Apple Music Has a Mess Problem. This App Wants to Clean It Up.

The Macro: Apple Music Is a Great Service With a Terrible Library Problem

Apple Music has somewhere north of 100 million tracks. That number gets cited as a feature. For most people who have used the service for more than two years, it is also quietly a curse.

Here is the actual problem. Apple Music pulls from multiple sources simultaneously: the iTunes store catalog, your own uploaded files, iCloud Music Library matches, and purchases dating back to 2003. These sources do not always agree on what constitutes a single album. Abbey Road might live in your library four times across standard, remastered, anniversary, and matched versions. Compilations bleed into artist pages. Live albums share track names with studio versions. The result is a library that grows organically in the worst possible direction.

Apple’s own tools for managing this are, charitably, minimal. The native iOS Music app lets you hide albums, delete downloads, and that is roughly where the toolkit ends. The desktop app gives you more, but not much more. And as Apple focuses its music hardware and software energy on spatial audio, Dolby Atmos integration, and AI-driven composition tools, the unglamorous problem of library hygiene has been left largely to users who just tolerate the mess.

The iOS market context matters here. According to multiple sources, Apple holds dominant U.S. smartphone market share, and iPhone shipments grew 10% year-over-year in 2025, the highest among the top five brands. More iPhones means more Apple Music subscribers means more bloated libraries. The audience for a cleanup tool is not niche.

What is surprising is that the competitive field in this specific category is nearly empty. There are no meaningful direct competitors I can point to for duplicate detection within Apple Music on iOS. Third-party library tools exist for Spotify, for local file management on desktop, and for podcast apps. On iPhone, for Apple Music specifically, the space is genuinely sparse.

The Micro: Four Features, One Honest Value Proposition

Song Sweeper does four things. It finds duplicate songs across different albums. It helps you unify various editions of the same album. It surfaces tracks you no longer play so you can clear them out. And it favorites your most-played songs, which is meant to improve Apple Music’s recommendation engine over time.

That last feature is the quietly interesting one.

Most people do not realize that Apple Music’s recommendations are heavily influenced by which tracks you have marked as favorites. If your favorites list is empty or stale, the algorithm is essentially guessing. Song Sweeper’s pitch is that cleaning your library and improving your recommendations are the same job, and automating the favorites piece closes a feedback loop that most users never even knew was open.

The duplicate detection logic works across album editions, which is the harder problem. Catching two files with identical names is easy. Catching that your standard 2009 remaster and your 2019 anniversary edition are functionally the same record requires something more considered. The app claims to handle this, though without access to the product website I cannot verify exactly how the matching works under the hood.

The interface is iPhone-native and the product is focused entirely on iOS, which is a deliberate constraint. No desktop companion, no web dashboard. You do the cleanup from the same device you listen on.

It got solid traction when it launched, landing in the top five on Product Hunt the day it went live.

The thing I find genuinely appealing about this product is its scope. It is not trying to be a full music management platform. It is not adding social features or AI-generated playlists. It is solving a specific, boring, real problem that Apple has declined to solve. That restraint usually either signals a very focused founder or a very small total opportunity. I am not sure which this is yet.

The Verdict

Song Sweeper is a reasonable solution to a real problem. I do not think it is going to define a category, but I also think the category has been ignored long enough that even a competent entry fills genuine whitespace.

The favoriting feature tied to recommendation improvement is the angle I would push harder if I were advising them. Duplicate cleanup is a one-time action. You do it, your library is clean, you never open the app again. Recommendation improvement is ongoing. That is the stickier use case and the stronger retention argument.

At 30 days, I want to know retention numbers. How many people who ran the initial cleanup returned within a month. At 60 days, I want to know whether the recommendation quality improvement is something users actually notice or just a marketing claim. At 90 days, I want to know how the team is handling the monetization question, because a one-time utility app has a very specific ceiling unless there is a subscription mechanic or a genuinely recurring use case underneath it.

The product I would compare this to is a good calendar cleanup tool. Useful, satisfying to use once, potentially hard to build a business around. The founders will need to answer that question sooner than they probably want to.

---

Also featured on HUGE: Google Wants to Be Your Beat Maker. Sure, Why Not. · Bitrig Lets You Build iPhone Apps From Your iPhone, and the Founders Literally Created SwiftUI · GoRiff Wants to Be Cursor for Music, and the DAW Industry Should Be Nervous