Auditing Your Assets: How to Run a Quick "Health Check" on Your Music Rights Data

Your catalog is only as valuable as the data behind it.

It's a truth that gets quietly ignored until something goes wrong…a royalty payment that never arrives, a sync placement that stalls because the rights information doesn't stack up, or a licensing deal that takes three times longer to close because nobody can agree on who owns what.

Poor music rights data isn't a niche problem. It affects independent publishers and major catalog owners alike, and in an industry where revenue is increasingly tied to the accuracy of metadata, a messy spreadsheet isn't just an admin headache, it's a commercial liability.

The good news is that a data "health check" doesn't have to mean a months-long audit. Small, consistent habits and the right systems in place can prevent the kind of decay that costs rights holders real money.

We sat down with three members of the Synchtank team to get their honest take on what bad data actually looks like in the wild, what it costs, and exactly where to start if your catalog needs attention.

---

Before we dive in, let's meet our contributors:

	Janet Landrum Royalties Specialist at Synchtank, with a background spanning several years at BMI in Los Angeles and over 20 years as Executive Director of Royalties at Twentieth Century Fox.
	Paul s'Jacob Product Manager at Synchtank, combining new product features and customer requests to continuously improve the platform. Prior to Synchtank, he worked for record labels, music publishers and collection societies.
	Rachael Walsh Media Asset Manager at Synchtank, focused on onboarding, auditing and maintaining large-scale music catalogs. Before Synchtank, she worked in sports broadcast media operations at DAZN managing high-volume, time-sensitive media assets.

"Having metadata" doesn't mean the data is healthy

The most common misconception Rachael encounters when onboarding a new catalog? That metadata existing at all means it's in good shape. "One of the biggest misconceptions is that 'having metadata' means the data is healthy. In reality, a catalog can look complete at first glance but still be extremely difficult to ingest cleanly or search properly."

"One of the biggest misconceptions is that 'having metadata' means the data is healthy. In reality, a catalog can look complete at first glance but still be extremely difficult to ingest cleanly or search properly." Rachael Walsh

The issues she finds are remarkably consistent across companies of all sizes. "The biggest recurring issue is inconsistency. Metadata might technically exist, but it hasn't usually been structured consistently over time. You'll often see:

• Different naming conventions across the same catalog

• Multiple delimiters being used interchangeably

• Missing or incomplete contributor data

• Broken parent/child relationships between tracks and versions

• Encoding issues causing corrupted characters

• Duplicate or conflicting IDs

• Legacy spreadsheets that have been edited by multiple teams over many years"

Scale makes all of it worse. "Something as simple as inconsistent separators or extra spaces can massively affect linking, search accuracy and automation downstream." The root causes, she adds, tend to be the same regardless of company size:

• Lack of standardisation over time

• Multiple teams contributing to the same data

• Catalog growth happening faster than governance processes

• Older metadata structures not evolving alongside newer delivery requirements

That's why onboarding is rarely just a technical ingest — it's usually a much wider data health exercise as well.

Paul sees the warning signs emerge in a very specific way from the product side. "Everyone in the company will know that there are data gaps as they try to carry out their daily work. When reports are pulled that need caveating or cleaning up because a vital piece of data is missing, when licences get delayed because the data is incomplete, when royalties aren't collected or able to be paid through — that's when you know you've got data issues."

The royalties problem hiding in plain sight

For Janet, the data problem she sees most consistently is one that plays out in the matching process, and its consequences are felt directly in what rights holders actually get paid.

"Metadata is what is keeping many administration teams from peak efficiency. This mainly affects the matching process with the million + line item source statements. Keeping up with that metadata matching, both on the royalty and registration side, keeps PROs and CMOs from identifying and paying out on works and keeps internal data teams from being able to process these statements in a timely matter."

