Keywords Getty and iStock will actually keep

Start with the part that decides everything else: you do not have to learn Getty's vocabulary. Open MetaBrain, drag your files in, wait. The matching happens on the way to the export file, and there is nothing to configure, no list to import and nothing to read first.

That matters because Getty Images and iStock work differently from every other agency you submit to.

The part nobody warns you about

Shutterstock, Adobe Stock and the rest take your keywords as you wrote them. Getty and iStock check every one of them against a controlled vocabulary - a fixed list of approved terms.

Nothing disappears quietly. It is worse than that: ESP shows you exactly what it could not match, in red, with a banner that says "Keywords in red are not found. Click for suggestions, remove, or recommend new keywords". Click a red chip and it turns into an editable field with live suggestions, and you resolve it by hand. One chip at a time.

On a real file of ours, 47 keywords came back as 21 accepted and 25 red. More than half of a finished, carefully keyworded photo turned into a to-do list.

The reasons are rarely dramatic. A plural where their list holds the singular. A phrase that is perfectly normal on other agencies and simply has no entry there. A word that means several things, waiting for you to say which one you meant. Multiply that by a batch of fifty files and you have lost an evening to clicking, not to shooting.

What MetaBrain does now

Before your Getty or iStock file is written, MetaBrain matches your keywords against the vocabulary those agencies accept, and writes them in the wording the agency expects rather than the wording you typed.

Three things happen in that step:

Your phrasing is translated, not thrown away. The idea you meant is preserved and expressed in their terms.

Ambiguous words are resolved by the picture. A single word can mean several different things in a controlled vocabulary, and the wrong reading is worse than no keyword at all. The choice is made from what the image is actually about.

Terms with no entry are left out here, not there. If a word has nowhere to land, it does not travel to ESP to be discarded in silence. You see a clean, honest set instead of a hopeful one.

One vocabulary serves both Getty and iStock, so the same file is ready for either. It works the same way in the desktop app and in Studio.

The point of all of it is the red list. Every keyword we can resolve before export is a chip you do not have to click, read and decide on at two in the morning.

How this compares

Here is the part where we could have claimed to be first and would have been wrong.

Vocabulary matching is not unique to us, and the tools that do it are good at it.

PhotoKeyworder.ai is the closest to this specific job. It generates titles, descriptions and keywords for photos, EPS vectors and video, says plainly that its metadata is built to be compatible with the controlled vocabulary, and ships ESP-compatible and DeepMeta-ready CSVs alongside metadata written straight into JPG, PNG and EPS. It handles batches of up to 500 files and its volume plans go a long way down in price per file. If Getty and iStock are your whole business, it is a serious tool built by people who clearly submit there themselves.

PixTagger does the same mapping for photos and video, runs in the browser and prices in credits. CyberStock advertises vocabulary compliance in its suggestions too.

So the honest summary is: this capability is expected now among the tools that take Getty seriously, and we keep building ours out release after release.

What is actually different here is what surrounds the step. Vocabulary matching sits inside one pipeline that also writes music metadata - measured tempo, key and loudness, cuts, watermarked previews - which none of the tools above do. The finished files go from the same app straight to the agencies over FTP or SFTP, so delivery is not a second subscription. The wording you keep correcting is learned and stops coming back. And there is a free tier of thirty files with no card, which is a cheaper way to find out the truth than any comparison table, including this one.

Where the others are ahead, plainly: PhotoKeyworder takes bigger batches in one go and costs less per file at high volume, a browser-only tool has nothing to install, and a single-purpose tool is simply easier to understand. Those are real advantages and they do not go away because we would prefer otherwise.

What we do not promise

Their vocabulary is theirs. It changes on their side, and a rare or very new term can still have no entry at all. We do not promise a perfect match on every keyword, and any tool that does is describing a list it does not control.

What we do promise is the part we own: your keywords are checked against what those agencies accept before the file leaves, and what cannot survive there is not quietly sent to die.

The practical difference

Nothing about your routine changes. You drop files, you get results, you export for Getty or iStock as you always did.

The difference shows up later, in the only place that counts: the file arrives with tags the agency can actually index, and it can be found by people looking for exactly what you shot.