Google Images SEO Checklist: How to Get Your Images Found in Search

Google Images SEO means making your images easy for Google to find, fetch, understand and show. In practice that comes down to a short list: put each image in a standard HTML img element with a crawlable src address, write accurate alt text, place the image near relevant text, keep the file fast and the URL stable, and do not block it from crawlers. None of this promises that an image will appear or rank. It removes the common technical reasons that images are missed.

This guide follows Google Search Central's image guidance and turns it into a checklist for site owners and developers. It includes markup to use, a 25-point table and a way to measure results in Search Console.

Before you begin

You will need:

Access to the page templates or CMS where images are added.
Access to your robots.txt file and, if you use one, your sitemap.
A verified Google Search Console property for the site.

The main risk is changing image URLs or blocking a folder by mistake.

How Google Images finds and shows an image

Google can only show an image it can crawl and index, and it generally links the result to the page that hosts it. Its guidance says it extracts information about an image from the page content, including captions and titles, so the page matters as much as the file.

The path has four stages: discovery (Google finds the image URL in your HTML or sitemap), fetching (Googlebot-Image requests the file), understanding (the file, alt text, captions and page context are read) and indexing. Problems usually arise from a CSS-only image, a robots.txt block, a server error, missing context or a landing page that is not indexable. Display in image results is never promised, even for a well-optimized image.

Use the HTML img element

Google's guidance says it crawls images referenced in the src attribute of an img element, and it does not index CSS images. If a picture is meant to be found, put it in markup:

<img src="https://example.com/images/blue-trail-shoe-side.jpg"
     alt="Blue Trailrunner 2 running shoe, side view"
     width="1200" height="800">

A picture element with an img fallback is also fine, and so is srcset, which lets the browser choose between sizes. Avoid patterns where the real address only appears in a data attribute and a script swaps it in after a click. If you lazy load, use native loading="lazy" and test that the images appear in rendered HTML. See Lazy Loading Images for SEO: Defer Images Without Hiding Them.

Write descriptive alt text

Alt text is the text equivalent of the image in the alt attribute. Google says it uses alt text with computer vision and the page content to understand an image, and as anchor text when the image is a link. Describe what the image shows in context, in plain words. Decorative images should use alt="". For rules and examples, read How to Write Alt Text: Examples for Every Type of Image.

Put the image near relevant text

Place an image close to the text it illustrates, under a heading that matches its subject. A page about trail running shoes gives a shoe photo more context than an image on a page about something else. A short figure and figcaption can help people and give Google extra context:

<figure>
  <img src="/images/trail-shoe-side.jpg" alt="Blue Trailrunner 2 shoe, side view" width="1200" height="800">
  <figcaption>The Trailrunner 2 has a rock plate under the forefoot.</figcaption>
</figure>

Do not repeat the caption word for word in the alt text. Also make sure the landing page itself is useful and indexable, because the image result links to it.

Use descriptive file names, as a light hint

Google recommends short, descriptive file names. A name such as blue-trail-running-shoe.jpg tells people and crawlers more than IMG_4821.jpg. Treat it as a light hint, not a lever. Separate words with hyphens and avoid long keyword chains. Renaming existing files changes their URLs, so do it only if you can redirect the old address. See SEO-Friendly URLs: How to Build Readable, Stable Web Addresses.

Choose sensible formats and sizes

Google lists the formats it supports for image search as BMP, GIF, JPEG, PNG, WebP, SVG and AVIF. Pick by content:

Content	Often a good fit	Notes
Photographs	JPEG, WebP or AVIF	Check current browser support for AVIF at caniuse.com.
Logos and flat graphics	SVG or PNG	SVG scales without blur.

Serve images no larger than the space they fill, with srcset and sizes. Set width and height attributes so the browser reserves space and reduces layout shift. Core Web Vitals "good" thresholds, measured at the 75th percentile, are Largest Contentful Paint (LCP) at or under 2.5 seconds, Cumulative Layout Shift (CLS) at or under 0.1 and Interaction to Next Paint (INP) at or under 200 milliseconds. Large images are a frequent cause of slow LCP. See Responsive Images with srcset and sizes: Serve the Right Size and Use WebP and AVIF Images to Improve Website Speed.

Keep image URLs stable and unique

If you change an image address, Google has to discover and process it again, and the old address may stop working. Avoid tokens in the path that change on every deployment, and redirect old image addresses when a move is unavoidable. If you use a CDN (content delivery network, a service that serves files from servers close to visitors), keep the image hostname the same over time.

Do not block images in robots.txt

A robots.txt file tells crawlers which paths they may fetch. Google's image crawler is called Googlebot-Image. This rule blocks it from a folder:

User-agent: Googlebot-Image
Disallow: /images/

Only use a rule like this if you really want those images out of Google Images. Broad rules such as Disallow: /assets/ can catch image folders by accident. If an image must be removed from results, a noindex for image files can be sent in an X-Robots-Tag header; see X-Robots-Tag Header: Noindex PDFs, Images and Other Files. To diagnose blocking, read Robots.txt Blocking Google or Bing? Diagnose It Step by Step.

Image sitemaps are optional

If your images are in normal img elements on crawlable pages, Google can usually find them without extra help. An image sitemap can help discovery in less common setups. Treat it as an extra, not a fix for markup that hides images. See Image, Video and News Sitemaps: When They Help and How to Build Them.

Add structured data only where it is valid

Structured data is code, usually JSON-LD, that labels what a page is about. For products, recipes and videos, Google's documentation includes image properties, and images on valid pages can sometimes be shown with badges or richer results. Add it only when the page truly is a product, recipe or video page, match what visitors can see and test it with the Rich Results Test. Each feature has its own image requirements, and markup does not guarantee a special result. See Product Schema for Shops: Merchant Listings and Product Snippets.

