Image SEO Checklist for Developers: Alt Text, Schema, Format and Speed

A technical image SEO checklist gives developers a precise, auditable set of criteria that images must satisfy before going to production — covering the signals Google's crawler uses to discover, understand, rank, and display images in both standard search results and Google Images.

This checklist is structured as six independent audit areas. Each can be checked independently and each has verifiable pass/fail criteria — not general advice. Run it against any page before publish or as part of a post-launch technical audit.

TL;DR — Six Audit Areas at a Glance

• Alt text — every non-decorative image must have a descriptive, keyword-relevant alt attribute; decorative images must have alt=""
• File naming — all image filenames must use lowercase hyphenated descriptive names before upload; never upload IMG_1234.jpg
• Schema markup — hero images, product images, and article featured images should carry ImageObject schema with url, width, height, and caption
• Image sitemaps — every indexable image must appear in an image sitemap or be embedded in the page sitemap via <image:image> tags
• Format and compression — serve WebP or AVIF; compress to under 150KB for content images using quality 80; verify with PageSpeed Insights
• Core Web Vitals — LCP image must have fetchpriority="high", no loading="lazy", and explicit width and height attributes

Checklist Area 1 — Image Alt Text SEO

Alt text is the highest-impact single attribute for image SEO. It is the primary signal Google's crawler uses to understand image content, and the primary signal screen readers use to communicate images to visually impaired users. Both use cases reward the same behaviour: specificity.

Rule 1.1 — Every non-decorative image must have an alt attribute. An <img> element without any alt attribute is an accessibility failure and an SEO gap. The crawler cannot classify the image and screen readers announce "image" with no description.

Rule 1.2 — Decorative images must have an empty alt attribute, not a missing one. An image used purely for visual decoration (dividers, abstract backgrounds, icon flourishes) should carry alt="" — the empty string signals to both screen readers and crawlers that the image carries no informational value and should be skipped.

Rule 1.3 — Descriptive alt text describes what is in the image, not what the page is about. The most common image alt text seo mistake is writing page-level keyword-stuffed alt attributes instead of image-level descriptions.

<!-- Fail: keyword stuffed, not descriptive -->
<img src="tool.webp" alt="free image converter webp avif no upload melotools">

<!-- Fail: too generic -->
<img src="tool.webp" alt="image tool">

<!-- Pass: describes the image accurately -->
<img src="tool.webp" alt="MeloTools drag-and-drop interface converting a JPG file to WebP in a browser window">

Rule 1.4 — Alt text should be under 125 characters. Screen readers typically truncate alt text at 125 characters. Keep descriptions concise and front-load the most important information.

Rule 1.5 — Include the primary keyword naturally in the alt text of the page's hero or most prominent image. For a blog post about image compression, the featured image alt text might read: "developer compressing JPEG images to WebP format using a browser-based image optimization tool." The keyword appears naturally because it accurately describes the image — not because it was inserted.

Rule 1.6 — Run an alt text audit using Screaming Frog free. Screaming Frog crawls up to 500 URLs free and exports all image alt attributes, flagging missing and empty ones. Run monthly and cross-check the export against this checklist.

Checklist Area 2 — Image File Name SEO

The image filename is one of the few on-page image SEO signals that must be set before upload — it cannot be changed after publication without breaking existing links and losing any link equity the URL has accumulated.

Rule 2.1 — Use lowercase hyphenated descriptive filenames. Google's URL guidelines recommend lowercase, hyphen-separated words. Underscores are not treated as word separators by Google — image_seo_checklist.jpg is treated as one long word, not three separate keywords.

Fail: IMG_4823.jpg
Fail: imageseochecklist.jpg
Fail: image_seo_checklist.jpg
Pass: image-seo-checklist-developers.jpg
Pass: jpg-to-webp-converter-browser.webp
Pass: avif-compression-comparison-2026.avif

Rule 2.2 — Describe the image content, not the page topic. "developer-compressing-images-webp.jpg" is better than "melotools-image-compression.jpg" because the former describes what is in the image while the latter describes the business.

Rule 2.3 — Use the target keyword in the filename of the page's featured image. If the page targets "image SEO checklist", the featured image filename should be something like image-seo-checklist-2026.webp — not hero-image.webp.

Rule 2.4 — Keep filenames under 80 characters. Extremely long filenames are truncated in some environments and add unnecessary URL length. Describe accurately but concisely.

Rule 2.5 — Rename files before uploading to WordPress, Notion, or any CMS. Every CMS preserves the original filename in the storage URL. A file uploaded as IMG_4823.jpg will forever live at that URL — you cannot rename it in the CMS without re-uploading and updating every reference.

Checklist Area 3 — Image Schema Markup

Image schema markup tells Google structured information about images that it cannot reliably infer from the <img> element alone — dimensions, captions, licensing, and the relationship between an image and the content it illustrates.

