accessibility must itself be accessible.

Section 22 requires organizations to notify the public and employees of the availability of accessible formats. For training video, this typically means an intranet or LMS landing-page notice: "Captioned versions of training courses are available. Contact [accessibility coordinator contact] to request an accessible format." If your LMS does not surface this notice automatically, your multi-year accessibility plan should include a timeline for adding it.

The multi-year accessibility plan: required structure

IASR Section 4 sets the structural requirements for the plan. Section 4(1) requires designated public-sector organizations and organizations with 50 or more Ontario employees to:

• Develop a written multi-year accessibility plan outlining their strategy to prevent and remove barriers and to meet their requirements under IASR;
• Post the plan on their website and provide it in an accessible format upon request;
• Review and update the plan at least once every five years for private-sector organizations, and at least annually for designated public-sector organizations;
• Establish, implement, maintain, and document a process for receiving and responding to feedback about the plan, and make information about the process publicly available.

Section 4 does not prescribe the exact sections or format of the plan, but the Ministry for Seniors and Accessibility's guidance and the ACA's compliance-review criteria provide a de facto template that has evolved through ten years of audits and advisory letters. Organizations that deviate significantly from this structure without good reason will draw audit inquiries. The annotated template below follows that structure and flags each annotation with the relevant statutory provision.

A common misunderstanding is that the plan needs to list only open items — things you have not yet done. In fact, the plan should include both accomplished items (showing progress since the prior version) and upcoming commitments. Auditors use the accomplished-items list to verify that annual updates actually occur. If your 2022 plan and your 2025 plan have identical "upcoming" sections with no movement on any commitment, that is a compliance failure independent of whether the underlying content was achieved.

The feedback process required by Section 4(1) must be substantive, not performative. A contact email address posted on the plan's web page does not automatically satisfy the requirement unless there is also a documented process for reviewing, responding to, and incorporating feedback. Your plan should describe: who receives feedback, within what timeframe they respond (the Ministry guidance suggests 15 business days for an initial response), how feedback is logged, and what decisions are made about incorporating it into the next plan update. If a student, employee, or member of the public submits feedback that your training video captions are inaccurate and the feedback is not addressed in the next plan update, that omission is documentable evidence of a process failure.

The annotated template

The following template is structured to satisfy the IASR Section 4 requirements as interpreted by the ACA's current audit criteria. Section headers in brackets indicate the relevant IASR provision. Annotations in italics explain what each section must contain and flag common errors. Copy and adapt for your organization — the bracketed placeholders are marked for replacement.

Section 1: Organization information and plan metadata

Annotation: Include this block at the top of the document so auditors can confirm the plan is current. The "plan period" should span 3–5 years for private-sector orgs; the annual review date is required for public-sector. Omitting the review date is the most common plan error flagged in ACA advisory letters.

Organization name: [Organization legal name]
Plan title: Multi-Year Accessibility Plan
Plan period covered: [e.g., 2024–2029]
Date of this version: [YYYY-MM-DD]
Date of last review: [YYYY-MM-DD]
Date of next scheduled review: [YYYY-MM-DD]
Accessibility coordinator: [Name and title, or "Contact accessibility@[organization].ca"]
Reporting obligation: [Designated public-sector organization / Large organization (50–249 employees) / Large organization (250+ employees)]

Section 2: Statement of commitment

Annotation: IASR s. 3 requires a statement of commitment to accessibility. The language below is based on the Ministry's guidance template and has been used without objection in ACA audits. Replace bracketed tokens. Do not dilute the language — statements like "We try to be accessible where possible" have drawn advisory letters citing insufficient commitment language. The statement must be posted publicly on your website and accessible formats must be made available on request (IASR s. 14).

[Organization name] is committed to treating all people in a way that allows them to maintain their dignity and independence. We believe in integration and equal opportunity. We are committed to meeting the needs of people with disabilities in a timely manner, and will do so by preventing and removing barriers to accessibility and by meeting our requirements under the Accessibility for Ontarians with Disabilities Act and its regulations.

This Multi-Year Accessibility Plan outlines our strategy to prevent and remove barriers to accessibility in all areas of our organization's operations, and to meet the requirements of the Integrated Accessibility Standards Regulation (O. Reg. 191/11). The plan is reviewed [annually / every three years] and updated to reflect progress and new commitments. We welcome feedback on the plan and on the accessibility of our programs and services — see Section 9 (Feedback Process) below.

