Google Image Search accounts for over 20% of all searches on Google (SparkToro, "Search Engine Referral Study," 2024). Yet most businesses ignore image optimization entirely — they upload images with cryptic file names, without alt text, in the wrong format, and far too large. That is wasted potential. Proper image SEO not only improves your images' visibility in Google Images — it also improves the page's overall rankings through faster load times, better user experience, and stronger content signals.
Why image SEO matters
Images are not just decoration on a webpage — they are a standalone search channel. Google Image Search is the second most-used search function after regular Google search. For industries like e-commerce, interior design, food, travel, and fashion, image search is a massive traffic source that many competitors overlook.
In my experience, good image optimization can increase a page's organic traffic by 15-30%. That is not just image search traffic — optimized images improve the page's Core Web Vitals score, which is a ranking factor for all organic search traffic.
Additionally, Google increasingly uses images in AI Overviews and visual SERP elements. Well-optimized images have a greater chance of being shown in these prominent positions. See our SEO guide for the broader context.
File naming conventions that describe the content
The first and simplest step: give your image files descriptive names before uploading them. Google uses the file name as a signal to understand the image's content. "IMG_4521.jpg" tells nothing. "black-leather-handbag-womens.jpg" tells Google exactly what the image shows.
Best practices for file names: Use hyphens to separate words (not underscores). Keep it short but descriptive — 3-6 words is ideal. Include the primary keyword if it fits naturally. Avoid filler words like "image-of" or "photo-of." Use lowercase and avoid special characters.
For websites with many images (e.g., e-commerce stores), I recommend establishing a naming convention early: [product-category]-[product]-[attribute]-[number].jpg. Example: "shoes-loafers-brown-leather-01.jpg." Consistency makes it easier to maintain and scale.
Alt text best practices: accessibility and SEO in one
Alt text (alternative text) is the single most important factor for image SEO. It serves two purposes: it makes the image accessible to visually impaired users who rely on screen readers, and it gives search engines a text description of the image's content.
Write alt text as if you were describing the image to someone who cannot see it. Be specific and concrete. Poor: alt="bag." Better: alt="black leather handbag with gold buckles by Mulberry." Best: alt="black leather handbag with two gold buckles and detachable shoulder strap by Mulberry, front view." Include relevant keywords naturally, but never list keywords — that is keyword stuffing and can result in penalties.
Exceptions for alt text: Decorative images (icons, background patterns) should have empty alt text (alt="") so screen readers skip them. Images already described in surrounding text need shorter alt text. Button and link images must have alt text that describes the link destination, not the image. For more on accessibility, see the glossary for definitions of alt text and related terms.
Image compression and modern formats
Image file size is one of the biggest factors for page speed. An uncompressed JPEG image can be 2-5 MB — that slows the page dramatically on mobile. With proper compression, you can reduce file size by 60-80% without visible quality loss.
Modern image formats to consider: WebP delivers 25-34% smaller file size than JPEG at equivalent quality (Google, "WebP Compression Study," 2023). AVIF delivers an additional 20% savings over WebP, but has lower browser support. JPEG XL is promising but not yet supported by Chrome. My recommendation for 2026: Use WebP as the default format with JPEG fallback. Consider AVIF for browsers that support it.
Compression tools: Squoosh (free, web-based), ShortPixel (WordPress plugin), Cloudinary (CDN with automatic optimization), ImageOptim (Mac). For WordPress sites, I recommend ShortPixel or Imagify, which automatically convert and compress all images on upload. Read more about performance optimization in our web performance guide.
Responsive images with srcset and sizes
Responsive images mean you serve different image sizes to different devices. A mobile user does not need a 2000px wide image designed for desktop — it wastes bandwidth and slows load time.
The HTML srcset attribute lets the browser choose the right image size based on the screen's resolution and size. The sizes attribute tells the browser how large the image will be displayed. Together, they enable the browser to load the smallest image that provides acceptable quality.
