Alamy splits image metadata into two fields that do different jobs. The caption is a sentence describing who or what is in the frame and what is happening. Tags are the individual words, literal and conceptual, that the search engine matches a buyer query against. Up to ten of those tags can be marked as supertags, which the engine weights more heavily than the rest.
Caption and tags are not interchangeable
Contributors arriving from Adobe Stock or Shutterstock tend to treat the caption as a title and the tags as a keyword dump. On Alamy that costs you twice. The caption is the first thing a buyer reads about the image, and Alamy states in its own guidance that caption quality affects how external search engines pick the image up and send traffic to it. The tag list is what the internal search engine matches against. A sharp caption with a thin tag list gets read by humans and missed by the engine; a padded tag list under a vague caption surfaces in results and then loses the sale on the page.
What belongs in an Alamy caption
Alamy asks for a caption that always states who or what is in the image and what is happening. Where the date and location carry weight, such as a historical frame or an image shot at a named event, those go in the caption too. The rule underneath all of it is accuracy: Alamy warns that wrong names of people, places or animals cause real problems, and points to cases where buyers licensed images for articles and educational texts on the strength of a caption that turned out to be incorrect. Researching the subject before you caption it is not pedantry here, it is risk management.
Three things Alamy does not want in the caption. Your own name and website do not belong there, because the credit line already carries them unless you sell under a pseudonym. Captions must be written in English only; Alamy machine-translates them into other languages, and adding your own translation can confuse that system and push your images down the results. And personal or sensitive data stays out, with narrow exceptions covered further down.
Tags: literal, conceptual, and knowing when to stop
Tags can be literal or conceptual, and Alamy wants as many relevant ones as you can genuinely justify. The qualifier matters more than the instruction. Alamy is explicit that you should not force words in simply to raise the count, and that tags with little or no relevance do not belong on the image at all.
When you run dry on a minimal frame, Alamy suggests moving from objects to attributes and interactions: smiling or frowning, young or old, running or sitting. After that come the conceptual terms a buyer actually searches when they have a layout to fill rather than a subject in mind. Copy space is the example Alamy uses itself, and it is one of the highest-intent tags in commercial stock, because an art director searching it has a headline that needs somewhere to sit.
Supertags are the only ranking lever you control directly
Inside Alamy Image Manager you can promote up to ten of your tags to supertags, and the search engine gives those extra priority. Alamy frames this as picking the most relevant words only and using them wisely, which is polite phrasing for a hard constraint: ten slots is a budget, and spending them on generic terms wastes the one weighting mechanism the platform hands you.
A workable way to spend the ten: the main subject, the action, the setting, one or two conceptual terms that describe what the image is about rather than what is in it, and whatever distinguishing detail would make a buyer pick your frame over the twenty next to it. Leave out the words that describe every photograph on the site. Nobody wins a search on photo, image or background.
Two search behaviours that catch experienced keyworders out
The engine does not stem plurals
A buyer searching cat will not be shown images tagged only cats. Alamy does not extend a query to its plural, so if the plural genuinely describes the image you have to add it as a separate tag. The trap is doing this reflexively. Alamy gives the example of a lifestyle shot of one child on a swing: child is a correct tag, children is not, because a buyer typing children wants more than one and will bounce off your result. Irrelevant appearances in search hurt your ranking later, so the plural is a judgement call about the picture, not a slot to fill.
Multi-word tags are searchable by their parts
Alamy does not exclude the constituent words of a tag from being matched. A tag of Banff National Park will surface for banff, for national park and for park. That saves you from exploding every phrase into its components. It also means a phrase tag quietly enters broad searches you may not want to be in, which is why Alamy notes it tries to keep a Rhinoceros Beetle image from ranking too high in plain rhinoceros results. Phrase tags are useful, but each one widens your exposure as well as narrowing it.
Personal data and the exceptions to the rule
Alamy does not allow personal or sensitive data in captions and tags, and it lists three narrow exceptions. Journalistic and news reporting is one, and Alamy uses a red carpet example, naming a public figure at a public event. Explicit consent is the second, typically a signed model release. The third covers other legitimate reasons where there is a clear public benefit, and the example given is naming an architect alongside their building.
This sits next to the release question rather than replacing it. Marking an image as released is a separate field, and Alamy does not ask you to upload the release itself, only to flag it and produce it if a buyer needs it. The wider set of fields Alamy expects on a submission is covered in our note on Alamy metadata requirements.
The fastest route is to keyword before you upload
Alamy extracts caption, keywords and date taken from the file itself. If that information is already embedded, the images are ready for sale as soon as they clear quality control and you never open Image Manager to annotate them. The field mapping is the part worth memorising: the IPTC description becomes the Alamy caption, and the keywords field becomes the tags. If embedded metadata is unfamiliar territory, start with IPTC and XMP basics.