Section 3: Accomplishments since last plan update

Annotation: List specific, verifiable accomplishments since the prior version of the plan. Vague items ("We improved accessibility") draw follow-up questions. Specific items ("We captioned 240 of 312 back-catalogue training videos in Q1–Q2 2025, bringing our LMS caption-coverage rate from 17% to 93%") demonstrate genuine progress and are the kind of evidence an ACA audit looks for. For training-video organizations, this section should include: caption coverage percentage before and after, number of videos remediated, captioning quality standard applied, and the LMS/platform hosting the captioned content.

Since the last update of this plan on [prior date], [Organization name] has accomplished the following:

• Information and Communications — Training Video Captioning: Captioned [X] of [Y] back-catalogue training videos in [LMS/platform], bringing caption coverage from [X%] to [Y%]. Captions conform to WCAG 2.0 Level AA Success Criterion 1.2.2 (synchronized prerecorded captions) and are served in SRT and VTT formats. [If applicable: Applied a company-specific captioning glossary across the [field] caption run, reducing proper-noun substitution errors by [Z%] relative to unassisted auto-captioning.]
• Employment Standard: [Description of any employment-standard accomplishment — e.g., updated job-posting template to include accessibility accommodation language per IASR s. 23.]
• Customer Service Standard: [Description — e.g., updated public-facing website chat to include TTY/relay service contact information per IASR s. 7.]
• [Additional accomplishments as applicable]

Section 4: Customer Service Standard

Annotation: IASR s. 6 (Customer Service Standard) requirements entered force for most organizations in 2012. Most organizations with a plan on file have already completed the core customer-service obligations. This section of the plan should reflect current status and any remaining commitments, not just check the boxes. The obligations relevant to training-video organizations: Section 7 (notification that assistance is available), Section 8 (training — all staff who provide services to the public), and the feedback process (already in scope of your plan itself). If your customer-facing staff include learners of online training who serve external customers, the intersection of customer-service and training-video captioning may appear here as well.

Current status: [Organization name] has implemented the Customer Service Standard requirements including: [list completed items — training of staff who interact with persons with disabilities; feedback processes; notice of service disruptions; service animals and support persons policies; assistive devices policy].

Upcoming commitments (this plan period):

• [Commitment 1 — e.g., Review and refresh staff training on the Customer Service Standard by [date] to ensure training reflects updated AODA obligations.]
• [Commitment 2 — if applicable]

Section 5: Information and Communications Standard — Accessible Formats

Annotation: IASR s. 14 (accessible formats and communication supports) applies to all organizations — it is not gated by the 50-employee threshold. The obligation is to provide information in accessible formats or with communication supports upon request and in a timely manner. For training-video organizations, the most common Section 14 exposure is an employee requesting a captioned version of a required training course that is posted without captions. If your organization has not yet completed back-catalogue captioning, this section must describe your process for fulfilling on-demand caption requests — not just your eventual plan to caption everything. The on-demand process is required now; the back-catalogue plan is a separate timeline.

On-request accessible format process: [Organization name] provides information and communications in accessible formats, or with communication supports, to persons with disabilities on request. Requests are submitted to [accessibility coordinator / HR / IT service desk]. We acknowledge requests within [5] business days and provide the accessible format or communication support within [15] business days, or as soon as reasonably possible.

For requests relating to training video, our process is: [e.g., For any required training course not yet captioned, we provide a verbatim transcript within 5 business days and a synchronized-caption version within 15 business days. The captioned version is then added to the LMS for all users.]

Upcoming commitments (this plan period):

• Complete back-catalogue captioning of all required training courses by [date], eliminating on-demand requests for mandatory training. [Milestone: 50% by Q2 [year]; 100% by Q4 [year].]
• Update the [LMS name] course-catalogue landing page to include an accessible-formats notice by [date].

Section 6: Information and Communications Standard — Accessible Websites and Web Content (WCAG 2.0 AA)

Annotation: This is the highest-stakes section for most training-video organizations. IASR s. 17 requires WCAG 2.0 AA conformance for new and significantly refreshed web content for public-sector organizations (effective 2016-01-01) and large private/NPO organizations with 50+ Ontario employees (effective 2021-01-01). "New content" means any content first posted after the applicable deadline. Training videos posted to your LMS after 2021-01-01 are "new content" and should have been captioned at time of posting, not retrospectively. The plan must address: (a) current conformance level, (b) back-catalogue remediation timeline, and (c) new-content process going forward. Failure to have a new-content process — that is, a process that ensures every new training video is captioned before it is posted — is the most common ICS audit finding for large organizations.