Modern CMS and frameworks (Next.js Image component, WordPress with a proper theme, Shopify) generate responsive images automatically. Check that your CMS actually does this — many older WordPress themes lack proper srcset implementation. For Core Web Vitals impact, see the CWV guide.
Lazy loading: Load images when they are needed
Lazy loading means images below the fold (not visible when the page first loads) are not loaded until the user scrolls down to them. This improves initial load dramatically, especially on pages with many images.
In 2026, lazy loading is built into browsers via the loading="lazy" attribute on img elements. You do not need JavaScript libraries for basic lazy loading anymore. Important: Do not lazy-load images above the fold (the topmost visible area). These should load immediately because they affect LCP (Largest Contentful Paint). Only images below the fold should have loading="lazy."
For images above the fold, you should instead use fetchpriority="high" to tell the browser that this image is important and should be prioritized. The combination of eager loading above the fold and lazy loading below the fold delivers optimal performance. See the glossary for definitions of lazy loading and WebP. Struggling with image speed and Core Web Vitals? We are happy to help.
Image sitemaps and structured data
An image sitemap helps Google discover images that might not otherwise be indexed — especially images loaded via JavaScript, images in carousels, or images on pages that are crawled infrequently.
You can either create a separate image sitemap or add image tags to the existing XML sitemap. Google supports the image:image element in sitemaps, where you can specify image URL, title, caption, geolocation, and license.
For structured data, you can use ImageObject schema to give Google richer information about your images. This is particularly useful for product images (combine with Product schema), recipe images (combine with Recipe schema), and editorial images. For a complete structured data guide, see the schema markup guide.
Image CDN benefits
An image CDN like Cloudinary, imgix, or Bunny Optimizer handles image optimization automatically. You upload the original, and the CDN serves optimized versions tailored to the user's device, browser, and connection. It is the most efficient solution for websites with many images.
Benefits of an image CDN: Automatic format conversion (WebP/AVIF based on browser). Automatic resizing based on device. Automatic compression. Global distribution from the nearest server. URL-based image manipulation (crop, resize, filters).
For European businesses, I recommend choosing a CDN with servers in Europe for optimal speed. Cloudflare (with Polish and Mirage enabled) is a good starting point because it is integrated into many hosting solutions and has servers across Europe.
Original images vs stock photos for SEO
Here is an insight many overlook: Google prefers original images. Stock photos used on thousands of other websites provide minimal SEO value because Google already knows them and recognizes they are not unique to your content.
In my experience, pages with original images rank better in Google Image Search than identical content with stock photos. That does not mean you need to photograph everything yourself — but it does mean you should prioritize original images for your most important pages: product pages, service pages, and key content.
Practical tips for getting more original images without a large budget: Take screenshots of tools and processes you discuss. Create simple diagrams and infographics in Canva or Figma. Use AI image generation (DALL-E, Midjourney) for conceptual illustrations. Photograph with your phone — authentic images are more valuable than perfect stock photos.
Image search ranking factors
Google has shared some of the most important ranking factors for image search (Google Search Central, "Image SEO Best Practices," 2025). The key ones are: relevance (the image matches the search term), context (the text surrounding the image), quality (resolution and originality), page placement (images higher on the page are prioritized), and the page's overall authority.
A factor many overlook: image caption (figcaption in HTML). Google heavily weighs the text directly below the image to understand the image's content. Always include a descriptive figcaption for important images — it is a strong ranking signal.
For e-commerce, product image quality is particularly important. Google Shopping requires white backgrounds and clear product images. Images with text overlays rank worse in general image search. Page speed's impact on image ranking is significant — slow pages get fewer images indexed. See our e-commerce SEO guide and SEO services for more.
Sources
SparkToro, "Search Engine Referral Study," 2024. Google, "WebP Compression Study," 2023. Google Search Central, "Image SEO Best Practices," 2025. HTTP Archive, "State of Images," 2025. Web.dev, "Fast Load Times — Optimizing Images," 2025.
