Image Alt Text Checker

Enter a URL and this lists every image tag on the page alongside its alt text, flagging the ones that are missing, say a single word, are file names, repeat themselves, or run longer than a listener will sit through.

Reading the result

Every image on the page gets one verdict, and the verdict names the work. Start with the Need attention filter, which hides everything already correct.

No alt attribute: the tag carries no alt at all, so a screen reader falls back to reading the file name. Describe the image, or write alt="" if it is decorative.
Whitespace only: the attribute holds a space, which keeps the image announced but leaves it unnamed. Write a description, or a true alt="".
Placeholder and File name: the alt text repeats what the browser already knows, or it is whatever the upload was called. Replace it with a description.
Redundant prefix: the alt text opens with "image of", which a screen reader has already announced. Cut the prefix and start with the subject.
One word and Very short: the alt text names the subject without describing it. Add what the image shows, as in "RankYak dashboard on the keyword overview" rather than "dashboard".
Keyword stuffing: the alt text reads as a list of search terms. A screen reader reads all of it aloud, so cut it back to the description.
Too long: the description runs past 125 characters. Keep the essential part in the alt text and move the rest into a caption, which everyone reads.
Tracking pixel: a 1x1 image that records a visit rather than showing anything. It is listed for completeness and counted nowhere, and it needs no alt text.
Decorative and Descriptive: nothing to do.

The Lazy loaded and No dimensions labels sit beside the verdict and are about speed rather than about alt text. Lazy loading is right for images further down the page and wrong for the first image a visitor sees; a missing width and height lets the page jump as each image lands.

Empty is a valid answer

Most alt text audits report a single number: how many images lack alt text. That number pushes people toward the wrong fix, which is writing a description for every image on the page including the ones that carry no meaning. A screen reader user then hears a list of decorative graphics before reaching the content.

The distinction this tool draws is between a missing attribute and an empty one. A missing alt leaves assistive technology guessing, and it will often fall back to reading the file name aloud. An empty alt="" is an explicit statement that the image is decorative, and it is exactly right for dividers, textures, and icons that sit beside a label saying the same thing.

So the work is a triage rather than a fill-in-the-blanks: decide which images carry meaning, describe those, and mark the rest as decorative. The result is both more accessible and shorter than describing everything.

Frequently Asked Questions

What makes good alt text?

A short description of what the image shows and why it is on the page, written for someone who cannot see it. Usually a sentence fragment of five to fifteen words. If the image carries information the surrounding text does not, the alt text must carry it too; if the image is purely decorative, an empty alt is the correct answer.

When should alt text be empty?

When the image adds nothing a reader would miss: background textures, decorative dividers, icons beside a label that already says the same thing. Writing alt="" tells a screen reader to skip it, which is a better experience than hearing "decorative swoosh". An empty alt is a deliberate statement; a missing alt attribute is not.

Why is a one-word alt text flagged?

Because a single word names the image without describing it. An alt of "RankYak" tells a screen reader user that something to do with RankYak is on the page and nothing more, while "RankYak logo" tells them what they are looking at. The check counts words rather than characters, so "logo" and "screenshot" are treated the same way: one is short and the other is not, but neither describes anything.

Japanese, Chinese, Korean, and Thai are written without spaces between words, so a whole caption in one of them is a single run of characters. Alt text in those scripts is measured in characters instead, and only the shortest labels are flagged.

What does the lazy loaded label mean?

It means the image carries loading="lazy", so the browser holds it back until the visitor scrolls near it. That is the right setting for images further down a page, because the rest of the page then arrives sooner. It is the wrong setting for the first image a visitor sees, which is often the largest element on the screen and the one Largest Contentful Paint measures.

Is missing alt text an SEO problem or an accessibility problem?

Both, and the accessibility side is the more serious one. For SEO, alt text helps Google understand the image and is how images rank in image search. For accessibility, it is how a screen reader user learns what the image shows, and a meaningful image left without a text alternative fails WCAG success criterion 1.1.1, Non-text Content, at Level A.

Should I put keywords in alt text?

Only where they describe the image honestly. Alt text stuffed with keywords is picked up as spam by Google and, more immediately, is read aloud to a person trying to use your site. If the target keyword genuinely describes what the picture shows, use it; if not, describe the picture.

Why does the tool flag images with no width and height?

Because the browser reserves space for an image before it arrives only when it knows the size, and everything below it moves once the image lands. That movement is what Cumulative Layout Shift measures, and Cumulative Layout Shift is one of the Core Web Vitals. Width and height attributes settle it, and so does a CSS aspect ratio. This tool reads the markup rather than the stylesheet, so a page that reserves the space in CSS still carries the flag.

Which images does this tool check?

The tool reads the HTML the page returns and lists the <img> tags in it. An image painted by CSS as a background has no alt attribute to check, an inline <svg> is described by a <title> element rather than by alt text, and an image a script inserts after the page loads is not in the markup the server sent, so none of those appear in the results. They still need an accessible name, so you should check them yourself.