WCAG 2.0 AA conformance status:

As of [date], [Organization name]'s web-based training-video library (hosted on [LMS/platform]) has the following caption coverage:

• Required courses with synchronized captions: [X] of [Y total] ([%])
• Optional/supplementary courses with synchronized captions: [X] of [Y total] ([%])
• Back-catalogue videos posted before [2021-01-01 or 2016-01-01] without captions: [X]
• Videos posted after the WCAG obligation effective date without captions: [X] — Note: videos posted after the obligation effective date without captions are in active non-conformance under IASR s. 17 until remediated.

Back-catalogue remediation timeline:

• Triage back-catalogue by regulatory exposure and course-completion rate by [date].
• Caption highest-priority required courses (top [X] by completion count) by [date].
• Caption all required courses by [date].
• Caption all supplementary courses by [date].

New-content process:

Effective [date], no new training video is posted to [LMS/platform] without synchronized captions conforming to WCAG 2.1 AA SC 1.2.2 at a minimum accuracy level of 99% on the DCMP (Described and Captioned Media Program) scoring rubric. The captioning workflow is: [describe — e.g., all videos are submitted to GlossCap for captioning with the organization's current glossary applied; captions are reviewed by the course author before publication; the publication workflow in the LMS enforces the review step before the course is available to learners]. Responsibility for enforcing the new-content process sits with [role — e.g., L&D Operations Lead].

Note on WCAG 2.0 vs 2.1: AODA s. 17 references WCAG 2.0 AA. [Organization name] has elected to conform to WCAG 2.1 AA for all new content, which satisfies the AODA WCAG 2.0 AA obligation and simultaneously satisfies our obligations under ADA Title II (28 CFR § 35.200, effective 2026-04-24) and the European Accessibility Act (EN 301 549 V3.2.1) for content served to US and EU learners. The captioning Success Criterion (SC 1.2.2) is identical across WCAG 2.0, 2.1, and 2.2, so the WCAG version distinction does not materially affect our captioning compliance posture.

Section 7: Employment Standard

Annotation: The Employment Standard (IASR Part III, ss. 22–35) covers recruitment, return-to-work, performance management, career development, accommodation, and information available to employees. For training-video organizations, the Employment Standard intersects with the ICS when: (a) training is required for employees as a condition of employment or advancement, and (b) an employee's disability prevents them from accessing the training in its current format. The Employment Standard's return-to-work process (IASR s. 28) explicitly includes accommodation of disability in access to information and training materials. Your plan should confirm that accessible training is part of your return-to-work accommodation framework, not just a separate ICS matter.

Upcoming commitments (this plan period):

• Update individualized accommodation plans (IASR s. 28) to include a training-video captioning section, confirming that any required training not available in accessible format is flagged to the L&D team for expedited captioning by [date].
• Include accessible-format notification language in all job postings by [date], per IASR s. 23.
• Confirm that performance management, career development, and redeployment information is available in accessible formats on request (IASR ss. 27, 29, 30).

Section 8: Transportation Standard and Design of Public Spaces Standard

Annotation: If your organization does not operate passenger transportation services or construct/redevelop built environment spaces open to the public, state that clearly. Auditors understand that these standards may not apply — but the plan should say so explicitly rather than omitting the sections, because omission implies you forgot rather than that you assessed and determined inapplicability.

Transportation Standard: [Organization name] does not operate conventional transit, specialized transit, or other passenger transportation services subject to the Transportation Standard (IASR Part IV). This standard is not applicable to our operations.

Design of Public Spaces Standard (O. Reg. 413/12): [Organization name] [does not construct or redevelop recreational trails, beach access routes, outdoor public eating areas, outdoor play spaces, exterior paths of travel, accessible parking, or service counters subject to this standard / does construct or redevelop the following public spaces and will comply with the Design of Public Spaces Standard as follows: (list commitments)].

Section 9: Feedback process

Annotation: IASR s. 4(1)(e) requires a documented process for receiving and responding to feedback about the plan. The feedback process must be posted publicly. It must be accessible (accessible formats on request for the feedback process itself). Feedback received must be logged and considered in the next plan update. If you received feedback in 2024 and it does not appear in any way in the 2025 plan update, that is a gap an auditor can document.

We welcome feedback on this plan, on our accessibility policies, and on the accessibility of our programs, services, training, and communications.

How to submit feedback:

• Email: [accessibility@organization.ca]
• Telephone: [phone number, TTY: if available]
• Mail: [mailing address], Attention: Accessibility Coordinator
• Online form: [URL if available]

Response process: We acknowledge feedback within [5] business days. We provide a substantive response within [15] business days or as soon as reasonably possible. Feedback received is logged by the Accessibility Coordinator, reviewed quarterly, and incorporated into the annual plan review where relevant.

