{"schema_version":"2.0","record_type":"article","canonical_url":"https://marketingwiki.ai/articles/email-image-captions-vs-alt-text","id":"email-image-captions-vs-alt-text","slug":"email-image-captions-vs-alt-text","title":"Email Image Captions, Alt Text, and Body Copy Have Different Jobs","description":"Decide what belongs in a visible image caption, HTML alt text, nearby body copy, or nowhere, with an email review worksheet.","dek":"A caption adds visible context, alt text supplies a text alternative, and body copy must carry information every recipient needs to act on.","category":"Email Accessibility","topics":["email accessibility","alt text","image captions","Migma","email content"],"publishedAt":"2026-09-02","updatedAt":"2026-09-14","lastVerifiedAt":"2026-09-14","readingMinutes":6,"author":"Marketing Wiki Research Automation","reviewer":null,"featured":false,"sources":[{"title":"Migma Product Changelog","url":"https://docs.migma.ai/changelog?utm_source=marketingwiki&utm_medium=referral&utm_campaign=email-image-captions-vs-alt-text"},{"title":"Migma Visual Email Editor","url":"https://docs.migma.ai/email-editor/visual-editor?utm_source=marketingwiki&utm_medium=referral&utm_campaign=email-image-captions-vs-alt-text"},{"title":"W3C Images Tutorial","url":"https://www.w3.org/WAI/tutorials/images/?utm_source=marketingwiki&utm_medium=referral&utm_campaign=email-image-captions-vs-alt-text"},{"title":"WCAG 2.2 Non-text Content","url":"https://www.w3.org/WAI/WCAG22/Understanding/non-text-content.html?utm_source=marketingwiki&utm_medium=referral&utm_campaign=email-image-captions-vs-alt-text"}],"wordCount":1124,"body":"An email image caption, its HTML alt text, and the surrounding body copy are not interchangeable. Use the caption for context every sighted reader should see, alt text for the image’s meaningful text alternative, and body copy for information the recipient needs regardless of whether the image loads or is perceived.\n\n> **Editorial disclosure:** Prepared by Marketing Wiki Research Automation under standing direct-publication authorization and not independently reviewed. Product capabilities are vendor-documented unless labeled otherwise; sources were refreshed on September 2, 2026.\n\nMigma added image captions to its Visual Editor on August 31, 2026. The same editor separately exposes alt text for an image. That separation is useful: it gives the team distinct fields, but the editor cannot decide which sentence belongs in which field.\n\n## The four destinations\n\n| Destination | Who receives it | Put this there | Do not put this there |\n| --- | --- | --- | --- |\n| Visible caption | Everyone who sees the rendered image and caption | Source, date, product variant, place, or short interpretation tied to the image | The only copy of a critical instruction or offer term |\n| Alt text | People using assistive technology and recipients whose image is unavailable, depending on client behavior | A concise equivalent of the image’s purpose in this message | “Image of,” decorative detail, or a second marketing paragraph |\n| Nearby body copy | Every recipient in the normal reading flow | Decision-critical fact, qualification, price condition, deadline, or action | Detail that only identifies a decorative flourish |\n| Empty alternative | Assistive technology should skip a decorative image when implemented correctly | Nothing, when the image adds no meaning and the email code marks it decorative | An informative or linked image |\n\nThe table is a writing tool. Final accessibility depends on the generated markup and receiving client, so inspect the result rather than approving field labels alone.\n\n## Start with the image’s job in this email\n\nW3C’s images guidance separates informative, decorative, functional, complex, and text images. The same asset can change category when its job changes.\n\nA shoe photo beside the product name and price may be partly illustrative; alt text such as “Blue trail shoe with orange sole” can convey the choice the image helps make. The same photo used as a linked product card is also functional, so the link’s accessible name and destination matter. A texture behind a fully written headline may be decorative. A chart summarizing experiment results is complex and needs more than a filename or one vague sentence.\n\nAsk: if the image disappeared, what meaning or action would the recipient lose? That answer defines the content that needs an equivalent. It does not automatically define the exact alt text; some essential information belongs in ordinary body copy so everyone gets it.\n\n## When a caption earns its place\n\nUse a visible caption when the note improves interpretation for all sighted readers but would interrupt the main paragraph. Common examples include:\n\n- “Prototype shown; final packaging may change.”\n- “Storefront in Girona, August 2026.”\n- “Dark-mode preview in Outlook for Windows.”\n- “Results cover the treatment group from the August 12 test.”\n\nMigma documents that captions inherit the brand’s font and colors. That is a presentation capability, not evidence that the caption has sufficient contrast, survives every export, or is programmatically associated with the image. Review the actual output.\n\nDo not hide an offer restriction in a tiny caption. If the recipient must know that a discount excludes subscriptions, put the restriction in the normal offer copy as well. The team should not depend on a visual relationship that may weaken when the email reflows.\n\n## Write alt text as an equivalent, not a label\n\nWeak alt text inventories pixels: “Image,” “banner,” or “woman holding product.” Useful alt text explains the purpose the image serves in this specific message: “Mara holding the compact brewer beside a carry-on bag.” If the nearby heading already says “Compact brewer,” remove repetition and keep only the additional meaning.\n\nFor an image containing words, move important words into real HTML text when possible. If the image is the only place a campaign date appears, a concise alternative is necessary but still leaves other readers vulnerable when images are blocked. Put the date in body copy.\n\nFor a chart, provide the conclusion and relevant values in nearby text or a linked accessible detail page. Alt text can identify the chart and summarize its purpose; it is rarely a good container for an entire dataset.\n\n## Handle linked images separately\n\nWhen an image is clickable, the control needs a useful accessible name. “Shop the blue trail shoe” describes the action better than “image” or a raw URL. Check how the destination platform nests the image and link, because the source editor may not be the final markup owner.\n\nIf a card includes linked image, linked heading, and linked button that all go to the same place, review the experience with images off and with assistive technology. Repeated identical destinations can create noise. This is an implementation review, not something a caption solves.\n\n## Image-copy worksheet\n\nComplete one row for every meaningful image:\n\n| Field | Reviewer entry |\n| --- | --- |\n| Image ID and email version | |\n| Purpose in this message | Informative / decorative / functional / complex / text |\n| Meaning lost if unavailable | |\n| Body copy carrying essential facts | |\n| Alt text or decorative treatment | |\n| Visible caption and reason | |\n| Link destination and accessible action | |\n| Mobile and dark-mode observation | |\n| Destination-platform test result | |\n\nA blank caption is not a failure. A duplicated caption is often a sign that the team has not decided where the information belongs.\n\n## Review in three passes\n\nFirst, read the email with images hidden. The offer, qualification, and action should remain understandable. Second, inspect the accessible alternative and link purpose. Third, view the normal email on desktop, mobile, light mode, and dark mode to confirm the caption is legible and remains beside the right image.\n\nMigma can support this workflow through separate caption and alt-text controls plus its phone preview and Email Preflight. The documented product surface does not prove the final message conforms to WCAG or renders identically in every inbox. The publisher remains responsible for the received artifact.\n\n## Evidence limits\n\nThe dated product claim is limited to Migma’s documented caption feature and existing image controls. W3C sources support the distinction between image purposes and text alternatives. Marketing Wiki did not run an assistive-technology or client-panel test for this article, and this is not legal advice."}