Fuzzy Match Tool - Feeling Fuzzy? Match Unmapped Data to your Catalogue

A common task in the music royalties world is trying to map lots of values from your distributors, DSPs, collection societies, partners, etc, to your catalogue. This often requires a manual, time-consuming process. Introducing: our free Fuzzy Match Tool.

Upload two files to find similar names or titles - even when they're not identical. Our tool gives you match scores, flags exact matches, and makes mapping easier than ever.

You have the option to paste your relevant data into the templates provided by the tool or - even better - if you are a Curve client, you can simply take a Sales/Income DB export from Curve for the unmapped data and a Catalogue page export for your catalogue data. Then, sit back, put the kettle on, and let the tool do the rest.

Step 1: Uploading Files

The first thing we do is upload two files.

• File One is your unmapped data—the messy stuff you’ve received from your DSPs or PROs. You can either use our template or, if you’re a Curve client, export invalid lines from Sales DB or Income DB and drop them right in.
• File Two is your catalogue data. Again, you can use the template or, for Curve clients, export your catalogue with titles, creator names, and identifiers like ISRCs, UPCs, or other main work identifiers in Track Separated, Release Separated or Works Export formats respectively.

Once both files are uploaded, we move on to configuration.

Step 2: Configuring the Tool

Here’s where you decide how the tool processes your data:

• There’s a threshold slider that sets how strict the matching should be. At 100%, only perfect matches show up. Lower it, and you’ll see more matches, including some that might need review. Usually, above 75% is a safe range.
• Next is the score weightage slider. This lets you prioritize title matches versus creator name matches. For sources like BMI, where a creator name might be missing, you can slide it all the way to title-only matching.
• If you're mapping works, you can add a single source across all lines, or upload a Curve sales file export to automatically match sources. This would enable you to export in Curve's Society Identifier format.
• You can leave the default settings if you like and just click "I’m Feeling Fuzzy" to start processing.

Step 3: Reviewing and Approving Matches

The tool immediately starts showing results in the table—you don’t need to wait for everything to finish. Here's what the results table contains:

• The results table shows the unmapped title in column 1, the catalogue match it found in column 2, and a similarity score. Higher the score, the better the match.
• You can hover over these entries to see identifiers and copy them if needed.
• There’s a net value column to help prioritize high-impact lines.
• Actions include approving, rejecting, picking alternate matches, or manually searching for matches.
• You can also filter by value in the search bar, sort by net value, and even auto-approve perfect matches as they come in.
• The noise filter lets you ignore common noise phrases or add your own.

Step 4: Magic Wand API Cleaning

The Magic Wand is a powerful tool that cleans up titles and names automatically. It uses YouTube, Spotify, and MusicBrainz APIs (for ISWC) to fetch accurate data based on identifiers. This is great for fixing garbled, incomplete or unmatchable rows.

Step 5: Downloading and Sharing Progress

Once you’ve approved matches:

• You can download your progress at any time in a CSV format to share with your team members or send for approval. This file can be loaded back into the tool and the session restores back to where it was downloaded.
• You can at this point, add another catalogue data file and rerun fuzzy matching on the pending lines.
• Once you are happy with the matching progress, you can download as a standard all results CSV or as an Alias or a Society Identifier Curve import (source is required for latter).

Some smart features:

• If a creator name is missing and there are identical titles in the catalogue, the tool warns you that this might not be a perfect match.
• Alternate matches are presented as a list to make decision-making easier.
• You can view unmatched rows separately. From there, you can force map, retry with a lower threshold, or use the Magic Wand.
• Undo your last action, if needed.
• View approved, rejected, pending, or unmatched lines with filters.
• The FAQ contains helpful info, and there’s a send feedback button for direct input.

And that’s the whole workflow: upload your files, configure the processing, review and approve matches, use the Magic Wand for tricky rows, assign sources if needed, and finally download or share your progress.

Now let’s jump in and see it in action!

Visit the tool at https://fuzzy.curveroyaltysystems.com