This feedback process and all documents describing it are available in accessible formats upon request.

Section 10: Review schedule and version history

Annotation: Include a version table. ACA auditors use this to verify that annual reviews (for public-sector) or regular reviews (for private-sector) actually occurred. A plan with version 1.0 dated 2020 and no version 2.0 is an immediate compliance flag for organizations that should have updated annually.

Version	Date	Summary of changes	Approved by
1.0	[date]	Initial plan	[Name, title]
2.0	[date]	[Summary — e.g., Updated ICS caption-coverage data; added new-content process; updated accomplishments section]	[Name, title]
3.0	[date]	[Summary]	[Name, title]

Next scheduled review date: [date]

Person responsible for next review: [Name and title]

Captioning-specific annex: what to add if your plan hasn't been updated since your training-video library grew

The single most common gap we see in AODA multi-year accessibility plans for organizations that have grown their training-video libraries since 2021 is the absence of a captioning-specific annex or subsection. The template's Section 6 above addresses WCAG 2.0 AA web-content conformance, but most plans written in 2021–2023 predate the post-ADA-Title-II-deadline urgency that pushed training-video captioning up the priority ladder. If your plan's ICS section says "we will caption new content going forward" but does not include: (a) current back-catalogue coverage data, (b) a specific back-catalogue remediation timeline with milestone dates, (c) a captioning quality standard (not just "we will add captions"), and (d) a new-content enforcement process, update it before your 2026 compliance report filing.

Captioning quality matters for AODA compliance in a way that most plans do not address. WCAG 2.0 AA SC 1.2.2 requires "captions" that are "synchronized" with the audio — but the Described and Captioned Media Program quality standard, which the ACA references in enforcement guidance, specifies that captions must be accurate (minimum 99% word accuracy), verbatim (not paraphrased), properly timed (synchronised to within 2 seconds of the audio), appropriately positioned, and formatted for readability. Auto-generated captions from YouTube, Microsoft Stream, or any LMS native captioning feature typically achieve 80–90% accuracy on general-vocabulary content and 65–80% accuracy on content with technical, medical, legal, or multilingual proper nouns. A YouTube-auto-captioned training video posted to your LMS is not WCAG 2.0 AA conformant under the substantive accuracy interpretation, even though it has captions in a technical sense.

The proper-noun problem is especially acute in Ontario training content because of the vocabulary density of provincial programs, regulatory citations, Indigenous community names, bilingual French-English terminology, and cross-border Canadian legal citations that appear in compliance training, healthcare training, university lecture content, and government-sector workforce training. Common failure modes in Ontario-specific content include:

• Provincial regulatory citations: PHIPA (Personal Health Information Protection Act), MFIPPA (Municipal Freedom of Information and Protection of Privacy Act), OHSA (Occupational Health and Safety Act), WSIB (Workplace Safety and Insurance Board), PIPEDA — ASR systems frequently mangle the acronyms and the full-form statute names, particularly when spoken in sequence with US regulatory citations that have different phonological profiles.
• Bilingual French-English proper nouns: L'Université de l'Ontario français, Conseil scolaire de district catholique du Centre-Sud, OSAP (Ontario Student Assistance Program), francophone community names across northern and eastern Ontario — ASR systems trained predominantly on English data systematically fail on French proper nouns spoken in otherwise English-language training content.
• Indigenous community names and treaty names: Anishinaabe, Haudenosaunee, Mississaugas of the Credit, Nishnawbe Aski Nation, Robinson-Huron Treaty, Williams Treaties — these proper nouns are increasingly expected in land-acknowledgement statements that appear at the opening of training videos for public-sector and university organizations, and auto-captioning failure on them is both a compliance failure and a respect issue that has drawn specific feedback in ACA advisory contexts.
• Healthcare drug and procedure vocabulary for Ontario health sector training: Provincial health program names (OHIP, Community Paramedicine, Ontario Health atHome), hospital network names (Unity Health, Trillium Health Partners, Sinai Health, St. Joseph's Health System, Mackenzie Health), and the same drug INN/brand-name failure modes documented in the drug-name ASR failure mode post for US healthcare content.
• Higher-education program and discipline vocabulary: Ontario university program names, college diploma names, research-institute names, and the dense technical vocabulary of STEM and health-sciences lecture capture content hosted on Brightspace (the dominant Ontario university LMS), Canvas, and Moodle. The university lecture captioning page documents this failure mode in detail.

