Accessible communications: captions, alt text and readable reports

Accessible communications means making sure people with disabilities can see, hear, read and understand what you publish. It is not a niche concern. The World Health Organization estimates that 1.3 billion people experience significant disability, which is 16% of the world's population, or one in six of us.¹

For organisations working with governments and the UN system, it is also a matter of rights. Article 21 of the Convention on the Rights of Persons with Disabilities asks states to provide information intended for the general public in accessible formats and technologies, in a timely manner and without additional cost.² The same article urges private entities that serve the public, including online, to do the same, and calls for recognising and promoting sign languages.² A report that a screen reader cannot open, or a campaign video without captions, quietly excludes the very people many programmes exist to serve.

The reference standard is the Web Content Accessibility Guidelines (WCAG) 2.2, published by the World Wide Web Consortium (W3C). It is written for web content, but its logic is just as useful for videos, photographs and downloadable reports. It is built on four principles: content must be perceivable, operable, understandable and robust.³

WCAG has three levels of conformance: A (lowest), AA and AAA (highest).³ In our view, level AA is the sensible target for any communications team, because it covers captions, audio description and minimum contrast. For everyday content, a handful of success criteria do most of the work:

• 1.1.1 Non-text content (Level A): images and other non-text content need a text alternative that serves the equivalent purpose.
• 1.2.2 Captions, prerecorded (Level A): recorded video with sound needs captions.
• 1.2.5 Audio description, prerecorded (Level AA): important visual information in video must be described.
• 1.4.1 Use of colour (Level A): colour must not be the only way information is shown.
• 1.4.3 Contrast, minimum (Level AA): body text needs a contrast ratio of at least 4.5:1, large text at least 3:1.

Captions are a text version of the speech and of the non-speech audio needed to understand a film, synchronised with the sound.⁴ The W3C uses captions and subtitles as the same term. Most are closed captions that viewers can switch on or off, while open captions are burned into the picture and always visible.⁴ Captions also help people who process written information better than audio.⁴

Automatic captions save time but, on their own, do not meet accessibility requirements unless they are confirmed to be fully accurate.⁴ A single missed word such as not can reverse a sentence. In our own practice, we treat machine captions as a first draft, then check names, place names, figures and Kinyarwanda, French or Swahili terms by hand, and add speaker names and meaningful sounds.

Blind viewers need the visual story too. The W3C recommends audio description for charts, on-screen text and speaker names, and a descriptive transcript, which is needed by people who are both Deaf and blind.⁵ It advises planning accessibility from the very start of a project and writing description into the script before filming.⁵ An interviewee who says their name and role on camera, for example, removes the need for a separate description of the lower third.

Alt text is the short description a screen reader reads aloud in place of an image. WCAG requires a text alternative for all non-text content that serves the equivalent purpose.³ The question is not "what is in the picture?" but "what is this picture doing here?"

The W3C's alt decision tree gives a clear rule set.⁶ For a simple photograph or graphic, write a brief description that conveys its meaning. If an image is in a link or button, describe where the link goes or what the button does. If an image is purely decorative, or repeats text that is already next to it, use an empty alt attribute so screen readers skip it. If an image contains text that appears nowhere else, put that text in the alt.⁶

For a field photograph, "Nurse checks a newborn's weight at a health centre in Nyamagabe" is far more useful than "image of woman". Keep it to a sentence, avoid "photo of", and do not repeat the caption word for word.

Annual reports and evaluations are often published only as designed PDFs. The UK Government Digital Service advises publishing in HTML wherever possible, noting that PDFs often do not work well with screen readers and that PDF and PDF/A cannot be made fully accessible to all users of assistive technology.⁷ A good approach is to publish the report as a web page and offer the PDF as a print version.

If you must publish a PDF, build accessibility into the source file in Word or your design software rather than trying to repair it at the end. Fixing a finished PDF is slow and rarely complete, so brief your designer on these points before layout begins. The same guidance recommends:⁷

• Use real heading styles (Heading 1, Heading 2) rather than bold text, and do not skip levels.
• Write link text that makes sense out of context, never "click here".
• Use tables only for data, keep them simple, avoid merged cells and mark the header row.
• Add alt text to images and charts, or make the same point in the text.
• Check colour contrast, then structure and tag the document in your PDF software.

Charts are where reports most often fail. A screen reader cannot read a bar chart, so the W3C advises giving the information in a complex image elsewhere on the page, for example as a short summary of the finding plus a data table.⁶ Do not rely on colour alone to separate categories: add direct labels or patterns.³ Check that text and key lines meet the contrast ratios in WCAG.³

Readable also means understandable. Plain language helps people with cognitive impairments and learning disabilities, and it helps everyone else. GOV.UK recommends explaining technical terms and acronyms the first time they appear and keeping sentences to around 25 words or fewer.⁷ For development audiences, a one-page summary in plain language, translated where needed, often reaches more people than the full report.

• Captions checked by a person, with speaker names and key sounds
• Visual information described in the script or in audio description
• A transcript published alongside every video
• Short, meaningful alt text on every informative image
• Empty alt on purely decorative images
• Reports published in HTML, or as tagged PDFs with real headings
• Text contrast of at least 4.5:1 and no information shown by colour alone
• Plain-language summary with acronyms explained