Good keywording is probably the single biggest lever you have over how often your stock photos get found. Agencies like Shutterstock and Adobe Stock are essentially search engines — they rely on the words attached to your images to figure out what a photo is about and when to surface it.
Most contributors either use too few keywords (leaving buyers unable to find the photo) or dump in whatever comes to mind without thinking about what someone actually searches for. Both kill discoverability.
This guide covers the practical side: how many keywords to use, which types matter, and the mistakes worth avoiding.
What keywords actually do
When a buyer searches "businesswoman in office meeting," the agency scans keyword lists, titles, and descriptions across millions of files. Photos tagged with "businesswoman," "meeting," "office," and "corporate" surface. Photos tagged only with "woman" or "person" don't rank nearly as well.
The gap between how you think about your own photo and how a buyer searches for it is usually bigger than you'd expect. Keywords are how you close that gap.
How many keywords should a stock photo have?
Shutterstock accepts up to 50 keywords. Adobe Stock has a similar ceiling. For most photos, 30 to 50 is the range that makes sense — enough to cover the real concepts without padding with irrelevant terms.
Some contributors aim for exactly 50 every time, treating it as a slot-filling exercise. Don't do that. A photo of a single red apple doesn't need 50 keywords. A busy street scene at night with crowds, neon lights, and rain does. Match keyword count to what's actually in the frame.
Under 15 keywords and you're almost certainly missing important search terms. If you're consistently maxing out at 50, scan the bottom third — there's usually filler in there.
Five keyword types worth thinking about
Brainstorming from scratch every time is slow and inconsistent. Work through these categories instead.
Subject keywords are what's literally in the frame. Person, dog, mountain, coffee cup — your core terms. Be thorough here. If there are three people in the photo, tag "group," "team," "people," "men," "women" as applicable. Include both the specific ("golden retriever") and the general ("dog," "pet").
Action and concept keywords describe what's happening or what idea the image represents. Running, working, celebrating, isolation, freedom, teamwork. These matter because buyers often don't have a specific subject in mind — they have a concept they're trying to illustrate.
Setting keywords cover where it's happening. Indoors, outdoors, kitchen, forest, urban, suburban, tropical. Add the season if it shows — a winter landscape and a summer one are different assets to different buyers.
Feeling and mood keywords — this is the most skipped category, which is a shame because buyers search by mood a lot. "Peaceful," "energetic," "melancholy," "tense," "cozy" — these words surface your photos in searches that subject-only tagging misses entirely.
Color keywords name the dominant colors. Red, warm tones, pastel, monochrome, golden. Art directors and designers search by color scheme when they need something that fits a layout. If you skip colors, you're invisible to a whole category of search behavior.
Singular vs. plural: pick one
Use singular form as the default. Agencies search "dog" and return photos tagged "dog" — their systems handle plural variations. Tagging both "dog" and "dogs" wastes two slots on the same concept.
A few compound terms read more naturally in plural form ("vegetables" over "vegetable," "mountains" over "mountain" when there's clearly a range). Use judgment. But singular is the default, not plural.
Specific beats general, except when it doesn't
"Labrador retriever" is more useful than "dog." "Software engineer" is more useful than "person." Specific terms pull in buyers who know exactly what they need and are ready to license something.
But don't only use specific terms. A buyer searching "dog" shouldn't miss your labrador retriever photo. Include both the specific term and at least one general category term — the specific one catches the targeted search, the general one catches the browsing one.
What not to do
Keyword stuffing with irrelevant terms. Tagging a photo of an empty park with "city," "business," or "family" is against agency terms and can flag your account. Beyond that, it just doesn't work. Buyers who land on a mismatched image keep scrolling.
Over-relying on synonyms. "Happy," "joyful," "cheerful," and "smiling" aren't quite the same thing, and using all four is fine. But adding "glad," "content," "pleased," and "delighted" on top of that is diminishing returns. Stop when you've covered the actual concepts.
Skipping the obvious. The coffee cup in the corner of the frame still needs the keyword "coffee." Don't assume the agency figures it out. It won't.
Using AI to speed up keywording
Manual keywording at scale is a real time problem. At 10 minutes per photo, 50 photos a week is over 8 hours of metadata work before you've written a single title or description.
AI tools like Picseta read your photos visually and generate a full keyword set automatically — including feeling tags and color terms, which are exactly what people skip when they're tired and just want to get through the pile. The workflow shifts from building from scratch to reviewing what came back. For most photos, that's under a minute per image.
See how it compares to doing it manually in our AI metadata guide.
A quick pre-submit check
Before clicking submit:
- At least 20 keywords for anything with more than one subject
- Subject terms, both specific and general
- At least 2–3 feeling/mood keywords
- At least 1–2 color keywords
- Singular form for most terms
- Nothing that isn't actually in the image
Keywording well manually takes 5–10 minutes per photo. Work through these categories a few times and the categories themselves become automatic. With AI handling the first pass, it's mostly a quick scan to catch anything off.
How many keywords should a stock photo have?
Aim for 30–50 relevant keywords. Shutterstock's limit is 50, and staying close to that without padding gives your photos the best search coverage. Under 15 and you're leaving real discoverability on the table.
Should stock photo keywords be singular or plural?
Singular form as the default. Most stock agencies handle pluralization automatically, so "dog" surfaces for searches on "dogs." Using both "dog" and "dogs" wastes a keyword slot.
What are feeling and color keywords for stock photos?
Feeling keywords describe the emotional tone — "peaceful," "energetic," "tense." Color keywords name the dominant colors visible in the image. Buyers often search by mood or color scheme rather than just subject, so these two categories matter more than most contributors realize. They're also the first ones people skip when keywording quickly.