Your plan's captioning-specific annex should describe your captioning methodology in enough detail to answer an auditor's question about how you achieve substantive accuracy — not just whether captions exist. The following template language works:

Captioning quality standard: [Organization name] requires that all training-video captions conform to WCAG 2.1 AA SC 1.2.2 at a minimum accuracy of 99% as measured against the DCMP scoring rubric. We do not use unreviewed auto-generated captions as the final caption product for any required training course.

Captioning methodology: Training videos are [sent to our captioning vendor / processed through our internal captioning workflow] with a vocabulary glossary that includes [organization-specific proper nouns — product names, system names, personnel titles, regulatory citations, program names]. The glossary is updated [quarterly / when new technical content is produced]. Caption output is reviewed by the content author before the course is published to the LMS. Caption files are stored in [SRT / VTT / both] format and are version-controlled alongside the video source file.

Back-catalogue remediation log: [Link to internal log or summary table showing videos remediated, date captioned, captioning vendor/method, and current accuracy spot-check result.]

The proper-noun failure mode in Ontario training content

The failure mode is consistent across all AI speech recognition systems: the words that matter most in domain-specific training content are exactly the words the model is least likely to have seen in sufficient training data. For Ontario training content, the proper-noun vocabulary surface is distinctive in several ways that go beyond what a US-focused captioning vendor will anticipate.

Ontario provincial program and regulatory vocabulary includes several dozen frequently-cited statutes and program names that are phonologically unusual in English — PHIPA sounds like "feepah" to a model without a vocabulary prior; MFIPPA sounds like "em-FIPPA" with a variable stress pattern depending on the speaker. The WSIB acronym is frequently transcribed as "W-SIB" or "wise-bib" or simply missing, and the full form (Workplace Safety and Insurance Board) is often split across speaker turns in training content with multiple presenters. Provincial legislation citation strings like "O. Reg. 191/11" or "R.S.O. 1990, c. H.19" require the model to correctly hear and transcribe an Ontario-specific citation format that does not appear in federal Canadian or US legislation.

French-language terms present a systematic failure mode for English-ASR systems. A training video that says "our D2L Brightspace platform supports both Anglophone and francophone learners through the Conseil scolaire de district catholique" will reliably produce garbled transcriptions of the French institution names — not because the ASR system can't hear French, but because the transition from English to French proper nouns mid-sentence creates a phonological context switch the English-trained model is not built for. This is particularly relevant for northern Ontario organizations, francophone universities, and any organization whose training content addresses Ontario's francophone-services obligations under the French Language Services Act (R.S.O. 1990, c. F.32).

Indigenous community names present a distinct challenge. Anishinaabe is a 6-syllable word with a stress pattern that does not map to common English words; ASR systems transcribe it as "anishnawbek," "anishi-naub," "a-nishi-nabe," or miss it entirely. Haudenosaunee is 5 syllables with similar issues. These words appear in land acknowledgements at the opening of training videos produced by Ontario universities, municipalities, school boards, hospitals, and government agencies — all AODA-designated public-sector organizations with the full WCAG 2.0 AA caption obligation. An auto-captioned land acknowledgement with mangled Indigenous community names is not just a compliance failure; it is a reputational and relational issue that organizations address when it is brought to their attention, usually too late in the audit cycle.

The remedy for all three failure modes — regulatory vocabulary, French proper nouns, and Indigenous community names — is the same: a vocabulary glossary that is applied during the captioning process as a prior. A 200-term glossary covering an organization's common acronyms, program names, community names, and French terminology will resolve 90%+ of these failure modes in a single pass. The residual errors (phonologically ambiguous terms that share a root with common English words, speaker-specific pronunciation patterns, non-standard acronym expansions) are handled by a human-review step before the caption file is published. This is the workflow GlossCap uses: see the live demo of glossary-biased captioning on domain-specific Ontario training content.

Compliance reporting: the ACA audit pattern and what happens at a finding

The Accessibility Compliance Authority is the Ontario government body responsible for enforcing AODA. It was established as the successor to the Accessibility Directorate of Ontario's compliance-enforcement function, with a mandate to conduct proactive audits in addition to responding to complaints. The ACA selects organizations for compliance reviews based on: industry sector (it has publicly announced sector-specific audit sweeps for healthcare, post-secondary, and financial services); size (large organizations face more frequent reviews than small ones); prior compliance-report history (organizations that filed late or with gaps are more likely to be selected for follow-up); and complaint intake (a complaint from an employee, student, or member of the public about inaccessible training content can trigger an individual-organization audit).