If the metadata is not embedded, Image Manager still lets you batch edit. Select a run of images from the same shoot or subject and apply captions, tags and license type across the whole selection at once. Alamy also accepts a spreadsheet: send your tags and account details to its image management team and they will apply the metadata for you, which is a reasonable fallback for an archive but a poor habit for daily work.
What the discoverability bar is actually measuring
This is the single most misread signal in Alamy Image Manager. The bar does not grade the quality of your metadata. Alamy says outright that it rises with the volume of searchable information you enter, nothing more. A poor discoverability warning means you have entered few tags and left optional fields blank, not that the tags you chose were weak.
Which makes chasing a full bar on every file the wrong target. Alamy is clear that not every image needs the full quota, and that relevance beats volume. A minimal still life honestly described with twelve tags is in better shape than the same image padded to forty with terms that will drag it into searches it cannot win. Padding does not get you rejected, but it does put your work in front of buyers who scroll past, and that is the pattern Alamy dampens. Outright rejections come from elsewhere, which we covered in Alamy rejection reasons.
When your edits take effect
Alamy refreshes its search index roughly every twenty-four hours. Captions and tags added today are live tomorrow, and the same delay applies when you go back and fix an old file. That cadence is worth knowing before you judge a change: if you re-tag a set of images and check the search that evening, you are looking at yesterday. Give it a day, then look.
A working pass over one image
- Write the caption as a sentence a buyer could read on its own: who or what, doing what, and where or when if it matters.
- Check every proper noun you used. A wrong place name in a caption can follow an image into print.
- Tag literally first, then by attribute and interaction, then conceptually. Stop when the next word would be a stretch.
- Add a plural only where the image genuinely shows more than one, since the engine will not infer it and a wrong plural invites the wrong buyer.
- Choose ten supertags deliberately, weighted toward what a buyer would type rather than what is technically present.
- Strip anything personal that is not covered by one of the three exceptions, and keep your own name out of the caption entirely.
Holding this together across several agencies
The awkward part of Alamy is that its rules pull against habits the other agencies reward. Alamy wants a descriptive caption where Adobe Stock wants a compact title. It punishes the reflexive plural that costs nothing elsewhere. It gives you ten weighted slots no other platform has. Keywording the same shoot once and shipping it everywhere unchanged means one of those agencies always gets the wrong shape of metadata.
That is the problem Rastock AI was built around. It generates the metadata, then shapes it per destination and delivers over FTP to 10+ agencies, with banned and mandatory keyword rules you set yourself and a rollback if a batch comes out wrong. Files and IPTC exports stay yours, and there is no revenue share. You can see how the metadata and delivery pipeline works, or check the plans and the 14-day free trial, which does not ask for a card.
Frequently asked questions
How many tags should I add to an Alamy image?
Alamy asks for as many relevant tags as the image genuinely supports, and explicitly tells contributors not to force words in to raise the count. There is no target number. A simple studio shot might honestly justify twelve tags; a busy street scene might justify forty. Irrelevant tags pull your image into searches it cannot win, which hurts its ranking over time.
What are Alamy supertags and how many can I use?
Supertags are the tags you mark as most relevant to an image inside Alamy Image Manager. You can designate up to ten, and the search engine gives them extra priority over your ordinary tags. Because the allowance is fixed, spend it on the words a buyer would actually type rather than on generic descriptors that every image on the site could claim.
Do I need to add both singular and plural tags on Alamy?
Only when the plural describes the image. Alamy does not stem search terms, so a query for cats will not match an image tagged only cat, and you must add the plural separately if it applies. But Alamy warns against doing this by habit: a photo of one child should carry child and not children, because a buyer searching the plural wants more than one and will pass over your result.
Can I write Alamy captions in a language other than English?
No. Alamy asks for English captions only and machine-translates them for buyers in other markets. Adding your own translation can confuse that system, and Alamy warns it may lead to your images being demoted in search results. The caption is one of the few Alamy fields where doing extra work actively harms you.
Why does Alamy say my image has poor discoverability?
The discoverability bar measures how much searchable information you have entered, not how good it is. A low reading means few tags and blank optional fields rather than badly chosen words. Alamy notes that not every image needs the full quota, so the fix is to add the information that genuinely applies rather than padding until the bar turns green.
Can Alamy read captions and tags from my file metadata?
Yes. Alamy extracts caption, keywords and date taken from embedded IPTC data, and images arriving with that information can go on sale without you touching Image Manager. The IPTC description field maps to the Alamy caption and the keywords field maps to tags. Embedding metadata before upload is the fastest route for anyone submitting regularly.