Hotlinking and duplicate copies

Hotlinking means another site embeds your image directly from your server, using your bandwidth. Keep one clear original for important images, use consistent URLs and consider server rules to limit direct embedding if bandwidth is a problem. Test such rules carefully, because strict ones can also block crawlers and your own CDN.

Step-by-step instructions

Step 1: Inventory your images

List each image address, its page, its size and its alt text. Start with product, article and landing pages.

What success looks like: you have a list of image URLs with page and size.

Step 2: Fix markup

Convert important CSS background images and script-swapped images to img elements with a real src. Add width and height.

What success looks like: each important image is in the page HTML as an img.

Step 3: Fix alt text and context

Write alt text for each informative image, set decorative ones to alt="" and move images next to relevant text.

What success looks like: every informative image has accurate alt text and sits near related content.

Step 4: Optimize files

Compress, resize and serve modern formats with srcset. Do not lazy load the first large image.

What success looks like: images are no larger than needed, and the page's LCP element loads promptly.

Step 5: Check access

Confirm robots.txt does not block image folders. Request an image address with curl -I https://example.com/images/blue-trail-shoe-side.jpg and confirm a 200 response and an image/... content type.

What success looks like: image files return 200 and are not disallowed.

Step 6: Measure in Search Console

Open the Performance report in Google Search Console, then set the Search type to Image. The report shows clicks, impressions, click-through rate and average position for image search. Filter by page to see which pages earn image impressions. Use URL Inspection on a key page to see whether Google can fetch and render it. Allow time for recrawling, since changes are not instant.

What success looks like: you have a baseline of image impressions and clicks to compare against later.

The 25-point Google Images SEO checklist

#	Check	Why it matters
1	Images are in `img` or `picture` elements	Google crawls `src`, not CSS backgrounds
2	`src` is a real, crawlable URL	Avoids script-only addresses
3	Images load without a click	Content hidden behind interaction may be missed
4	Important images appear in rendered HTML	Confirms Google can see them
5	Alt text describes the image	Context for people and search
6	Decorative images use `alt=""`	Keeps noise out
7	Alt text is free of keyword stuffing	Matches Google's advice
8	Image sits near related text	Page context aids understanding
9	Captions used where helpful	Adds context for visitors
10	File names are short and descriptive	Light hint
11	File names use hyphens and lowercase	Readable, consistent URLs
12	Format is supported by Google	BMP, GIF, JPEG, PNG, WebP, SVG, AVIF
13	Format suits the content	Smaller files, better quality
14	Images are not larger than shown	Saves bytes
15	`srcset` and `sizes` provide options	Right size per device
16	`width` and `height` are set	Reduces layout shift
17	First large image is not lazy loaded	Protects LCP
18	Below-the-fold images lazy load safely	Saves load time
19	LCP, CLS and INP are in the good range	Page experience for users
20	Image URLs are stable	Avoids recrawling and broken results
21	Old image URLs redirect if moved	Keeps existing signals
22	`robots.txt` does not block image folders	Allows fetching
23	Landing pages are indexable and useful	Image results link to them
24	Optional image sitemap is valid	Helps discovery
25	Structured data is valid and matches the page	Only for products, recipes, videos

Using Crawvix for an image check

Crawvix's Image Doctor lists the images written into the HTML of the pages of a full scan of a verified website (up to 2,000), checks up to 200 of them for status, type and file size, and reports missing or generic alt text, broken images, large files, missing width and height, a lazily loaded first image, older formats and generic file names. Each finding comes with evidence and what to do, and there is a CSV export. It only reads a website and never changes it. The page is /image-doctor.

Final verification checklist

Important images are img elements with working src addresses.
Files are the right size, in a sensible format, with width and height set.
The first large image is not lazy loaded.
Image URLs return 200, are stable and are not blocked by robots.txt.
The Search Console Performance report is set to the Image search type and a baseline is saved.

Common mistakes to avoid

Using CSS backgrounds for content images. Google does not index CSS images.
Keyword-stuffed alt text. It hurts people and goes against Google's guidance.
Blocking image folders in robots.txt. Check broad Disallow rules.
Changing image URLs often. Keep addresses stable.
Lazy loading the hero image. It can delay LCP.

Frequently asked questions

How do I get my images to show up in Google Images?

Put them in img elements on indexable pages, add accurate alt text, place them near relevant text and make sure they are not blocked. Google decides what to index and show, so no steps ensure inclusion.

Does alt text help Google Images SEO?

Google says it uses alt text, with page content and computer vision, to understand images. Write it for people first and keep it accurate, without keyword stuffing.

Do I need an image sitemap?

No. It is optional. If your images are in normal HTML on crawlable pages, Google can often find them without one, but a sitemap can help discovery in less typical setups.

Do file names matter for image SEO?

They help a little. Google recommends descriptive names, so blue-trail-running-shoe.jpg beats IMG_4821.jpg, but treat the name as a light hint and not a ranking lever.

How do I see image search traffic in Search Console?

Open the Performance report, choose Search type, and select Image. You will see clicks, impressions, click-through rate and position for image results, and can filter by page or query.

Why are my images not indexed?

Common causes are CSS backgrounds, script-only loading, robots.txt blocks, errors when fetching the file, or a landing page that is not indexed. Test the page with URL Inspection and fetch the image address directly.

Official references

Next step: Run Image Doctor on a verified website to see image problems with evidence, or scan the whole website for wider technical checks. This guide provides technical information, not a guarantee of rankings, traffic or index inclusion.