The ACA audit methodology for web-content compliance involves an auditor selecting a sample of web content from the organization's public website and accessible portions of its intranet (if the auditor has credentials to access the intranet — this is negotiated during the audit intake process). The sample is tested against WCAG 2.0 AA using a combination of automated tools (axe-core, WAVE, or similar) and manual inspection. For video content, the auditor plays the video and reviews the caption track against the DCMP scoring criteria. A 10-video sample is typical. If the auditor finds that 3 of 10 sampled videos lack captions or have substantively inaccurate captions, the finding is recorded as "non-conformant" under IASR s. 17.

A compliance finding does not immediately result in a fine. The ACA's enforcement sequence is: (1) issue an advisory letter or notice of non-compliance to the organization, stating the specific IASR requirement that was not met and requesting a response within a specified period (typically 30–60 days); (2) if the organization responds with a remediation plan and timeline that the ACA considers credible, the ACA monitors progress and may conduct a follow-up audit; (3) if the organization does not respond, denies the finding, or fails to remediate within the agreed timeline, the ACA issues a compliance order; (4) compliance orders are public — they are posted on the ACA's website and searchable by organization name; (5) if the compliance order is not followed, the ACA may seek a monetary penalty of up to $50,000 per day for an individual and $100,000 per day for a corporation, per contravention.

In practice, most organizations that receive an advisory letter for training-video captioning non-compliance do so for the first time in a follow-up audit after a 2026 compliance report filing, rather than as the result of a complaint. The 2026 filing window is therefore both a filing deadline and a de facto audit trigger: organizations that file and confirm "we are WCAG 2.0 AA conformant" while having large portions of their training-video library without captions are filing inaccurate reports. The ACA cross-checks compliance-report claims against audit sampling, and an inaccurate report worsens the remediation trajectory.

For cross-border organizations, an AODA compliance finding does not automatically constitute an ADA Title II or EAA violation, but the same facts that establish AODA non-conformance will typically establish non-conformance under ADA Title II (if the organization is a US public entity) or EAA (if the organization serves EU consumers). Running a single captioning project that closes the AODA gap closes all three simultaneously. The cross-walk section below explains the relevant differences to watch for.

Cross-walk: AODA vs ADA Title II vs EAA for cross-border orgs

Organizations with Ontario operations that also have US public-entity status (public universities with US campuses, Canadian subsidiaries of US state instrumentalities) or EU consumer-facing operations hold simultaneous obligations. The three regimes are closely related but not identical, and the differences matter for how you write your plan and vendor contracts.

The practical implication for cross-border organizations is that a single captioning project conforming to WCAG 2.1 AA SC 1.2.2 (the most stringent common denominator) satisfies all three regimes' caption obligation simultaneously. The plan-document requirement is regime-specific: AODA requires a multi-year accessibility plan (this template); ADA Title II requires a transition plan for entities over 50 employees; the EAA requires an accessibility statement (see the annotated EU accessibility statement template in the sister post to this one). These three documents are distinct and non-interchangeable — you need all three if you hold obligations under all three regimes — but the underlying captioning compliance posture that satisfies all three is the same: WCAG 2.1 AA SC 1.2.2 at ≥99% accuracy.

For Ontario universities with US campus partnerships, the ADA Title II post-2026 compliance obligation under 28 CFR § 35.200 applies to the US public-entity operations. Section 504 applies to any operations receiving US federal financial assistance. The university captioning vendor guide covers the multi-framework RFP process in detail. At the vendor level, the important cross-regime requirement is that your captioning vendor's data processing agreement satisfies both PIPEDA/PHIPA (for Ontario data) and any applicable US regulation (FERPA for student data, HIPAA BAA for clinical data, GDPR Article 28 for EU-origin data) — single vendor, three data-processing frameworks. The captioning RFP playbook provides the security-section questionnaire template for exactly this scenario.

The captioning procurement that typically follows the plan update

Organizations that update their AODA multi-year accessibility plan to include substantive captioning commitments typically discover, in the process of writing the plan's ICS section, that they do not yet have a captioning vendor relationship, or that their existing relationship does not include glossary-based captioning for Ontario-specific vocabulary, or that their current LMS workflow does not enforce the new-content captioning step before publication. The plan update and the vendor selection process happen simultaneously, and the plan's ICS section ends up describing a workflow that does not yet exist operationally. That is fine — the plan is forward-looking — but the commitments in the plan need to be achievable within the stated timeline. Promising in the plan that all back-catalogue videos will be captioned by 2025-12-31 when you have 3,000 videos and no vendor in place is a commitment you will not meet, and a missed commitment creates a worse compliance posture than a longer timeline honestly stated.

