Find duplicate assets in your DAM.

A digital asset management system should hold one trusted version of each asset. Twinsift finds the duplicates, renditions, crops and edits that got in anyway, and links each one to its original.

• 99.9%of duplicates found
• 96%+of crops traced to their source
• 0files deleted without your approval

Accuracy measured on a test library of 250,995 files.

How duplicates get into a DAM

• A photo is downloaded for a project and uploaded again later under a new name.
• Agencies deliver crops, banners and colour-graded versions as new assets.
• Web renditions are exported, edited and imported as if they were originals.
• Libraries are merged after a migration or an acquisition.

Most DAM systems check for an exact copy at upload. A resized, re-saved or cropped version passes that check and becomes one more asset to tag, search through and license.

What duplicates cost

• Storage and DAM licences priced by volume.
• Search results full of near-identical images, so teams pick the wrong one.
• Rights you can't track. When a licence or model release ends, every crop and banner made from that photo has to go too.
• Slower and more expensive migrations.

What Twinsift does

Twinsift compares what is in the pixels, so it finds copies whatever their name, format or size. Each group gets one keeper, chosen by true resolution, format, bit depth, compression and sharpness. Every other file is labelled a duplicate, rendition, crop, banner or edit, and your rules decide what happens to each kind.

In our tests on a 250,995-file library, Twinsift linked 50,078 of 50,082 duplicates and 96.5% of crops to their original.

Connecting your DAM

If your DAM keeps its files in your own Amazon S3 bucket, Twinsift can read the bucket directly. For any other DAM, we scan an export of the library, and you get a report to act on in the DAM.

Questions