You upload a product photo. It loads on the page. Customers see it. Job done.
Except the photo only exists for one audience: people who open that specific page and look at it with their eyes. For every other way the internet processes information — search indexing, screen readers, social media crawlers, image search, Pinterest, AI assistants — the photo is either invisible or described by whatever text the platform can scrape from nearby content. Usually the filename. Usually something like IMG_4892.jpg.
That is not a presence on the internet. That is a ghost.
What alt text actually is
Alt text is not an SEO trick. It is not a keyword-stuffing opportunity. It is a text description of an image, originally created so that people who cannot see the image — because of a visual impairment, a slow connection, or a broken link — still know what it contains.
That original purpose still matters. Around 2.2 billion people globally have some form of vision impairment. Screen readers are not a niche technology; they are how a significant portion of the internet experiences the web. When a screen reader hits an image without alt text, it either skips it entirely or reads the filename aloud. Neither is useful. Neither communicates what the product looks like.
But alt text has grown well beyond accessibility. Google uses it as a primary signal for understanding what an image contains. Pinterest reads it to categorize pins. Social platforms use it when rendering link previews. AI tools consume it to understand product catalogs. Alt text has become the canonical text description of a visual asset, and every system that processes the web depends on it.
When you leave alt text blank, you are not just failing accessibility standards. You are removing your images from every system that could surface them to new audiences.
Alt text is not metadata. It is a description written for a reader who cannot see the image. Write it for that reader and every other system benefits automatically.
The Google Images problem most stores ignore
Here is a number that should bother you: Google Images accounts for roughly 22% of all web searches. Not 22% of Google searches — 22% of all searches on the internet. It is the second largest search engine on the planet, behind Google’s main search and ahead of YouTube, Bing, and everything else.
Think about what happens when someone searches “blue linen blazer” in Google Images. Google cannot look at your product photo and determine that it shows a blue linen blazer. It makes guesses using computer vision, and those guesses have improved, but the primary signal it relies on is the alt text attached to the image tag.
If your alt text says “blue linen blazer, relaxed fit, notch lapel,” your image competes for that query. If your alt text is empty, or says “product-image-3.jpg,” or says “blazer,” you are invisible to the second largest search engine on earth. Not poorly ranked. Invisible.
This is traffic you are not competing for. Not losing — never entering the race.
Why most alt text is bad even when it exists
The stores that do write alt text tend to fall into two traps.
The first is keyword stuffing. “Blue linen blazer men’s blazer men’s fashion men’s clothing sustainable fashion summer blazer linen jacket.” This reads like a tag cloud, not a description. Google recognizes it as spam. Screen readers read every word of it aloud, which is a miserable experience for the person listening. It helps nobody.
The second is being too vague. “Product image.” “Front view.” “Photo of a jacket.” These are technically descriptions, but they carry no useful information. They do not tell Google what distinguishes this image from millions of others. They do not help a visually impaired customer understand what the product actually looks like.
Good alt text sits between those extremes. It is specific enough to be useful and natural enough to be read aloud without embarrassment. It describes what someone would see if they could see the image, the same way you would describe it to a person standing next to you.
“Blue linen blazer with notch lapels, unstructured fit, shown on a model against a white background.”
That is a sentence a person would say. It is also exactly what Google needs to understand the image. The best alt text serves both audiences simultaneously because both audiences want the same thing: an accurate, plain-language description of what is in the picture.
The scale problem is real
This is where most advice about alt text falls apart. It is easy to say “write good alt text for every image.” It is another thing to do it when your store has 400 products with four images each. That is 1,600 descriptions to write.
Most merchants face this arithmetic and choose one of three paths: write nothing, copy the product title into every alt field, or pay someone to write them all once and never touch them again.
None of these work at scale. Writing nothing is abandonment. Copying the product title is lazy — if you have five images of the same product, each showing a different angle, they should not all say “Blue Linen Blazer.” The first image might show the full front view. The second might show the fabric texture up close. The third might show the back panel. Each image tells a different part of the story, and the alt text should reflect that.