The practical steps for organizations running this process concurrently:

1. Inventory: Pull the list of all training videos in your LMS (most LMS administrators can export a course catalogue with video metadata including duration and publication date). Separate by: (a) already captioned with verified-quality captions, (b) has auto-generated captions only, (c) no captions at all. Note publication date against the WCAG obligation effective date (2021-01-01 for large private/NPO orgs; 2016-01-01 for public sector).
2. Triage: Sort by regulatory exposure (required courses vs elective), completion rate (high-completion courses affect more people and draw more ACA sampling weight), and content type (healthcare, engineering, legal, and multilingual content have the highest proper-noun failure rates and should be prioritized for human-review captioning rather than auto-only).
3. Vendor selection: Issue an RFP or request for quote. The 14-question captioning RFP template covers accuracy floor, glossary architecture, Ontario-specific vocabulary handling, format support (SRT/VTT/TTML/WebVTT for the major Ontario LMSs), LMS integration depth, security and data residency (PIPEDA/PHIPA compliance for Ontario health data, OSAP/FERPA compliance for student data), SLA, pricing, and termination/portability. The full first-person RFP walkthrough is in the captioning RFP playbook.
4. Timeline commitment: Once you have a vendor quote for the back-catalogue volume and a per-video turnaround time, you can commit to a realistic back-catalogue remediation timeline in your plan. A vendor processing 50 hours of video per week on a 2,000-hour back-catalogue closes the gap in 40 weeks. A vendor processing 10 hours per week takes five years. Your plan commitment should reflect your actual vendor capacity.
5. New-content enforcement: The most durable fix is a workflow change in your LMS that prevents a course from being published to learners until the caption file has been uploaded and approved. Most LMSs support this through a course-status workflow or a publishing checklist. For Brightspace, this is a Workflow step in Release Conditions; for Canvas LMS, it is a Publishing pre-check; for Moodle, it is a Lesson or Activity completion requirement. Describe the specific workflow implementation in your plan's ICS section — not just "we will ensure new content is captioned" but "we have implemented a [Canvas/Brightspace/Moodle] publishing workflow that requires a caption file to be uploaded before the course status can be set to Published."

Organizations at the 50–100 employee tier typically find that the back-catalogue is manageable (100–500 hours of video) and the per-video captioning cost at a professional vendor is $1.25–$3.00/minute — a $10,000–$90,000 one-time project, depending on content complexity and whether human review is included on all videos or only high-priority ones. The ongoing cost for new content is much smaller — if your organization produces 5 hours of new training video per month and all of it is captioned at $2.00/minute, that is $600/month for captions on all new content. Compared to the FTE cost of manual caption correction (1–2 hours of correction per video hour at $40/hour burdened cost = $200–$400/hour of video), a professional captioning workflow at $120/hour of video generates a labour savings that pays for itself within the first year. The hidden half-FTE cost post documents this math in detail.

30-day publication plan

If your organization's multi-year accessibility plan is out of date and you need to file an updated compliance report by 2026-12-31, this is a workable 30-day sprint to get the plan updated and posted before the 2026 reporting window closes.

FAQ

Our plan was published in 2021 and we've had an annual review note added each year. Is that sufficient?

It depends on what the annual review note says. If the review note states "plan reviewed; no changes required" but your training-video library has grown substantially and your captioning commitments from 2021 were either not met or not updated, the review is technically present but substantively empty. The ACA's enforcement guidance treats an annual review that does not update commitments to reflect actual progress as an incomplete review. If your 2021 plan committed to "caption all required courses by 2023" and your 2024 and 2025 review notes say "no changes required" but 40% of required courses are still uncaptioned, you have a compliance gap that a 2026 audit will identify. Review notes need to say what was assessed, what progress was made, and what the updated commitments are — not just that a review occurred.

We're a small SaaS company with 55 Ontario employees — do we really need a full multi-year accessibility plan?

Yes, if you have 50 or more Ontario employees. The multi-year accessibility plan requirement (IASR s. 4) applies to all private-sector organizations with 50 or more Ontario employees. The plan can be proportionate to your size and operations — a 55-employee SaaS company does not need the same depth as a 5,000-person hospital system — but the core sections are required regardless: statement of commitment, ICS compliance status (including web-content and training-video captions), Employment Standard status, applicable feedback process, and review schedule. The ACA has issued advisory letters to organizations with 50–100 employees for missing plans, particularly in the software and technology sector following sector-specific audit sweeps.

AODA says WCAG 2.0 AA. Our ADA Title II obligations say WCAG 2.1 AA. Which do we target?

