A recent in-depth investigation has revealed significant disparities in how alternative text (alt text) for non-textual content is preserved when Microsoft Word and PowerPoint documents are converted to PDF format using various methods. The study, which involved over 100 distinct test cases across Word and PowerPoint, followed by conversion into three different PDF types—Adobe Acrobat PDFMaker (Windows), Adobe Create PDF cloud service (Mac), and Microsoft’s native Save as PDF feature—has identified crucial insights and potential pitfalls for content creators and accessibility advocates alike. With over 400 individual checks performed, the findings underscore the critical importance of understanding these conversion nuances to ensure digital documents remain accessible to all users.
The process of adding alt text in both Microsoft Word and PowerPoint applications is generally consistent, offering a straightforward user experience. Users can typically access the alt text functionality by selecting an object, such as an image or shape, which then reveals a contextual “Format” tab in the ribbon. Within this tab, a dedicated “Alt Text” option is available. Alternatively, a right-click on the object provides access to the same option through a context menu. Once accessed, a dedicated pane appears, featuring a field for entering descriptive alt text. Crucially, this pane also includes a checkbox to designate content as “decorative,” which is vital for informing assistive technologies that the element does not convey meaningful information and can be skipped. Depending on user settings, the pane may also offer AI-generated alt text suggestions, a feature designed to streamline the process of adding descriptions.
For users on Windows, a particularly useful feature allows for the Alt Text pane to remain open and dockable, enabling users to work on other tools like the Accessibility Assistant without losing their alt text context. This is achieved by accessing a pane options icon, typically located in the upper-right corner of the pane, and selecting the “Move out of Tab” option. This persistent display is a significant workflow enhancement, particularly when undertaking comprehensive accessibility reviews.
The Accessibility Assistant within Microsoft Office applications serves as a powerful tool for identifying and rectifying accessibility issues, including missing or potentially problematic alt text. By navigating to the Review tab and initiating the Accessibility Assistant, users can have the application automatically flag content requiring attention. The assistant groups alt text-related issues into a single category, guiding users through the process of adding or correcting alt text for each identified image. This automated flagging, when combined with manual review, represents the most efficient approach to ensuring robust alt text implementation.
Microsoft Word further enhances this process with an inline prompt feature, providing real-time feedback for accessibility fixes. This feature, accessible via the File tab (or Options) and then Accessibility, can be enabled to display small icons next to elements that require accessibility adjustments. Clicking these icons opens a dialog box where users can directly add or modify alt text, approve AI-generated suggestions, or mark elements as decorative. While this feature is mouse-only, its immediate feedback loop significantly aids in proactive accessibility remediation.
A critical point of attention for users employing the inline prompt feature, or indeed any manual alt text input, is the necessity of explicitly saving these changes. Unlike the persistent Alt Text pane which may offer auto-save functionality, changes made through inline prompts or dialog boxes often require a manual save action. Users must select the “Save” button (or equivalent) within the prompt dialog or the main document save function to ensure their alt text modifications are permanently recorded. Failure to do so can result in lost effort and incomplete accessibility.
The study’s core focus was the integrity of alt text post-conversion to PDF, evaluating three primary methods: Adobe Acrobat PDFMaker (Windows), Adobe’s Create PDF cloud service (Mac), and Microsoft’s built-in Save as PDF export. Historically, Adobe’s PDFMaker has been the benchmark for creating well-structured PDFs, often facilitating easier review within Adobe Acrobat. However, recent developments, including Microsoft’s announced improvements to PDF accessibility in late 2024, have shifted the landscape. Microsoft’s native PDF exporting feature, particularly in its latest iterations, has demonstrated remarkable proficiency in preserving alt text, in some instances surpassing Adobe’s offerings.
Microsoft’s native PDF conversion, whether through the “Save as PDF” function or the dedicated “Export” option, emerged as the clear frontrunner for alt text preservation. The analysis indicates that this method maintains alternative text for nearly all image types, with emoji being the sole exception identified as failing to retain their alt text during the conversion process. This represents a significant leap forward for Microsoft’s commitment to accessible document creation.
