Why Your Rekordbox Library Didn't Fully Match

You just dragged your Rekordbox XML into Mixgraph, waited for the upload to churn through, and the number staring back at you isn't the one you expected. A chunk of your tracks matched straight away. A chunk didn't. Your first instinct is that something broke — that the library you've spent years building, crate by crate, only got partially recognised by whatever's on the other end.

Nothing broke. An unmatched track means one specific thing: that exact release, with that exact title and artist, isn't sitting in Mixgraph's catalogue yet — at least not under the tags you have it filed as. That's a normal outcome, not an error, and if you read it right it actually tells you something useful about your own collection.

What matching is actually checking for

When your library imports, each track gets checked against the catalogue on title, artist, and mix version together — not just the filename sitting in your Rekordbox folder. Rekordbox filenames are messy by nature: track numbers stuck on the front, bracketed tags trailing off the end, ID3 metadata that got overwritten by three different pieces of software over the years. A file called “03 Artist - Track Name (Extended Mix) [Label].mp3” and a catalogue entry called “Track Name” by the same artist are obviously the same track to you. Matching has to work that out cleanly before it can say yes.

Mix version is the part people underestimate. The Original Mix, the Extended Mix, a specific remix, and a radio edit are treated as genuinely different tracks, because they often play differently — different intro length, different arrangement, sometimes a different key. If your copy is tagged in a way that doesn't cleanly map onto what a release is actually called, that ambiguity is enough to hold a match back rather than guess and risk pairing you with the wrong version.

What an unmatched track still gets: BPM and key scoring

Here's the part worth knowing before you worry any further: an unmatched track isn't set aside somewhere useless. It stays in your library as a first-class entry, sitting right alongside your matched tracks, and it still gets scored — just on BPM and key rather than the full chemistry read a fully matched track gets. That's enough to answer the question you actually need answered most of the time: will this mix cleanly into what comes next.

Open your library after an import and you can browse and filter matched and unmatched tracks separately, so you always know which read you're working with for a given track. And if you want to sanity-check a specific transition into or out of an unmatched track by hand — say you're not sure whether it'll sit well against a track you know well — Score My Transition will run the same BPM and key check on any two tracks you give it, matched or not.

Why matching stalls on rare edits, white labels, and deep cuts

The tracks that don't match tend to fall into a recognisable handful of categories, and none of them are your fault. Unofficial edits and bootlegs that never had a proper release. White labels and promo-only pressings that were only ever meant for a handful of DJ boxes. VIP mixes and dubs that got passed around at gigs or on forums but never landed on a streaming or store platform under a clean title. Regional exclusives that only ever released through one label in one territory. Older vinyl-only cuts you own physically or ripped years ago, from before half of today's discovery platforms existed.

What all of those share is that there's no clean official release for matching to hook onto. A catalogue built from real, discoverable releases can only match what actually exists as a release somewhere — it can't reverse-engineer a title and artist out of a bootleg that was never formally credited in the first place. That's not a gap in how matching works. It's a gap in what's publishable to begin with.

Reading your unmatch rate as a signal about your collection

This is the reframe worth sitting with: a high unmatch rate isn't a mark against your library, it's a mark of how deep it goes. DJs who've been buried in one scene for years — trance heads with crates of early-2000s imports, techno collectors chasing white labels from small European labels, anyone who buys direct from artists at gigs — tend to see more unmatched tracks than someone with a broadly mainstream house collection. That's not because their library is disorganised. It's because they've gone further into the edges of a genre than any general catalogue currently reaches.

If your unmatch rate looks high, don't read it as “my library confused the importer.” Read it as a rough map of where your collection has genuine depth — the corners of your scene that a mainstream catalogue hasn't caught up to yet. That's worth knowing about yourself as a collector, not something to explain away.

Deciding which unmatched tracks are worth chasing

You don't need to chase every unmatched track down, and you shouldn't try — that's a good way to spend an evening on tracks you'd never actually notice missing. Be selective about which ones you push on. The tracks worth chasing are the ones you reach for constantly: the anchor tracks you build sets around, the ones you'd be genuinely annoyed to plan a night without full chemistry on. If a track only shows up once or twice a year in your sets, BPM and key scoring is plenty — spend your effort elsewhere.

A quick gut check works well here: if you can picture the last three sets you built and one unmatched track kept showing up in your head as a “where does this go” moment, that's your shortlist. Everything else can sit quietly on basic scoring without costing you anything.

Submitting a track request for the ones that matter

For the tracks that make your shortlist, Track Requests is the direct next step. Search for the release rather than typing a loose free-text guess — picking the actual result means Mixgraph is chasing the exact version you have, mix and all, instead of a title that might resolve to the wrong edit. Once a request gets picked up and lands in the catalogue, that track upgrades from BPM-and-key scoring to a full chemistry read automatically — you don't need to re-import anything or do a second pass on your library.

It won't catch everything. Genuine one-off bootlegs and forum-only dubs may never clear that bar, and that's fine — they were never going to be catalogue tracks in the way an EP release is. But the deep cuts and rarer official releases that are simply missing rather than unreleasable are exactly what requests are for.

Building sets with a mixed library

None of this should slow down how you actually build a set. Flow Builder treats matched and unmatched tracks as equals — both drop into the same flow, sit in the same slots, and get scored against their neighbours using whichever read they've got. A flow that opens on three unmatched deep cuts and closes on five fully matched catalogue tracks is a completely normal flow, not a compromise.

Treat your unmatch rate as information rather than a problem to solve in one sitting. Chase the anchors, leave the rest on basic scoring, and keep building. If you want the wider picture of how to shape a full set once your library's sorted — energy, order, harmonic movement, all of it — the full set-planning guide picks up from here.