Find the images on a page that have no alt text at all. Paste a URL and this free check reads the page's main content as the server sends it, counts every <img> with no alt attribute, and passes decorative images marked alt="", because that is the correct markup for them.
<img> with no alt. Describe the ones that carry information; mark the decorative ones empty. Both are shown below.<!-- an image that carries information: say what it shows --> <img src="team-at-closing.jpg" alt="Our two partners signing the purchase agreement"> <!-- a decorative image: empty alt, so screen readers skip it --> <img src="divider.png" alt="">
1 check. Does any <img> in the page's main content lack an alt attribute. One GET of the URL as VerandBot/1.0, redirects followed, 12 second timeout, the served HTML only: no JavaScript is run. The main content is the first <article> or <main> holding 250 characters of text or more, otherwise the body with header, nav, footer and aside removed.
An image passes when its tag carries alt with a quoted value, and alt="" is a value: it marks the image as decorative, so it passes. Missing means no alt attribute at all. A fail comes back as a count, N of M images, not a list of file names.
Whether the alt text is accurate: this checks presence, not quality, so a pass does not mean the page meets WCAG. CSS background images, inline SVG, and images in <noscript>, iframes or added by script. Labels given by aria-label instead of alt, and an unquoted or bare alt, read as missing. Logos in the header and other images outside the main content are not counted, and on a listing page only the first <article> is read.
One fetch, one region, one question per image tag. The card is the tool in motion on an example article, looped, and each step lights up while the card is doing it.
<img src="distributions-chart.png" alt="Bar chart of quarterly distributions, see the table">
The tool asks for the URL you typed, identifying itself as VerandBot, following redirects, giving up after twelve seconds. It reads the HTML the server sends. No JavaScript runs, so an image a script inserts after load is never seen.
Scripts, styles, inline SVG, noscript blocks, iframes and forms are dropped. The first <article> or <main> with at least 250 characters of text becomes the region; failing that, the body minus its header, nav, footer and aside.
Each <img> in the region is tested for an alt attribute with a quoted value. An empty value is still a value, so a divider marked alt="" passes. Only a tag with no alt at all counts as missing.
A fail reads "1 of 4 images have no alt text": the count, not the file names. The highlight on this card is for the example. On your page, the fix is a choice per image: describe it, or mark it decorative.
Alt text is the one sentence a page gives for each picture, read aloud to people who cannot see it and read as text by every crawler. Here is what it is for, the difference most checkers blur, and how to write it for a firm whose images carry claims.
Image alt text is the text alternative for an image, written in the alt attribute of the <img> element. It is an attribute, not a tag, but "alt tag" is the name most people search for, so both appear on this page and mean the same thing. It lives in the HTML beside the image's address:
<!-- informative: the text does the image's job --> <img src="k1-part-3.png" alt="Schedule K-1, Part III, box 1 highlighted"> <!-- decorative: empty, so assistive technology skips it --> <img src="leaf-divider.svg" alt=""> <!-- missing: no alt attribute at all --> <img src="IMG_4471.jpg">
Three states, and they are not two versions of the same thing. The first tells a reader what the image shows. The second tells assistive technology the image is decoration and should be passed over. The third says nothing, and leaves every reader of the HTML to guess.
Screen readers. A person using a screen reader hears the alt text in place of the image. When the attribute is missing there is nothing to read, and the W3C's own failure technique for this case, F65, notes that some assistive technology compensates by reading out the file name instead. IMG_4471.jpg, spoken aloud, is the sound of a missing alt attribute.
Image search. Google's image SEO guidance calls alt text the most important attribute for giving an image metadata, and says Google uses it along with computer vision and the page's content to understand what an image is about.
Anyone whose images don't load. On a slow connection or a blocked image host, the browser shows the alt text where the picture should be.
Every crawler that reads HTML. Search crawlers and the crawlers behind AI assistants receive the page as text. The alt attribute is text; the picture is not. Whatever a chart says, a machine reading the HTML gets only what the alt text and the surrounding copy say about it.
alt="": the difference most checkers blurNot every image needs describing. A divider, a background flourish, an icon sitting beside a text label that already says the same thing: describing those makes a screen reader stop to announce noise. The accessible markup for a decorative image is an empty alt attribute, alt="", which the W3C documents as technique H67. It is a deliberate instruction: skip this.
A decorative image is not the same as an image with no alt. Leave the attribute off and assistive technology is left guessing, which is how file names end up read aloud. Some ranking pages say decorative images "don't need" alt text, which is true of the description and wrong about the attribute. The attribute stays; its value is empty.
That is why this checker reports only a missing attribute as a fail. A checker that flags every alt="" punishes the correct markup, and trains site owners to stuff descriptions onto dividers to make a warning go away. The cost of getting it right is the other half: an informative image someone marked alt="", a chart or a product photo, passes here, because presence is what a machine can check and meaning is what a person has to.
Describe what the image shows and why it is on this page. The same photograph of a building is "Front entrance of the firm's office" on a contact page and "Three-story garden apartment building, 1980s brick" in an article about property age. Context decides.
A caption (<figcaption>) is visible to everyone and usually adds context the image can't: a date, a source, a credit. Alt text replaces the image for someone who can't see it. The title attribute shows as a hover tooltip, is unreliable on touch screens and is not a substitute for alt. When a caption already describes the image fully, the alt text can be short, but it should not simply repeat the caption word for word, or a screen reader user hears it twice.