Rule 3.1 — Add ImageObject schema to all hero and featured images.

The minimum required ImageObject schema for a blog post featured image:

{
  "@context": "https://schema.org",
  "@type": "ImageObject",
  "url": "https://melotools.com/images/image-seo-checklist-2026.webp",
  "width": 1200,
  "height": 630,
  "caption": "Developer reviewing an image SEO checklist covering alt text, schema markup, and image sitemaps"
}

Rule 3.2 — Nest ImageObject inside Article schema for blog posts. Google's Article schema supports a image property that accepts an ImageObject. This nested structure tells Google which image represents the article — important for Google Discover and rich results:

{
  "@context": "https://schema.org",
  "@type": "Article",
  "headline": "Image SEO Checklist for Developers",
  "image": {
    "@type": "ImageObject",
    "url": "https://melotools.com/images/image-seo-checklist-2026.webp",
    "width": 1200,
    "height": 630
  },
  "author": {
    "@type": "Person",
    "name": "Dushyant Patel"
  },
  "datePublished": "2026-03-31"
}

Rule 3.3 — Add image property to Product schema for e-commerce product images. Product schema without an image property is incomplete. Google uses the image in shopping results, product carousels, and Google Merchant Center displays.

Rule 3.4 — Validate schema using Google's Rich Results Test. Go to search.google.com/test/rich-results, paste your page URL, and confirm the ImageObject is detected without errors. Warnings about recommended fields are acceptable; errors must be resolved.

Rule 3.5 — Do not block schema images in robots.txt. If the image URL path is blocked in robots.txt, Google cannot crawl the image to verify the schema claim. Confirm that your image CDN or storage bucket is not disallowed.

Checklist Area 4 — Image Sitemap SEO

An image sitemap tells Google about images on your site that it might not discover through standard page crawling — images loaded by JavaScript, images in non-standard HTML attributes, or images on pages with low internal link equity.

Rule 4.1 — Include image tags in your existing page sitemap. The simplest image sitemap implementation adds <image:image> tags inside the existing <url> block for each page. No separate sitemap file required:

<?xml version="1.0" encoding="UTF-8"?>
<urlset xmlns="http://www.sitemaps.org/schemas/sitemap/0.9"
        xmlns:image="http://www.google.com/schemas/sitemap-image/1.1">
  <url>
    <loc>https://melotools.com/guide/image-seo-checklist</loc>
    <image:image>
      <image:loc>https://melotools.com/images/image-seo-checklist-2026.webp</image:loc>
      <image:title>Image SEO Checklist for Developers 2026</image:title>
      <image:caption>Developer using a technical image SEO checklist covering schema markup, sitemaps, and alt text rules</image:caption>
    </image:image>
  </url>
</urlset>

Rule 4.2 — Only include images that are publicly accessible. Images behind authentication, images served only via JavaScript with no <img> fallback, or images on disallowed paths should not appear in the image sitemap. Google cannot crawl them and the sitemap entry is wasted.

Rule 4.3 — Submit the sitemap in Google Search Console. Go to Search Console → Sitemaps → Add a new sitemap. If you added image tags to an existing sitemap that is already submitted, Google will pick up the changes on the next crawl. If you created a dedicated image sitemap, submit it explicitly.

Rule 4.4 — Confirm image indexation in GSC. In Google Search Console → Enhancements → Images — review the image indexation report. URLs with indexation errors or warnings identify specific pages where image schema or sitemap tags have issues. Address errors before warnings.

Rule 4.5 — Do not include images from third-party CDNs you do not control. If your images are served from an unverified third-party domain, Google may not associate them with your site. Use a CDN where you control the domain, or serve images from your own subdomain.

Checklist Area 5 — Format, Compression and Speed

This area references the full compression workflow covered in the complete guide to image compression without quality loss. The checklist items here are the pass/fail criteria only — not the full methodology.

Rule 5.1 — All production images must be in WebP or AVIF format. JPEG and PNG remain acceptable as final fallbacks in <picture> elements. Serving JPEG or PNG as the primary format when WebP or AVIF are supported by the browser is a PageSpeed Insights failure that directly affects Lighthouse scores.

Rule 5.2 — Content images must be under 150KB. Hero images under 200KB. These are the practical targets for most web content at quality 80 in WebP format. Images above these thresholds are flagged by PageSpeed Insights under "Efficiently encode images." Convert to WebP and compress using MeloTools image compressor — processes locally in your browser with no upload.

Rule 5.3 — Run PageSpeed Insights on every key page before launch. Go to pagespeed.web.dev → paste your URL → check specifically: "Serve images in next-gen formats", "Properly size images", "Efficiently encode images", and "Defer offscreen images." Each is a direct image SEO signal.

