Back to TechSheets
Beyond the Green Checkmark: Crafting Truly Accessible Alt Text for Web

Beyond the Green Checkmark: Crafting Truly Accessible Alt Text for Web

Thanga MariappanSenior Architect
9 min read
Aug 26, 2026

Beyond the Green Checkmark: Crafting Truly Accessible Alt Text for Web

As a Senior Front-End Architect, few things are as simultaneously reassuring and misleading as the ubiquitous 'green checkmark' from an automated accessibility scanner. We’ve all seen it: a tool flags an image as having alt text, breathes a sigh of relief, and moves on. However, as the recent GitHub blog post aptly highlighted, "Your alt text passes automated checks. That doesn't mean it's any good." This isn't just a nuance; it's a fundamental gap in our understanding and implementation of inclusive design.

Today, Wednesday, August 26, 2026, with the rapid evolution of web technologies and an increasing focus on user experience, the quality of our accessibility efforts — particularly something as foundational as alt text — has never been more critical. Automated tools are essential, but they are just the first line of defense. True accessibility for images requires a deep, human understanding of context, purpose, and the diverse needs of our users.

This deep-dive isn't about if you should use alt text, but how to write alt text that genuinely elevates the user experience for everyone, especially those relying on assistive technologies. We'll go beyond basic compliance to explore the nuances, pitfalls, and best practices for creating truly meaningful image descriptions.

The Illusion of the Green Checkmark: Why Automated Tools Fall Short

Automated accessibility checkers are invaluable for catching obvious violations: missing alt attributes, insufficient contrast, or incorrect ARIA roles. For alt text, they are designed to answer a simple question: "Is an alt attribute present?" They typically cannot, and should not be expected to, answer: "Is this alt text useful and accurate in context?"

Consider these examples that would likely pass an automated check but provide little to no value:

  • <img src="hero-banner.jpg" alt="hero-banner">
  • <img src="product-image-xyz.png" alt="image">
  • <img src="chart-data-2026.svg" alt="chart">
  • <img src="decorative-swirl.png" alt="a decorative swirl image">

While an alt attribute exists, the content itself is either redundant (repeating the filename), generic, or utterly devoid of the descriptive information a screen reader user needs. Automated tools confirm presence; human judgment evaluates utility and relevance. This disparity creates a false sense of security, leading developers to believe their work is accessible when, in fact, it creates significant barriers for many users.

The Unspoken Contract: Alt Text's True Mission

At its core, alt text serves as a direct textual replacement for an image. Its mission is threefold:

  1. For Screen Reader Users: To convey the equivalent information or function of the image, allowing users who cannot see the image to understand its purpose and content within the surrounding context.
  2. For Broken Images: When an image fails to load, the alt text is displayed, providing a textual fallback that explains what should have been there.
  3. For Search Engine Optimization (SEO): Search engines use alt text to understand image content, contributing to better image search rankings and overall page relevance. However, this is a secondary benefit; accessibility and user experience must always take precedence.

The crucial insight is that alt text is not merely a label; it's a surrogate. It must offer an equivalent experience to seeing the image, allowing the user to grasp its meaning without visual input.

The Alt Text Matrix: A Categorical Approach to Crafting Contextual Descriptions

Effective alt text isn't a one-size-fits-all solution. It depends entirely on the image's purpose and content. Let's break down common image categories and their optimal alt text strategies.

1. Decorative Images: The Empty Alt Attribute

These are images that serve purely aesthetic purposes, providing no information essential to understanding the page content. Think background patterns, generic dividers, or purely stylistic elements.

Strategy: Use an empty alt attribute (alt=""). This tells screen readers to ignore the image completely, preventing unnecessary clutter in the auditory experience.

Good Example:

<img src="gradient-bg.svg" alt="" aria-hidden="true">

Using aria-hidden="true" as well can further ensure decorative images are ignored by assistive technologies, particularly when they might have complex SVG structures that could otherwise be announced.

Bad Example:

<img src="spacer.gif" alt="a transparent spacer gif">

This unnecessarily announces a non-informative image.

2. Informative Images: Direct and Concise Descriptions

Informative images convey essential information that contributes to the page's meaning. This category includes product photos, article illustrations, simple charts, or screenshots.

Strategy: Provide a concise, descriptive phrase that accurately summarizes the image's content and its relevance to the surrounding text. Avoid redundancy.

Good Example (Product Image):

<img src="ergonomic-keyboard.jpg" alt="Close-up view of an ergonomic mechanical keyboard with a split layout and wrist rests.">

Bad Example (Product Image):

<img src="ergonomic-keyboard.jpg" alt="keyboard image for sale">

This provides generic, marketing-speak that doesn't describe the product.

Good Example (Simple Chart):

<img src="quarterly-sales.png" alt="Bar chart showing Q3 sales increase by 15% from Q2, reaching 1.2 million units.">

Bad Example (Simple Chart):

<img src="quarterly-sales.png" alt="chart of sales data">

This misses the critical data takeaway the visual provides.

3. Functional Images: Conveying Action

Functional images are interactive elements like buttons, links, or controls that are represented graphically (e.g., a magnifying glass icon for search, a cog icon for settings).

Strategy: The alt text should describe the action or purpose of the element, not just what it visually depicts.

