Writing alt text for images you didn’t take or choose
A designer drops six photos into the shared folder at 4:40 on a Friday. A client emails a “brand image” with no context beyond the filename. Someone on the team wants to repost a customer’s photo by Monday. In every case you’re the one writing the alt text, and you weren’t there when the picture was taken.
That describes most alt text. The person writing the description is rarely the person who shot the image or picked it. Here is how to do it well from the outside.
Describe what’s visible, not what you assume
The temptation with an unfamiliar photo is to fill the gaps with plausible detail. Resist it. If you don’t know that the two people in the frame are a nurse and a patient, they are two people — one in scrubs, one in a hospital gown. If you can’t tell whether the building is the main hall or the annex, it is a brick building with an arched entrance.
Guessing wrong is worse than being general. A sighted reader can quietly correct a mislabeled caption against the image in front of them. A screen reader user gets your guess and nothing else.
Ask the person who chose it what it’s for
The fastest route to good alt text for an image you didn’t select is one question to whoever did: what is this picture doing in this post?
The answer usually rewrites the description. “A photo of the new lobby” and “proof the renovation finished on time” are the same JPG and two different sentences. Purpose is the part you can’t read off the pixels, and it lives in the head of the person who picked the file.
Reposts and customer photos
User-generated content adds a second rule: don’t promote the original caption into your description. If a customer wrote “finally got my hands on these,” that’s their voice, not a description of anything. Write what is in the frame — the product, the setting, the action — and keep credit and handles in the caption, where sighted readers see them too.
The same restraint applies to names. If the original post doesn’t identify the people in it, yours shouldn’t either. “Two people holding up a wrapped package” is honest; inventing a first name is not.
Make the description travel with the file
The durable fix is upstream. Since 2021 the IPTC photo metadata standard has included an Alt Text (Accessibility) property — a real embedded field, capped at 250 characters, that moves with the image into whatever system opens it next. A growing number of digital asset managers read and write it. If you commission photography or license stock at any volume, asking for that field to be filled costs one line in a brief and saves the guessing later.
Watch the other end of the handoff too. In WordPress, alt text set in the Media Library is copied into the post at the moment you insert the image — editing it in the library afterward does nothing to posts that are already published. Whoever writes the description has to write it before insertion, or fix each post by hand.
None of this requires having been behind the camera. It requires describing honestly, asking one question of the person who chose the shot, and putting the answer somewhere it won’t get lost between the folder and the feed.
SnipAlt writes publish-ready alt text for anything on your screen — snip, paste, done. Start with a 7-day free trial — card required, cancel anytime.
Start free trial