The third option — writing them once and forgetting — decays as your catalog evolves. New products get added without alt text. Old products get new photos without updated descriptions. Coverage degrades silently, and nobody notices because nobody audits alt text on a Tuesday afternoon.
This is a systems problem, not a writing problem. You need a process that generates accurate alt text as part of your product workflow, not as a separate project you run once a year when someone mentions accessibility.
What your images actually need to say
Not every product image requires the same kind of description. What you write should change based on what the image shows.
- Primary product image. Describe the product the way a customer would. Include the product type, key visual features — color, material, shape — and any notable design details. “Navy cotton crew-neck sweater with ribbed cuffs and hem.”
- Lifestyle or context image. Describe the scene. “Woman wearing the navy crew-neck sweater while sitting at a cafe table.” This matters because Google indexes the context — someone searching “cafe outfit” might discover your sweater through this image.
- Detail or close-up shot. Describe the specific detail being highlighted. “Close-up of the ribbed knit texture on the navy crew-neck sweater sleeve.”
- Scale or sizing reference. Describe the comparison being shown. “Navy crew-neck sweater laid flat next to a ruler, showing a 24-inch chest width.”
- Color variant swatch. Describe the specific color. “The crew-neck sweater in forest green.” Do not just say “green” — be specific enough to match what a customer actually sees.
Each image earns its own description because each image communicates something different. When you collapse them all into the same product title, you waste the distinct opportunity each photo creates.
The quality of your alt text is a direct reflection of the quality of your product data. Stores with structured tags write better alt text faster because the raw information already exists — it just needs to be composed into a sentence.
Tags make alt text possible at scale
Here is where structured product data connects to alt text in a way most stores never consider.
If your products are tagged with attributes like color, material, product type, fit, and occasion, generating accurate alt text becomes a data problem instead of a creative writing problem. You already know the product is a “navy cotton crew-neck sweater.” That information lives in your tags. The alt text for the primary image is a composition of those attributes into a natural sentence.
This does not mean alt text should be blindly auto-generated. It means good tagging gives you the raw material to produce first drafts that are already 80% accurate, instead of starting from a blank field 1,600 times. You review and refine instead of inventing from scratch.
The stores with the best alt text coverage are not the ones with the most diligent copywriters. They are the ones with the best product data. Tags, attributes, and structured metadata feed the systems that generate and maintain descriptions over time. Without that foundation, alt text is a manual chore that will always fall behind your catalog.
A twenty-minute audit
You do not need to overhaul your entire catalog today. Start with a focused check.
- Pick your 20 best-selling products. Open each product page and inspect the images. How many have alt text? How many have useful alt text versus a filename or the product title pasted into every field?
- Search for your own products on Google Images using specific attributes — “red silk scarf,” not just your brand name. See which images appear. The ones that show up probably have decent alt text. The ones missing probably do not.
- Check your theme’s default behavior. Some Shopify themes automatically use the product title as alt text when no custom value is set. This is better than nothing but still means every image of the same product carries an identical description.
- Prioritize by traffic impact. Your homepage hero images, collection banners, and top-selling product photos get the most impressions. Fix those first. A store with strong alt text on its top 50 products and gaps elsewhere is in better shape than a store with mediocre alt text everywhere.
Stop decorating and start describing
Alt text is not a decoration. It is not a field you fill in to satisfy an SEO plugin’s audit score or check a compliance box. It is a description of what your image shows, written for an audience that includes humans who cannot see and machines that cannot interpret pixels.
When you treat it as a description instead of an optimization, the quality improves and the anxiety disappears. You stop worrying about keyword density and start writing what is actually in the picture. That is what every audience — human and algorithmic — needs from you.
Your product images are rich with information. Color, material, fit, texture, context, detail. Right now that information is trapped inside pixels that only sighted visitors on your specific page can interpret. Alt text releases it into the layer of the web that search engines, screen readers, social platforms, and AI systems can actually read.
Until you write it, your images are not on the internet. They are on your page. And those are not the same thing.