How to track what DJs are playing

Each method trades coverage against effort, and the right choice depends on which DJs you follow, how often they post, and whether you want a one-time answer or an ongoing record. No single method wins on every dimension, so most serious crate-diggers end up using two: a database for the headline-DJ archive and an automated tool for the working DJs whose sets never get logged.

The sections below examine each method in detail so you can make an informed choice for your own roster of DJs.

Before comparing methods it helps to be precise about what you are trying to do, because the word “tracking” means two quite different things depending on the context.

The first meaning is reactive and one-time: you saw a specific post, you want to know the name of a specific track, and once you have the answer you are done. That is a lookup task. Any of the four methods can handle it, and the best choice for a single post is usually to read the caption first, check the comments second, and reach for audio recognition only if those come up empty. The flagship guide on how to find what a DJ is playing on Instagram covers this step-by-step process in detail.

The second meaning is proactive and ongoing: you follow a roster of DJs for the records they break, and you want every new post they put up to be identified automatically. This is a different problem. A lookup tool answers a question you already thought to ask; an ongoing tracker builds a running record whether you remember to check or not. Most of this guide is about the second problem, because that is where the methods diverge most sharply in coverage and effort.

The distinction also changes what “good coverage” means. For a one-time lookup, coverage means: does the answer exist somewhere I can find it? For ongoing tracking, coverage means: what fraction of a DJ's posts get identified, over the long run, without me doing work on each one? A manual database can have great coverage for one famous set and zero coverage for the hundred Instagram clips that same DJ posted this month.

Sites like 1001Tracklists are the starting point for most crate-diggers. The model is straightforward: community members type in tracklists from sets they attend, record, or stream, and the database accumulates them over time. For the right kind of set the coverage is genuinely impressive: you can find a complete 90-minute festival set with time-codes, format tags (vinyl / CDJ / USB), and comment threads debating the IDs.

Where the model breaks down is the manual-submission gap. Every tracklist in the database exists because a person chose to log it. That creates three structural biases:

• Headliner bias. Big festival slots and syndicated radio shows get logged because many fans care. A local or up-and-coming DJ rarely has fans dedicated enough to transcribe sets, so their entries are sparse or absent.
• Short-form blind spot. The 30-to-60-second Instagram clip where most DJs break new records today is almost never logged. It is too short to call a “set” and too fleeting to attract a transcriber. This is the biggest gap for active crate-digging.
• Lag and incomplete entries. Even logged sets often carry unresolved ID placeholders for tracks the submitter could not name. A set you want to reference might be 80% complete, which is useful but not complete.

The verdict: an indispensable archive for the DJs who get logged, and nearly useless for the long tail. For a deeper comparison of what a manual database offers versus an automated alternative, see the 1001Tracklists alternative guide.

The second method takes a different angle: instead of relying on a human to log a set, it runs audio recognition automatically over a full mix recording. Tools in this category such as TrackSniff let you paste a URL or upload an audio file, then return a timed tracklist built from fingerprint matches. For a long mix you already have, this is genuinely powerful: it processes the whole recording in minutes and surfaces every released track it can fingerprint, with timestamps.

The constraint is the per-mix model. You supply one audio file, you get one tracklist. To keep up with a DJ who posts three Instagram clips a week, you would need to manually retrieve each clip, feed it to the tool, and review the output — repeated indefinitely. That is a lot of work to automate what should be automatic. The other limit is the same one that constrains consumer Shazam: heavily blended transitions, pitch-shifted audio, and unreleased tracks have no reference fingerprint to match, so they return no result even when the audio is clean.

Per-mix fingerprinting wins when: you have a single long mix in audio form, you want a timestamped tracklist quickly, and you do not need it to repeat forever. It is a great one-shot tool and a poor ongoing one. For a head-to-head breakdown of the per-mix model versus continuous auto-tracking, see the TrackSniff alternative guide.

The simplest method is also the oldest: open the post, read what the DJ wrote in the caption, scroll the comments, and look for a fan who typed the track name. No tools, no account, no cost. For a DJ who consistently labels their tracks or has an active fan community, this is the fastest path to an answer.

The problem is cost at scale. A caption check on one post takes 30 seconds. Checking 50 posts across 10 DJs once a week takes hours — and you still have to remember to do it. When a DJ's captions are silent and the comment section has not solved it yet, you are left with nothing and have to return later. There is also a signal-quality problem: a confident-sounding comment is not the same as a confirmed ID, and a wrong guess that gets liked can stick to a record and spread.

For the full ranking of caption and comment signals by reliability, and a step-by-step process for IDing a single post, see the guide to finding what a DJ is playing on Instagram. Manual caption-checking is the right method for one or two DJs you follow closely and have time to watch. It does not scale to a roster.

The fourth method closes the gaps in the other three by automating the checks on every new post rather than waiting for a human to log it, supply audio, or remember to look. CrateWire is built around this category: you follow a DJ, and for each new public Instagram post it runs a multi-signal waterfall automatically.

The waterfall works in order of confidence. It checks Instagram's own music metadata first, since an IG-attributed track is already named and needs no inference. Then it parses the caption for an explicit ID label or an artist-title pattern. Then it scans the comments for written IDs, weighting the DJ's own replies above unverified fan guesses. Finally, if text signals are empty or ambiguous, it fingerprints the audio against a reference catalog. Each signal that names the same track adds to a confidence score; the track surfaces only once the blended score passes a threshold, so a borderline single-source match does not masquerade as a confirmed ID.

The output is a running tracklist per DJ — every identified track with its confidence score — and it updates every time a new post goes up. An unreleased ID that exists nowhere today will resolve the moment a caption is added, a fan comments the name, or the track hits a catalog and the fingerprint matches. You can browse the public DJ tracklists to see the method's output before signing up.

Honest limits apply: a brand-new unreleased ID with no reference anywhere cannot be identified by any tool until a signal appears. What the automated approach removes is the manual labor of re-checking, and it catches the answer the moment it becomes available, rather than the next time you remember to look.

A single DJ's tracklist tells you what that DJ has been playing. Following ten DJs gives you ten separate lists. What neither gives you, without manual aggregation, is an answer to the question working DJs and A&R people care about most: which unreleased track are multiple DJs playing right now?

When a forthcoming track starts spreading through the DJ circuit, it tends to show up across several sets within a short window before the release or announcement. In the vinyl era you spotted this by watching who was importing the same acetate. Today it happens in Instagram posts and short clips, where the captions say ID or name a promo that most of your audience has not heard yet. If you are only looking at one DJ at a time, you see the clip and move on. If you are aggregating across many, the pattern jumps out.

This is why cross-DJ aggregation changes the value of tracking. When CrateWire runs the same waterfall across all the DJs you follow, it can surface tracks that appear across multiple DJs within the same week — before they reach streaming platforms, before they trend on social, and often before most listeners have heard them. A cross-DJ trending feed built this way turns individual tracklists into an early-signal network. You can get a preview of what this looks like on the public DJ tracklists hub.

The right method depends on two variables: what kind of DJ you are tracking, and what your goal is. Here is a quick decision matrix to cut to the right answer without working through all four methods every time.

For most working crate-diggers the answer is two tools in parallel: a manual database for the famous DJs with documented archives, and an automated tracker for the rest of the roster where manual coverage does not exist. The automated tool does not replace the database for its strong cases; it covers the long tail the database never touches.