However, the Microsoft conversion process is not without its nuances. While the alt text itself is largely preserved, the inclusion of extraneous information within the alt text string has been observed. For elements other than standard pictures, the conversion process often appends the original image type to the defined alt text. For instance, a shape described with alt text "Two-way communication" might be rendered in the PDF as "Two-way communication (Arrow: Left-Right)," with the "(Arrow: Left-Right)" component being an automatic addition reflecting the shape’s type. While this can occasionally be helpful, it frequently leads to repetitive or even confusing descriptions, especially given the vast array of named shapes available in Office applications—estimated to be around 150 distinct types.
In contrast, Adobe’s conversion tools, the PDFMaker on Windows and the Create PDF cloud service on Mac, showed a more varied performance, with a common deficiency being the loss or corruption of alt text during the conversion. This inconsistency means that relying solely on Adobe’s tools for alt text preservation can be a gamble, potentially requiring significant post-conversion remediation.
A critical workaround identified for enhancing alt text preservation, particularly when using Adobe’s conversion tools, is to convert non-picture objects into actual Pictures within Word or PowerPoint. A “Picture” in Office refers to a specific image type, encompassing photographs and most images pasted from external sources. The defining characteristic of a Picture is that selecting it triggers the “Picture Format” tab in the ribbon. When an object is designated as a Picture, its alternative text is consistently retained across both PDFMaker and the Adobe cloud service conversions.
The method to achieve this involves copying or cutting the object and then pasting it back into the document as a Picture. On Windows, this is accomplished via the Paste Options menu, typically found under the Home tab, by selecting “Paste Special” and then choosing “Picture (Enhanced Metafile)” or a similar picture format. On Mac, the process involves right-clicking (or accessing the Edit menu) and selecting “Paste Special” followed by one of the “Picture” file types.
However, this “Picture” conversion strategy has limitations. Certain object types, such as audio and video elements, cannot be effectively pasted as Pictures without losing their core functionality, rendering this workaround unsuitable for them. Furthermore, once an object is pasted as a Picture, it becomes immutable; it can no longer be edited in its original form. Therefore, users are strongly advised to create a duplicate version of their document before applying this conversion as a final step before PDF creation.
A particularly crucial caveat associated with converting objects to Pictures is that the original alt text is lost during this paste operation. To mitigate this, users must first copy the image (not cut it), paste it as a Picture, then manually copy and paste the original alt text into the new Picture object’s alt text field, and finally delete the original, un-converted object. This multi-step process, while effective, adds considerable time and complexity to the workflow.
The investigation also delved into specific “gotchas” for various image types, revealing a multitude of less prominent alt text bugs, especially when using Adobe’s Acrobat tab for PDF creation. While a comprehensive list is beyond the scope of this report, several notable quirks and pitfalls were highlighted. These include issues with charts and graphs, SmartArt graphics, and complex diagrams, all of which can suffer from inconsistent alt text rendering or complete loss of descriptive information depending on the conversion method and the specific object. For instance, complex charts might have their data labels misinterpreted or their overall narrative lost, while SmartArt graphics may revert to generic descriptions that fail to convey the intended message.
In essence, when dealing with alt text in Microsoft Word and PowerPoint, several core principles should guide content creators:
-
Prioritize Native PDF Export: For optimal alt text preservation, utilize Microsoft’s built-in PDF export features (“Save as PDF” or “Export”) whenever possible. This method has shown the most robust results, significantly reducing the risk of alt text loss.
-
Understand Conversion Tool Differences: Be aware that Adobe’s PDF creation tools, while historically robust for overall PDF structure, can be unreliable for alt text integrity. If Adobe tools are necessary, employ workarounds like converting objects to Pictures, but be mindful of the associated complexities and data loss risks.
-
Leverage Accessibility Assistants: Regularly employ the Accessibility Assistant within Office applications to identify and rectify missing or inadequate alt text. This proactive approach is fundamental to creating accessible documents from the outset.
-
Test and Verify: Always test your converted PDFs using accessibility checkers and screen readers to confirm that alt text is present, accurate, and effectively conveys the intended information for all users, including those relying on assistive technologies. The findings from this study underscore that digital accessibility is not solely about authoring content but also about ensuring its faithful representation across different formats and platforms. The continued evolution of software and conversion tools necessitates ongoing vigilance and a deep understanding of their implications for inclusivity.
