Check the image resource behind a visually complete article

Decide which images carry editorial meaning

Google’s image guidance, checked 7 November 2026, describes discovery through HTML image elements and a src fallback for responsive image implementations; CSS backgrounds are not indexed as images. These are discovery conditions, not a promise of image-search placement. The acceptance exercise concerns public editorial images that the publisher has already chosen to make discoverable.

Start with the page’s visual inventory. Identify the lead photograph, explanatory diagram and gallery items whose meaning belongs to the story. Separate them from ornamental gradients, textures and layout flourishes. A decorative background does not need to become a searchable illustration merely because it exists. For each editorial image, record the approved asset identity and its relationship to the nearby text so a technically valid but unrelated file cannot pass.

Trace the rendered picture back to a resource

Create one manifest row per meaningful image: article URL, content role, element carrying the image, fallback asset address and responsive candidates. Ask the implementation owner to inspect the generated page rather than copying a CMS upload reference. A media-library identifier is not evidence that the published template exposes the correct resource.

Compare the picture shown on a narrow screen with the fallback representation and the wide-screen candidate. They may be different crops of the same editorial asset, but should not quietly become different stories. Note a crop that removes a necessary label or subject for editorial review. Keep accessibility acceptance separate: suitable alternative text matters, but it cannot establish that an actual discoverable image resource has been supplied.

Check the asset independently of the page shell

Open the approved public asset through the normal delivery path in a clean test context. Record whether the result is the expected image, an access challenge, a placeholder or an error document. Compare its final address with the manifest. A successful article response does not establish that a separate asset host is functioning, and a warm browser may conceal a resource that is no longer retrievable.

Include one image loaded from the publisher’s own host and one from an approved asset service if both are used. Test the actual generated addresses, not a convenient original file that bypasses transformation. If an asset requires restricted access by design, document that boundary; do not remove protection to satisfy a discovery test. The remedy may be an approved public derivative or a decision that the image is outside the discovery objective.

Accept the page-to-image relationship

Resolve findings by object: template representation, wrong asset mapping, broken derivative or unavailable delivery address. This makes responsibility clearer than a general instruction to improve image SEO. Ask for a replacement fixture after correction and retain the old manifest row so reviewers can see precisely which connection changed.

Finish by checking a second article that uses the same component but a different image arrangement. One hero fix should not conceal a broken gallery implementation. Record the component version and accepted public resources, while keeping search visibility as a separate observation. The deliverable is a reliable path from the article’s intended visual meaning to an identifiable image file, not a collection of unsupported ranking predictions.

Make the reasoning concrete.

A fictional trade journal publishes a diagram as a CSS background in its new article header. The mobile screenshot looks correct, while the page’s image manifest contains only the logo.

The journal accepts the revised header only after the diagram has its own manifest entry and retrievable approved asset. It does not convert decorative textures or claim that the change guarantees image-search traffic.

Take the questions into your next brief.

Use these prompts in your own notes. Nothing is submitted from this page.

1. Role

   Which image conveys story content rather than decoration?

2. Representation

   What generated element and fallback identify it?

3. Candidates

   Do responsive versions retain the same editorial meaning?

4. Retrieval

   What arrives from the actual public asset address?

5. Decision

   Which component or asset mapping needs correction?

Watch for the wrong conclusion.

• Approving a picture solely because it appears in a screenshot.
• Testing the original upload instead of the published derivative.
• Making a restricted asset public without authorization.

Sources & scope

Workflows, fixtures, worksheets and examples are original editorial analysis. Sources support the narrow attributed statements, not measured business results. Platform guidance is scoped to the documented implementation and date checked.

• Google image SEO best practices →Primary Google technical documentation · Supports: Supports HTML image discovery, the responsive src fallback and the distinction from CSS images. The manifest and acceptance scenarios are original editorial analysis.