About
Built by a stock contributor, for stock contributors
Metadata World is an independent toolbox made by Saqlain Noorani, a working microstock contributor who got tired of losing entire evenings to the unpaid part of the job. The photography and the editing were never the bottleneck; the bottleneck was everything that happens between exporting a batch and pressing submit.
Why this exists
A single shoot can easily produce two hundred keepers. Each one needs a title that reads like a sentence rather than a keyword dump, a description that survives review, somewhere between twenty-five and fifty ranked keywords, the right category, and a CSV row in the exact column order each marketplace expects. Adobe Stock wants one format, Shutterstock wants another, and both reject batches for reasons that only become obvious after the rejection email arrives — trademarks in a title, repeated phrases, non-English characters, a video that is technically sharp but 23.94 fps instead of an accepted frame rate.
Doing that manually is slow and inconsistent. Doing it with a generic chat assistant is faster but produces metadata that ignores marketplace limits and gets flagged. So the first version of this site was a private script for one contributor's own submissions. Everything since then has been shaped by real rejections: each new check exists because a batch failed review and the reason was worth encoding permanently.
What the toolbox does
The metadata studio is the centre of it: drop in photos, vectors or clips, generate titles, descriptions, ranked keywords and categories with your own AI key, review every row, then export a CSV that matches the target marketplace column-for-column. A compliance pass runs before export and blocks the things reviewers reject most — brand names, duplicated phrases, stray special characters, over-length titles.
Around it sits the rest of the toolbox, each piece covering a different part of the submission routine: a spec checker for resolution, DPI and codec rules; a duplicate scanner that clusters near-identical frames so you submit the best of a series instead of the whole burst; a metadata embedder that writes titles and keywords into the files themselves as XMP; an approval sorter that scores technical quality and commercial viability before you waste a review slot; an earnings analyser for the CSVs marketplaces hand back; a batch renamer that keeps filenames in step with CSV rows; converters, audio strippers and prompt tools for the odd jobs in between.
How it is built
Almost everything runs inside your browser. Image conversion, hashing, duplicate detection, spec checks, XMP writing and CSV parsing never leave your machine. When AI is genuinely needed, you bring your own provider key, so you pay the provider directly at cost and no plan tier stands between you and your own files. Original media is never stored on our side — compressed previews or keyframes are relayed to the provider you picked, and nothing else.
No account is required to use any tool. Projects and settings live in your browser's own storage; signing in with Google is optional and only lets keys and project metadata be backed up, encrypted, and restored on another device. You can read the details on the privacy page.
Honest limits
No tool can promise acceptance. Reviewers make judgement calls, marketplace rules change, and AI-written metadata still needs a human read before submission — which is why every generated row stays editable and nothing is exported without your click. Metadata World is an independent tool and is not affiliated with or endorsed by Adobe or Shutterstock.
If something is wrong, missing or rejected in a way the tools did not catch, that feedback is the most useful thing you can send. The contact page reaches the person who builds it.