← Back to Blog

Image SEO Is A Five-Dimensional Audit, Not A One-Line Check

Image SEO Is A Five-Dimensional Audit, Not A One-Line Check

Open any SEO audit tool. Search "image SEO." It'll check alt text and call it done.

Google Images ranks on five dimensions. Alt is one. The other four — filename, dimensions, surrounding text context, and image-sitemap inclusion — most SMBs never even hear about. Each is worth a measurable rank delta on its own, and the combined effect is large enough that a properly-audited site can pull 15-30% of its traffic from Google Images while a one-dimensional one pulls almost none.

What the Image Search Traffic Optimizer does

You paste a URL. The tool:

  1. Extracts every <img> tag from the rendered HTML.
  2. Scores each image across 5 dimensions:
    • Filename quality — descriptive multi-word vs camera/CMS gibberish vs hash
    • Alt text — present, descriptive, appropriate length (8-140 chars)
    • Dimensions — has width/height, ≥200px (Google's threshold), ≥1200px preferred
    • Surrounding context — nearby figcaption / heading / paragraph within 300 chars
    • Image-sitemap inclusion — checks /sitemap.xml for the image: namespace
  3. Aggregates a 0-100 score per image and a page-level average.
  4. Flags page-level findings (missing image sitemap, etc.).
  5. Emits an AI prompt with per-image rename suggestions, alt-text drafts, figcaption placements.

The five dimensions, ranked by ranking weight

1. Alt text (highest weight). Google's official position. Missing or "image" / "photo" / "graphic" / a single word is essentially missing. Aim for 8-140 chars describing what's in the image and why it's there.

2. Surrounding context. Google reads the text adjacent to an image to disambiguate. A photo of a roof titled "shingles.jpg" wrapped in an article about hail damage repair is contextualized as roof-damage-from-hail. Same image with no surrounding text is just "shingles."

3. Filename. roof-hail-damage-twin-falls.jpg ranks better than IMG_1234.jpg or b2c45f9a.jpg. Filename appears in the image URL which Google parses. Hyphenate, use 3+ descriptive words.

4. Dimensions. Google Images won't rank images smaller than ~200×200. It prefers 1200px+ on the long edge. Always set width and height attributes (also helps CLS).

5. Image-sitemap inclusion. A separate XML signal that says "here is an image; here is its caption; here is the page it's on." Without it Google relies entirely on crawling. Adding it typically lifts image-search impressions 15-25% within 30 days.

What the score thresholds mean

Per-image score:

  • 80-100 — well-optimized. Each dimension scoring ≥70.
  • 60-80 — workable. Usually one weak dimension dragging the average.
  • 40-60 — multiple weak dimensions. Real opportunity.
  • Under 40 — non-competitive. Image is ballast, not asset.

Page-level average:

  • 70+ — site is image-search-ready.
  • 50-70 — typical SMB site that has paid attention to alt only.
  • Under 50 — image-search invisible. Single biggest improvement vector for sites with photo-heavy verticals (real estate, photography, restaurants, retail, hospitality).

The 30-day image-traffic upgrade path

Week 1: filenames. Rename your top 20 most-trafficked page's images. Before-uploading: descriptive multi-word. Already-uploaded: rename + 301 the old URL or add a canonical via rel="canonical" on the image itself (rare; usually rename + redirect).

Week 2: alt text. Pass the audit's per-image alt drafts back through your CMS. Aim for 8-140 chars, no "image" / "photo" filler, no keyword-stuffing.

Week 3: surrounding context. For your hero images, add a <figure> with a <figcaption> directly under each. The 30-character caption you already have in your head goes right there.

Week 4: image-sitemap. Either generate image-sitemap.xml and reference it from /robots.txt, or extend your main sitemap.xml with the image: namespace. The audit detects either form.

By day 60, image-search impressions should be visibly higher in GSC. By day 90, image-search clicks should follow.

What the audit can't see

The tool reads what comes back from the proxy fetch. Three blind spots:

  • Lazy-loaded images that load via JS only — these don't appear in the initial HTML. Google's renderer mostly catches them, but the audit doesn't.
  • Images served as CSS background — same problem. Background-image is invisible to the audit and to Google Images.
  • Picture / source elements with multiple sizes — the audit reads the first <img> inside a <picture> and that's the most-likely-rendered variant; nuance about srcset is lost.

For lazy-loaded sites, run the audit against the page's pre-rendered HTML if you have an SSR or static-export option.

Related reading

Fact-check notes and sources

This post is informational, not SEO-consulting advice. Mentions of Google Images are nominative fair use. No affiliation is implied.

← Back to Blog

Accessibility Options

Text Size
High Contrast
Reduce Motion
Reading Guide
Link Highlighting
Accessibility Statement

J.A. Watte is committed to ensuring digital accessibility for people with disabilities. This site conforms to WCAG 2.1 and 2.2 Level AA guidelines.

Measures Taken

  • Semantic HTML with proper heading hierarchy
  • ARIA labels and roles for interactive components
  • Color contrast ratios meeting WCAG AA (4.5:1)
  • Full keyboard navigation support
  • Skip navigation link
  • Visible focus indicators (3:1 contrast)
  • 44px minimum touch/click targets
  • Dark/light theme with system preference detection
  • Responsive design for all devices
  • Reduced motion support (CSS + toggle)
  • Text size customization (14px–20px)
  • Print stylesheet

Feedback

Contact: jwatte.com/contact

Full Accessibility StatementPrivacy Policy

Last updated: April 2026