Image SEO: What Most Sites Skip

Images are the most commonly neglected SEO asset on most sites. They're also one of the most impactful: they affect Core Web Vitals scores, accessibility compliance, Google Image Search visibility, and how well AI crawlers understand your content. Most sites treat images as design elements and leave all of that on the table.

A technically sound image implementation isn't complicated. It's a short list of requirements that most developers didn't learn as SEO fundamentals, so they're absent from most codebases until someone explicitly adds them.

Alt text: what it does and how to write it

Alt text (the alt attribute on an img element) serves two functions: it describes the image to screen readers for users who can't see it, and it describes the image to search engine crawlers that can't see it either. Both audiences benefit from the same thing: a concise, specific description of what the image shows.

Good alt text: "Line graph showing organic traffic growth from 1,200 to 8,400 monthly sessions over 14 months." Bad alt text: "graph" or "image001.jpg" or a keyword-stuffed phrase with no descriptive value. Decorative images (dividers, background patterns, icons with adjacent text labels) should use empty alt text (alt="") to signal to screen readers that there's nothing to announce.

This is why accessibility and SEO are the same work: good alt text serves both users with visual impairments and search crawlers simultaneously. Sites that implement alt text for accessibility compliance get the SEO benefit automatically.

File format and compression

WebP is the current standard for web images. It produces 25–35% smaller files than JPEG at equivalent quality, and 60–80% smaller than PNG for photographic content. Smaller files mean faster load times, which affect Core Web Vitals — specifically Largest Contentful Paint (LCP), which Google measures against the largest image in the viewport.

AVIF is newer and produces even smaller files, but browser support is slightly less universal. For broad compatibility in 2026, WebP is the practical choice. Serving AVIF with WebP fallback via the picture element is the advanced implementation for sites with the pipeline to support it.

File names

File names are a weak but real signal. An image file named seo-audit-checklist-screenshot.webp provides more context than IMG_4823.jpg or screenshot-2024-03-12.png. Crawlers read file names; so does Google Image Search. Descriptive, hyphenated file names cost nothing and contribute marginally to topical relevance. Name the file what the image actually shows.

Width, height, and CLS

Cumulative Layout Shift (CLS) — one of Google's Core Web Vitals — is caused by content moving on the page as it loads. The most common cause: images without explicit width and height attributes. When a browser loads a page and encounters an image without dimensions, it doesn't know how much space to reserve. When the image loads, everything below it shifts. That shift is CLS.

The fix is simple: always include width and height attributes on every img element, set to the image's actual or intended display dimensions. The browser uses these to reserve the correct space before the image loads. No layout shift, no CLS penalty.

Loading strategy

Images above the fold — visible without scrolling on load — should use loading="eager" (or no loading attribute, since eager is the default) and, for the largest above-fold image, fetchpriority="high". This tells the browser to prioritize fetching this image early, improving LCP.

Images below the fold should use loading="lazy". This defers image loading until the user scrolls near the image, reducing initial page load time without any visible impact on the user experience. The combination of eager loading for above-fold + lazy loading for below-fold is the standard implementation for optimized LCP and reduced initial page weight.

Image schema for structured data

The ImageObject type in schema.org allows you to associate images explicitly with a page's structured data. For Article schema, including an image property pointing to your feature image with width, height, and caption is a requirement for eligibility in Google News and a strong signal for Knowledge Panel image selection. For product pages, product images with full schema implementation are required for Rich Results. The technical SEO audit should include a check that feature images are referenced in Article schema, not just in the HTML.

---

Related reading

• The Problem with Generic Tech Recommendations
• If the Recommendation Came Before the Questions, It's Not Advisory

Frequently asked questions about image seo

Do images affect Core Web Vitals?

Yes, significantly. The Largest Contentful Paint (LCP) element is the largest image or text block visible in the viewport — and on most pages, it's an image. LCP is one of Google's three Core Web Vitals and a confirmed ranking signal. Slow-loading hero images, unoptimized file sizes, and missing fetchpriority="high" on the LCP element are among the most common CWV failures.

Should I use descriptive or keyword-optimized alt text?

Descriptive. Keyword-optimized alt text that doesn't accurately describe the image is a form of over-optimization that Google's image understanding has become good at detecting. Write alt text that describes the image as you would explain it to someone who can't see it. Natural descriptions that happen to include relevant terms perform better than keyword-stuffed alt text that doesn't describe the actual image.

What's the ideal image size for blog posts?

For feature and OG social images, 1200×630px is the standard — it fills Twitter/X, LinkedIn, and Facebook card formats without cropping, and is the recommended size for Open Graph previews. For in-article images, match the display width (typically 700–900px for blog columns) at 2x resolution for high-density displays, then compress to WebP. A well-compressed WebP at 1200px wide should be under 150KB; under 80KB is achievable for most photography.

Does Google Image Search drive meaningful traffic?

It depends heavily on vertical. For visual-heavy industries — interior design, food, fashion, real estate, products — image search is a significant traffic source. For business and professional services content, less so. Either way, image SEO done correctly doesn't cost meaningfully more than image SEO done poorly — so the question is less "is it worth it" and more "why would you leave it done incorrectly."