
Alt text does not by itself determine where a reader encounters an image. Check whether the photo is floating rather than inline with the document text. Microsoft's screen-reader guidance identifies access differences for those layouts and documents the Selection pane for locating objects. Review the actual reading path before rewriting an already useful description.
Key Takeaways
- Image description and image placement are separate checks.
- Consider inline placement when the image belongs in the document's text sequence.
- Verify the delivered Word file and any PDF edition separately.
Describe where the image belongs
Imagine a proposed installation guide containing a paragraph about a connector, an annotated photo of that connector and a follow-up instruction. The photo has alt text, but continuous reading reaches the follow-up without encountering the image where the editor expected it.
Record the intended sequence in plain language. The goal is for the reader to understand the relevant image before acting on the next instruction. It is not enough to find the description somewhere in an object inventory if ordinary reading places it out of context.
Inspect the layout and object path
Select the photo and inspect its layout mode. Microsoft's guidance on inserting pictures includes placing the picture inline with text and adding alt text. Work on a copy when changing layout, since the page composition may move.
For the documented Word for Windows route, open Home > Select > Selection Pane, or press Alt+F10. Microsoft's screen-reader walkthrough also uses Alt+H, then S, L, P. Use arrow keys to move through the list and Spacebar to select the intended object. These are Windows instructions, not a promise of identical Mac, web or mobile shortcuts. Finding an object in this pane confirms an access route; it does not establish that continuous reading encounters it at the correct paragraph. In the Windows instructions, the pane lists objects on the current page. Do not interpret an empty list on another page as proof that the image was deleted from the whole document. Platform-specific controls and assistive behavior should be checked against the actual setup.
Review the description's purpose
After placement is suitable, read the alternative alongside the nearby instruction. It should convey the image's relevant meaning without inventing hidden details. The alternative text versus generation prompts matters for artwork that originated in an AI workflow: a production prompt is not automatically a useful document description.
Do not label a task-critical connector photo decorative to make a warning disappear. If the explanation needs visual relationships beyond a short alternative, provide accurate surrounding text that supports the reader's action.
Accept the complete reading sequence
For the hypothetical guide, acceptance means the intended reader encounters the connector explanation, photo meaning and next instruction in a coherent order. Also inspect the visible layout after the change. A repaired reading path should not cover a caption or separate it from the wrong image.
If a PDF is delivered, choose an appropriate Office export route and inspect the figure-caption reading association in that artifact. This guide does not certify an unidentified document or every screen reader. Keep the exact reviewed editions identifiable for later revisions.
Related tasks
- Should I Use My Generation Prompt as Alt Text for an AI Image?
- Why Does a Screen Reader Read My Photo-Catalog Caption Before the Wrong Image?
- Why Did My Word or PowerPoint PDF Lose Alt Text and Document Structure?
Sources and scope
- Use a screen reader to select and read text boxes and images in Word, Microsoft Support: Microsoft identifies reading-path differences between floating and inline objects and documents the Selection pane for locating images.
- Use a screen reader to insert a picture or image in Word, Microsoft Support: The documented image workflow combines inline placement with an appropriate alternative description.
- Use the Selection pane to manage objects in documents: Platform-scoped Selection pane menu and keyboard entry.
Based on Microsoft Support reviewed October 31, 2026. Reading behavior depends on the actual Word and assistive-technology setup, which was not tested here.
Prepared by Imagild Editorial, the company editorial team responsible for these guides. AI assisted research and drafting. This is documentation-based guidance with proposed examples, not a claim of hands-on testing or an independent product review. Report a correction.
Cover: AI-generated conceptual editorial illustration; not a real application screenshot, observed output or before-and-after test.