Good Example (Search Button):

<button type="submit">
  <img src="magnifying-glass.svg" alt="Search">
</button>

Bad Example (Search Button):

<button type="submit">
  <img src="magnifying-glass.svg" alt="magnifying glass icon">
</button>

Announcing "magnifying glass icon" doesn't tell the user what pressing it does.

4. Complex Images: Charts, Graphs, and Infographics

These are the trickiest. Complex images contain substantial data or information that cannot be adequately conveyed in a short alt attribute.

Strategy:

  • Short alt with a Long Description: Provide a concise alt attribute that identifies the image (e.g., alt="Quarterly Revenue Trends Graph") and then link to or embed a more detailed textual description elsewhere on the page.
  • Adjacent Text: Integrate the key information from the image directly into the surrounding paragraph text, making the image supplementary.
  • Data Tables: For complex charts, provide the raw data in an accessible HTML table that can be read by screen readers.
  • aria-describedby: Use aria-describedby to associate the image with an element containing a more detailed description.

Conceptual Example (Long Description):

<img src="complex-data-visual.svg" alt="Annual stock performance chart for Acme Corp, 2020-2025. See full details below.">
<div id="stock-details">
  <h3>Annual Stock Performance Data</h3>
  <p>Detailed breakdown of Acme Corp stock values from 2020 to 2025...</p>
  <!-- Or a data table here -->
</div>

Conceptual Example (aria-describedby):

<img src="flowchart.svg" alt="System architecture flowchart" aria-describedby="flowchart-desc">
<p id="flowchart-desc">This flowchart illustrates the data flow from client-side application through API gateways, microservices, and database layers, highlighting authentication and authorization steps at each stage.</p>

For truly complex visuals, a combination of these approaches might be necessary to ensure no user is left behind.

Common Anti-Patterns and How to Sidestep Them

Beyond just missing context, certain alt text patterns actively detract from accessibility:

  • Keyword Stuffing: Don't cram irrelevant keywords for SEO. This makes the alt text unintelligible for screen readers and offers minimal SEO benefit while risking penalties.
  • Redundancy: Avoid starting alt text with "Image of..." or "Photo of...". Screen readers already announce that an image is present. Just get straight to the description.
  • Overly Verbose Descriptions (when not complex): For simple informative images, keep it concise. A single sentence or phrase is often enough. Don't write a paragraph for a simple icon.
  • Generic Phrases: "Click here," "Learn More" are problematic if the alt text doesn't provide additional context. Combine alt text with surrounding text for clarity.

Beyond Automation: Cultivating a Culture of Human Review

Since automated tools have their limits, integrating human review is paramount. As architects, we must foster environments where accessibility is a shared responsibility, not an afterthought.

  • Developer Checklists: Incorporate specific prompts into your definition of done, reminding developers to manually review alt text for context and clarity.
  • Peer Reviews: During code reviews, make alt text quality a discussion point. Another pair of eyes can often spot ambiguities or omissions.
  • Manual Testing with Screen Readers: Encourage developers to install and experiment with screen readers (NVDA, JAWS, VoiceOver, Narrator). Experiencing the web through assistive technology is the most powerful way to build empathy and improve quality.
  • Dedicated Accessibility Audits: Periodically engage accessibility specialists or conduct internal audits to get a comprehensive view of your product's accessibility posture.

The Bigger Picture: A Holistic Approach to Image Accessibility

While alt text is fundamental, remember it's one piece of a larger puzzle. Consider these additional factors for comprehensive image accessibility:

  • <figure> and <figcaption>: For images that are referenced in the main content and require a caption, use these semantic HTML5 elements. The figcaption provides a visible, accessible description for everyone.

    <figure>
      <img src="company-hq.jpg" alt="Exterior view of TechSheet's modern office building with large glass windows.">
      <figcaption>TechSheet's new headquarters in downtown San Francisco.</figcaption>
    </figure>
    
  • ARIA Attributes: Beyond aria-hidden and aria-describedby, explore aria-labelledby for more complex associations, especially with custom components that use images as labels.

  • Responsive Images (srcset, sizes): Ensure that alt text is consistent across different resolutions and image versions served by srcset and sizes attributes. The content of the image, and thus its alternative, should remain consistent regardless of its visual dimensions.

Key Takeaways

  • Automated checks are a start, not the finish. They validate presence, not meaning.
  • Context is king. alt text must match the image's purpose (decorative, informative, functional, complex).
  • Prioritize user experience. alt text is for people first, then search engines.
  • Human review is indispensable. Integrate manual checks, peer reviews, and screen reader testing into your workflow.
  • Think holistically. alt text is part of a broader image accessibility strategy.

What You Should Do Today

  1. Audit Your Core Images: Pick a critical page on your current project and manually review the alt text for every image. Ask: Does it genuinely convey meaning to someone who can't see it?
  2. Educate Your Team: Share this post with your fellow front-end developers and designers. Spark a conversation about improving alt text quality.
  3. Experiment with a Screen Reader: If you haven't recently, spend 15 minutes navigating a web page using NVDA (Windows) or VoiceOver (macOS). Pay close attention to how images are announced.
  4. Update Your PR Template: Add a checklist item requiring alt text review for all new or modified images.