Target WCAG 2.1 AA. WCAG 2.1 is a strict superset of WCAG 2.0 — a document that conforms to WCAG 2.1 AA automatically conforms to WCAG 2.0 AA. Conforming to 2.1 AA satisfies both your AODA obligation and your ADA Title II / EAA obligation simultaneously. For the captioning component specifically, SC 1.2.2 is identical in both versions, so the caption-quality standard you apply to satisfy 2.1 AA also satisfies 2.0 AA. The only practical issue is if your plan language states "we conform to WCAG 2.0 AA" — consider updating it to "we conform to WCAG 2.1 AA (which exceeds the WCAG 2.0 AA standard required by IASR s. 17)" to make clear that you have exceeded, not merely met, the AODA technical floor.

Our LMS vendor says their auto-captioning is WCAG 2.0 AA compliant. Is that sufficient?

In most cases, no. LMS vendors who claim WCAG 2.0 AA compliance for their auto-captioning feature are typically making a claim about the accessibility of the caption interface — that the caption display controls conform to WCAG — not a claim about the accuracy of the caption content. SC 1.2.2 requires "captions" but does not specify a minimum accuracy level in the criterion text. The ACA's enforcement guidance and the DCMP quality standard both interpret "captions" to mean substantively accurate captions (≥99% word accuracy), not auto-generated output at 80–90% accuracy with proper-noun failure rates of 30–40% on domain-specific content. If you rely on your LMS vendor's auto-captioning feature as your WCAG compliance mechanism, request their documented accuracy rate on content similar to your training library, with specific data on proper-noun accuracy. If they cannot provide this, assume the auto-captioning output is not WCAG-conformant for your content and add a human-review step.

We have a French-language obligation under the Ontario French Language Services Act in addition to AODA. Does the multi-year accessibility plan need to address both?

The AODA multi-year accessibility plan and French Language Services Act obligations are separate compliance instruments — you need a plan for each. However, the captioning component has a practical intersection. If you produce French-language training video (or bilingual French-English training video) for Ontario francophone employees or service recipients, those videos need captions that are accurate in French — not just a French text track generated by running a French ASR on French audio. The proper-noun failure mode in French training content includes the same regulatory citations, institution names, and program names as English content, compounded by the phonological characteristics of Québécois French (which many ASR systems trained on standard European French do not handle well) and by the code-switching patterns in bilingual Ontario content. The glossary-based captioning methodology applies to French content with a French-language glossary — French acronyms (LSSTO, AADNC, INMS), French program names, and French regulatory citations all need glossary coverage equivalent to the English-language glossary your English training content uses.

Our next compliance report is due 2026-12-31. If we update the plan in November 2026, is that too late?

It is not too late to file, but it is riskier than updating now. ACA auditors who receive a 2026 compliance report from an organization whose plan is dated November 2026 may ask whether the plan reflects genuine ongoing accessibility work or whether it was assembled solely to satisfy the filing deadline. An organization that updates its plan annually and has a version history showing 2023, 2024, and 2025 reviews is in a clearly stronger position than one producing a plan for the first time in November 2026. Additionally, if your back-catalogue captioning commitment requires 6–12 months of vendor work, starting in November 2026 means your stated commitments will be dated 2027–2028, which are plausible but will be subject to follow-up audit. Starting the captioning project now and filing the 2026 report with demonstrable progress (e.g., "back-catalogue captioning commenced [date]; 60% complete as of filing date") is a substantially better compliance posture than starting and filing simultaneously with 0% completion.

We're a Brightspace-heavy Ontario university. Are there specific AODA captioning considerations beyond the standard template?

Yes. Ontario universities are designated public-sector organizations under AODA, which means the WCAG 2.0 AA obligation effective date was 2016-01-01 — five years earlier than for large private organizations. A back-catalogue of lecture capture videos posted between 2016 and 2021 that were not captioned represents ten years of non-conformance in an active ACA enforcement environment for the post-secondary sector. The ACA has conducted sector-specific audit sweeps for Ontario post-secondary institutions; universities that are selected will face scrutiny on their entire LMS video library, not just content posted after 2021. Additionally, the IDEA Part D and Section 504 crossover (for any US-federal-financial-assistance-receiving programs) means that US-linked Ontario universities may face parallel OCR complaints from students with disabilities enrolled in joint programs. The Brightspace captioning guide covers the specific AODA obligations for D2L-hosted content, the AODA three-year compliance reporting cycle for universities, and the Accessibility Compliance Authority sampling pattern for post-secondary web content. For the procurement side, the university captioning vendor guide covers the multi-framework RFP process for institutions holding AODA, Section 504, Section 508, ADA Title II, and FERPA obligations simultaneously.