Rule 5.4 — Convert all JPEG and PNG images to WebP before uploading. Use MeloTools image converter — the JPG to WebP converter and PNG to WebP converter run in the browser. Nothing is uploaded to a server. For maximum compression efficiency, convert key images to AVIF using the JPG to AVIF converter.

Checklist Area 6 — Core Web Vitals Image Rules

Rule 6.1 — The LCP image must have fetchpriority="high". The Largest Contentful Paint element is almost always an image. Adding fetchpriority="high" tells the browser to begin downloading this image as early as possible — before scripts, stylesheets, and other non-critical resources. This is the highest-impact single attribute for improving LCP scores.

<img
  src="hero.webp"
  alt="MeloTools browser-based image optimization tool interface"
  width="1200"
  height="630"
  fetchpriority="high"
>

Rule 6.2 — The LCP image must NOT have loading="lazy". loading="lazy" defers image loading until the element is near the viewport. Applied to the LCP image, it actively delays the most important content on the page. Never apply lazy loading to above-the-fold images.

Rule 6.3 — All images must have explicit width and height attributes. These attributes allow the browser to calculate and reserve the correct space for the image before it loads — preventing layout shift (CLS). The values match the image's intrinsic dimensions in pixels.

Rule 6.4 — All below-the-fold images must have loading="lazy". Every image that is not visible in the initial viewport benefits from lazy loading. This reduces the number of concurrent network requests on page load and improves initial paint performance. The complete lazy loading guide covers the implementation patterns and common mistakes in depth.

Rule 6.5 — Verify CLS with Chrome DevTools Layout Shift regions. Open Chrome DevTools → Rendering tab → enable "Layout Shift Regions." Load the page and observe whether any images cause highlighted red regions to appear as they load. Each highlighted region is a CLS contributor that must be fixed with explicit width/height attributes or aspect-ratio CSS.

The 30-Minute Page Audit Using This Checklist

Run this sequence on any page in under 30 minutes:

1. Open Screaming Frog → crawl the page URL → export Images report → check for missing alt attributes and non-descriptive filenames (Areas 1 and 2) — 5 minutes
2. Open View Source → search for @type → verify ImageObject and Article schema are present and correctly nested (Area 3) — 3 minutes
3. Paste the page URL into Google's Rich Results Test → confirm schema is valid — 2 minutes
4. Open Google Search Console → Sitemaps → verify the page's images appear in the image sitemap — 3 minutes
5. Paste the page URL into PageSpeed Insights → note image-specific opportunity items (Area 5) — 5 minutes
6. Open Chrome DevTools → Network tab → filter by Img → check format (WebP/AVIF) and file sizes → enable Layout Shift Regions → check for CLS (Area 6) — 10 minutes
7. Fix any failures using the rules above — 2 minutes to identify fixes, time to implement varies

For the broader image optimisation workflow covering compression settings, format conversion, and responsive image generation in detail, the SEO image optimization checklist for developers covers the production workflow that feeds into this technical audit.

Frequently Asked Questions

What is the most important item on the image SEO checklist?

Alt text is the highest single-impact item because it is the primary signal Google uses to understand image content and the primary accessibility signal for screen readers. Every non-decorative image must have a descriptive alt attribute that accurately describes what is in the image. Missing or generic alt text is both an SEO gap and an accessibility failure that affects all users relying on assistive technology.

Does image file naming actually affect SEO rankings?

Image filenames contribute a weak but real keyword signal to both Google Image Search rankings and page-level content relevance scoring. More importantly, filenames cannot be changed after publication without breaking URLs and losing accumulated link equity. Naming images correctly before upload costs zero additional time and protects against a permanent SEO gap.

What is ImageObject schema and which images need it?

ImageObject schema is structured data markup that explicitly tells Google the URL, dimensions, and caption of an image. Hero images, article featured images, and product images benefit most — these are the images Google uses in rich results, Google Discover cards, and image carousels. Recipe images and video thumbnails also require ImageObject within their respective schema types.

Do I need a separate image sitemap or can I add images to my existing sitemap?

Adding <image:image> tags inside your existing page sitemap entries is sufficient for most sites. A separate dedicated image sitemap is only beneficial if your site has images that are not accessible through standard page crawling — images loaded by JavaScript without HTML fallbacks, or images on pages with very low internal link equity. In both cases the image tags follow the same XML structure.

How do I check if Google has indexed my images?

In Google Search Console → Enhancements → Images shows indexation status for images in your submitted sitemaps. For spot-checking individual images, use Google Image Search with site:yourdomain.com — images that appear in results are indexed. The Google Rich Results Test also shows whether your ImageObject schema is being parsed correctly, which is a prerequisite for rich result eligibility.