When master owners and publishers are working from misaligned records, the untangling process usually starts with universal identifiers, but those come with their own complications. "Both teams rely on universal identifiers to try to keep this information straight. A sound recording's ISRC matches up with a composition's ISWC, but there are issues with both standardized numbers. Occasionally multiple ISWC's are issued for a composition, same can happen on the ISRC side. Where there should be a 1:1 ISWC to ISRC relationship, it doesn't always work out that way. A good publisher does their best to keep on top of their records at all CMO and PRO outlets and runs audits looking for outliers and mismatches and multiple registrations (resulting in duplicated and often incorrect submissions)."

"Current systems need to be able to adapt and utilize multiple metadata combinations in order to properly match works for royalty obligations. Often, territory, received share, catalog number, recording/ ISRC and others can affect how much you pay out and to whom." Janet Landrum

She's seen the consequences up close. "At one company I worked at we had a difficult copyright wherein we collected all royalties, but were required to pay out different versions (catalog numbers/ ISRC's) to varied third parties. The lack of information provided by collection agencies made that job very difficult. Not to mention that the royalty system we used to ingest those statements was only using the source's work identifier to match. Current systems need to be able to adapt and utilize multiple metadata combinations in order to properly match works for royalty obligations. Often, territory, received share, catalog number, recording/ ISRC and others can affect how much you pay out and to whom."

Where to start when you only have an afternoon

If clean data feels like a mountain, Rachael's advice is to stop looking at the mountain and find the recurring pattern instead.

"The best quick wins are usually the ones that improve consistency immediately without requiring a complete catalog overhaul. If someone only had an afternoon, I'd recommend focusing on the foundational cleanup areas that tend to have the biggest operational impact:

• Standardising separators and formatting

• Removing duplicate spaces and trailing punctuation

• Checking contributor formatting consistency

• Reviewing parent/child relationships for alternate versions and cutdowns

• Identifying truncated metadata fields caused by legacy exports or spreadsheet limitations

Even relatively small cleanup exercises can dramatically improve ingest success rates, searchability and reporting."

Another quick win is pattern recognition. "If you can find one recurring issue affecting thousands of rows, you've potentially solved a major operational problem in one pass."

The mindset shift matters as much as the method. "I'd also encourage catalog managers not to aim for perfection immediately. The goal is really to create structure and consistency first. Once the data becomes predictable, automation and validation tools become far more effective. A healthy catalog is rarely created through one giant cleanup project — it's usually the result of lots of smaller, repeatable improvements over time."

"A good music administrator with a few hundred tracks will know each one in depth. The trick is knowing how to scale that in-depth knowledge for thousands of songs, and that needs to be through well curated metadata." Paul s'Jacob

Paul frames the scaling challenge similarly. "A good music administrator with a few hundred tracks will know each one in depth. The trick is knowing how to scale that in-depth knowledge for thousands of songs, and that needs to be through well curated metadata. As with many things, expanding rapidly is often done at speed and sometimes quality suffers as a result. Handling thousands of tracks requires quick and simple identification of data issues and bulk methods to fix them. Due diligence and data review prior to taking on a catalog can help, or at least planning for the required clean-up after ingestion."

What automation can do — and what it can't

Platforms can process at scale. But Rachael is clear about where human judgement stays essential.

"Automation is incredibly powerful for identifying patterns, validating formatting rules and flagging anomalies at scale. But human judgement is still essential when it comes to context, intent and commercial understanding. For example, an automated system might correctly identify that two track titles don't match exactly, but a human still needs to determine whether that difference is intentional, meaningful or simply an error."

The same applies across a range of decisions that look simple on the surface:

• Parent/child version relationships

• Composer and publisher splits

• Genre and mood interpretation

• Duplicate detection

• Rights ownership conflicts

• Catalog hierarchy decisions

• Naming conventions that make sense commercially rather than just technically

“Humans also understand nuance. A platform can tell you something is inconsistent, but it can't always tell you whether changing it would create bigger downstream problems."

There's also the human layer that systems simply can't replicate. "I also think communication is a hugely overlooked part of metadata operations. Platforms can process data, but people still need to interpret client intent, make judgement calls and understand how different teams actually use that data operationally."