These are the images that most need real alt text and most often get a file name. A chart's alt text should say what the chart is and what it shows at the level the page relies on: "Bar chart of quarterly distributions, 2021 to 2025, detailed in the table below." A screenshot's alt text says what the screenshot demonstrates, not every pixel. When the data matter, put them in a table or in the prose next to the chart, and let the alt text point there.
For a regulated firm there is a second reason to be careful here. Alt text is content. If a chart shows performance figures, alt text that restates those figures carries the claim to screen readers and to every crawler reading the HTML, without the footnote or disclosure that sits beside the chart on screen. Describe the chart's subject and point to the table, rather than repeating its numbers, unless the disclosure travels with them. Treat alt text on any chart of returns, outcomes or results as copy, and review it with compliance the way you review the caption.
Every alt checker that reads HTML shares one blind spot, and it is worth knowing where it is.
background-image has no alt attribute to check. Google's guidance says plainly that it doesn't index CSS images. If a background image carries meaning, it should be an <img>, or its meaning should be in the text on top of it.<svg> takes its accessible name from a <title> inside it or from role="img" with aria-label. This checker drops SVG before it counts, so it neither passes nor fails them.<img> named with aria-label can be accessible without an alt attribute. This checker reads only alt, so it counts that image as missing. Adding the same text as alt resolves it.WordPress. In the block editor, select the image and fill in the Alternative text field in the block settings sidebar. Alt text set in the Media Library is copied in when the image is inserted, so editing it in the library later does not change posts that already use the image; fix each post. Leave the field empty for a decorative image and WordPress writes alt="". Static sites and templates. Add the attribute in the HTML or the template that renders the image; one missing alt in a shared template repeats on every page that uses it. Other site builders put an alt field in the image's settings panel; look for it before you publish, not after.
Missing alternative text is one of the most common failures that WebAIM's annual Million report detects when it scans home pages, and it is among the easiest for any automated scan to find. Law firms, medical practices and other professional practices receive web accessibility demand letters, and a scan that turns up images with no alt text is an easy exhibit. Fixing missing alt text removes that finding. It does not make a site accessible on its own: color contrast, form labels, keyboard access and the quality of the alt text itself all need a person. Use this check as the first pass, and an accessibility review as the second.
alt="IMG_4471" or alt="hero-final-v2" passes a presence check and helps no one.alt="".Six things that are true of this tool, each one backed by a line in the code that runs it.
The request carries a URL and nothing else. There is no account and no session behind the tool, and the result is on the page, not in a report you have to ask for.
Only an image with no alt attribute counts as missing. A decorative image marked alt="" passes, because that is the accessible markup, so the tool never pushes you to describe a divider.
Each image tag is tested with a fixed pattern for an alt attribute. No model reads the page, so the same HTML gives the same count every time.
This is the same page check Verand's deep crawl runs on every page of every customer site every two weeks, called directly on one URL. Not a lighter demo version.
Presence, not quality. CSS backgrounds, inline SVG, script-inserted images and ARIA labels are out of scope, and the card says so beside the result rather than in a footnote.
A run is one page fetch and a parse, so it is never metered. The one limit is a courtesy to the sites being fetched: 20 checks a minute per visitor.
What alt text is, when it should be empty, and how the checker tells the two apart.
Yes, they are the same thing. Image alt text is the text alternative written in the alt attribute of an img element, read by screen readers in place of the image and shown by browsers when the image fails to load. Strictly it is an attribute, not a tag, but "alt tag" is the common name. This checker looks for that attribute on every image in a page's main content and reports the images that have none.
They need the attribute, with an empty value: alt="". That tells assistive technology to skip the image, which is what you want for a divider, a background flourish or an icon beside a text label that says the same thing. Leaving the attribute off entirely is a different result, because assistive technology then has nothing to go on and may read the file name aloud. This checker passes alt="" and fails only a missing attribute.
An empty alt attribute, alt="", is a deliberate instruction that the image is decorative. A missing one is no instruction at all. The W3C documents the first as an accessible technique (H67) and the second as a failure (F65). When you check alt text with this tool, only the missing case counts against the page. The flip side: an informative image someone marked alt="" also passes, because the checker confirms the attribute is there, not that its value is right.
There is no official limit. The working habit is about one short sentence that says what the image shows and why it is on the page. If an image needs a paragraph to explain, such as a detailed chart, write the paragraph or a table in the page itself and let the alt text name the chart and point to it. This checker does not measure length; an image alt text checker that reads presence cannot judge whether the words are the right ones.
Both, in that order. Google's image SEO guidance says it uses alt text, with computer vision and the page's content, to understand what an image shows, which matters for image search. The same guidance warns that filling alt attributes with keywords makes a poor experience and may cause a site to be seen as spam. Write it for the person who can't see the image, and it does its search job as a side effect.
Name the chart and what it shows, then point to where the data are: "Bar chart of quarterly distributions, 2021 to 2025, detailed in the table below." Put the figures in a table or in the prose, where everyone can read them. If the chart shows performance, returns or outcomes, remember that alt text carries the claim to screen readers and crawlers without the footnote beside the chart, so describe the subject rather than repeating the numbers, and review it with compliance like any other copy.
Alt text is one line of a page that has to earn its place. Verand writes articles from your own expertise and credentials, so Google and the AI assistants have something of yours to name. It then tracks where you rank and where ChatGPT, Gemini, Google AI Overviews, Google AI Mode, Perplexity and Claude mention you, and gates every draft so a claim your regulator would not allow never publishes.
Content built to rank in
Google and get cited by
ChatGPT
Perplexity
Gemini
Claude, with every claim checked before it goes live.