How do I keep a caption and alt text from repeating each other?

Start by asking why the image appears on this page and what a reader needs from it. Alt text should convey the image’s essential information or function in context; that may be its meaning, rather than a literal inventory of everything visible. What counts as informative depends partly on what the surrounding page already communicates. [S1]

Let the caption add something useful, such as context, attribution, or interpretation. These are editorial options, not a universal formula: web accessibility guidance does not prescribe one fixed rule for caption writing. [S1]

A simple check: read the nearby text, caption, and alt text together. Does each contribute something useful? If two lines convey the same fact in the same way, consider revising one. Some overlap may still be useful for clarity; the aim is not different wording at any cost.

Suppose the page explains that volunteers planted a garden in a formerly unused lot, and the photo shows several people planting seedlings. These sample details are hypothetical; use only facts you can verify about your own image and page.

Possible caption: “Volunteers plant seedlings in the [neighborhood] garden during its opening day.”

Possible alt text: “Volunteers kneel along a garden bed, placing seedlings into the soil.”

Here the caption adds event and location context, while the alt text conveys relevant visual information. If nearby text already explains that the photo shows volunteers planting seedlings, ask whether repeating those details in the alt text adds anything. The best alternative depends on what information the image contributes in context. [S1] [S2]

Now suppose the image is a link to the garden’s event schedule. Its function may matter more than a detailed visual description. Write an alternative that communicates the link’s purpose, and use the caption for any separate context. Only name a destination if the link actually goes there.

Ask whether both pieces are needed. If a caption adds no useful context, you need not write one just to paraphrase the alt text. And W3C says purely decorative web images need no alternative text. [S2]

Don’t decide an image is decorative just because it looks attractive. If it identifies a person, illustrates a point, communicates information, or performs a function such as linking, consider what a reader would miss without it. Review the image alongside the page text and judge what it contributes there. The guidance supports making that decision in context; it does not prescribe one wording for every page. [S1]

Keep claims consistent, too. If the caption calls the event an opening day but you cannot confirm that, verify or remove the detail. Don’t make the alt text compensate for an uncertain caption.

Yes. Fill this in for each image, then compare the drafts with the surrounding page text.

Caption-and-description division-of-labor worksheet

• Image and page purpose: What is shown, and why is it here?
• Nearby text: What relevant information is already communicated?
• Image’s essential information or function: What does a reader need from it in this context?
• Alt text draft: Communicate that information or function without unnecessary detail.
• Caption’s separate contribution: What useful context, attribution, interpretation, or other information can it add?
• Overlap: What is repeated, and does keeping it help clarity?

For example, if nearby text already states the date and location, a caption might credit the photographer while the alt text conveys the image’s relevant content. If the image is purely decorative, note that rather than forcing a description. W3C’s advice concerns web images and context; it does not require captions and alt text always to use different words. [S1] [S2]