John Mueller from Google explained how to use alt text for SEO effectively and why there may be better approaches than relying on AI to scale it.

Google’s John Mueller answers to a question on Reddit about image alt text and SEO, providing a detailed explanation of the importance of alt text and why using AI to generate it automatically might fall short.

Alt Text

A user asked whether alt text is still relevant to search optimization. To understand the role of alt text in SEO, it’s helpful to explore why it exists in the first place.

HTML elements serve as the building blocks of a web page, which can be enhanced with attributes that add additional information. For instance, the <A> tag is an anchor element that becomes a clickable link when paired with the “href” attribute. Similarly, the <A> tag can also have a “nofollow” attribute. W3C, which establishes HTML standards, offers an insightful explanation:

For complex images such as graphs or diagrams, two-part alternative text might be necessary. The alt text briefly describes the image, while the surrounding text offers a more detailed explanation of the content. Another approach is to link to a longer description.

The “alt” in alt text stands for “alternate” or “alternative” content. Specifically, it refers to an HTML attribute of the image element. Its primary purpose is to provide alternate information about an image for users who cannot view it, ensuring that the content still makes sense in the web page context.

While web content is often thought of as primarily text, images are equally important when they carry “informative value” that aids the visitor’s understanding of the page topic.

Question About Alt Text and SEO

The question stems from a user’s understanding that Google now employs advanced algorithms to “see” and interpret images. Given these advancements in computer vision and image recognition, the user wonders whether alt text has become redundant or unnecessary. The specific question asked was:

“Image alt text in the era of computer vision:

Are image alt texts still relevant for SEO with all the computer vision and image recognition advancements? Is there any information about Google or other search engines using machine learning models to crawl images, rather than relying on user-provided alt texts?”

The user is essentially questioning whether alt text is still necessary for SEO, or if modern image recognition technology makes it superfluous.

Context Is Key in SEO

The question raised is both reasonable and valid, considering the advancements in Google’s ability to “see” and understand images through AI. However, what may be missing is the understanding of the context in which Google applies these AI-driven image recognition capabilities. Google’s documentation highlights the use of AI vision in tools like Google Lens, Google Translate, and other specific search surfaces. However, it doesn’t explicitly mention the use of these capabilities within regular Google search results (thanks to @schachin for pointing out that documentation!).

John Mueller’s response helps fill in this gap by explaining that the surrounding text on a page provides crucial context for what an image represents. AI image recognition alone doesn’t capture the full meaning or relevance of the image within the context of the webpage.

Here’s Mueller’s explanation:

**“For image search, there’s the context that comes from the page + image combination that matters.

A photo of a beach might be a relaxing poster, it might be the beach from a hotel, or it could be the site of a chemical spill. Just knowing that the image is of a beach doesn’t really give sufficient background information to display it in imagesearch appropriately. A lot of the meaning comes from the page, and the alt text is unique in that it directly connects the image to the page with context.

Unless your site is a photo agency, traffic for ‘photo of a beach’ isn’t going to be very useful. But for a hotel, having ‘hotelwith beach in X’ can be highly relevant. Much of that context comes from the rest of the page, but the alt attribute is a unique opportunity to provide clear context.

(And with that … if you use AI to create alt texts based on the image file, and get ‘photo of a beach’ as the alt text, you’re not getting the most out of the alt text for either users or search engines.)”**

Mueller’s response perfectly illustrates why alt text remains crucial for SEO. Alt text doesn’t just describe what’s in an image; it shows how the image connects directly to the content of the webpage. This contextual relevance is key for search engines to understand how an image fits within the broader topic of the page.

Why AI Fails for Alt Text

John Mueller highlights a significant shortcoming when using AI to scale alt text: while AI can describe an image, it often fails to label the image within the context of the surrounding content. Alt text should not just describe what the imageshows, but communicate its relevance to the overall page content. This approach aligns with the best practices recommended by the W3C, and is essential for both SEO and accessibility purposes.

Mueller emphasizes that alt text should provide an informative description in relation to the context of the page, not just a generic label. For example, AI might label an image as “photo of a beach,” but that doesn’t capture whether the beach is part of a hotel, a vacation spot, or a disaster site.

For both SEO and accessibility, the goal is to offer a meaningful description that enhances the user’s understanding of the page. If you’re finding this confusing or overwhelming, consider checking out our monthly SEO packages and let the experts handle it for you!