From a product perspective, Paul explains how Synchtank is built to make that human oversight easier rather than bypass it. "Synchtank includes duplicate detection that surfaces potential duplicate records before they become a problem. When new tracks or albums have been added, the system cross-references existing catalog data and flags likely matches for review — meaning a catalog manager can catch and resolve duplicates at the point of entry rather than discovering them after the fact during a client delivery or licensing request. For larger catalogs, filtered views and reporting tools let managers proactively audit for duplication across specific fields like ISRC, title, or artist name."

Validation is baked in at entry, too. "Some field-level validation rules are enforced at the point of data entry, so the system won't accept incorrectly formatted values — for example, an ISRC in the wrong format, a missing mandatory field, or a date entered outside an expected range. For catalog managers, this means errors are caught immediately rather than propagating silently through the system. We also allow users to define mandatory fields required for metadata delivery outside the system, meaning you're not only keeping good metadata within your own system but also propagating good metadata to your partners."

When correction is needed at scale: "Where a catalog manager needs to apply the same change across hundreds or thousands of records — correcting a composer credit, updating a publisher name, or adding a missing genre tag — Synchtank's bulk update tools allow this to be done in a single operation rather than record by record. Updates can typically be scoped by filter first, so managers can review exactly which records will be affected before committing the change, reducing the risk of unintended edits at scale."

"We're also building tools to integrate with third-party data sources — such as rights databases or delivery partners — to pull in reference data that can be compared against what's held in the catalog." Paul s'Jacob

Paul also points to where the platform is heading next. "We're also building tools to integrate with third-party data sources — such as rights databases or delivery partners — to pull in reference data that can be compared against what's held in the catalog. This gives catalog managers a way to verify that key identifiers, credits, or ownership information align with upstream or downstream sources, and to spot discrepancies without having to manually cross-reference external systems. Where gaps or mismatches are found, the relevant records can be flagged for correction directly within the platform."

The commercial cost of getting it wrong

The link between clean data and commercial performance is one Paul thinks gets consistently underestimated.

"Getting one's own metadata in order is the first step to ensuring that all your downstream partners, agents, representatives and societies all have the best data they can have to ensure they can represent your catalog in the most effective manner. Better metadata means quicker identification of repertoire leading to faster approval and licensing and more accurate matching when processing royalties."

Rachael sees the same dynamic from the operational side — and flags how much is at stake as the industry changes. "As AI-generated content increases and catalogs continue growing exponentially, the ability to organise, verify and contextualise assets at scale becomes critical. The challenge won't simply be storing more content — it'll be understanding what that content actually is, where it came from, who owns it and how it should be surfaced."

That means the infrastructure underpinning catalogs needs to evolve too. "Metadata structures will need to become more standardised, more scalable, more validation-focused, more adaptable to new asset types and delivery models. Search, recommendation systems, rights management and AI governance all rely heavily on structured data underneath the surface."

Her background in broadcast reinforced the stakes. "In live sports media environments, asset organisation and accuracy directly affect operational delivery in real time. In music and rights management, the timelines may be different, but the principle is very similar: if the underlying data is unreliable, every downstream process becomes harder to manage."

"The companies investing in clean, structured and well-governed metadata now are going to be in a much stronger position long term — not just operationally, but commercially as well." Rachael Walsh

On preparedness, she's candid. "I think many companies are still catching up. A lot of catalogs were originally built for much smaller ecosystems and weren't designed with today's scale, automation requirements or AI-related complexities in mind. The companies investing in clean, structured and well-governed metadata now are going to be in a much stronger position long term — not just operationally, but commercially as well."

Her conclusion carries a weight that anyone who's been on the wrong end of a data problem will recognise immediately: "Good metadata has quietly become infrastructure. When it's working properly, nobody notices it. But when it isn't, every operational problem downstream becomes exponentially harder to solve."

---

So, what did we learn?

• Start with structure, not perfection: The goal isn't a flawless catalog overnight, it's consistency first. Standardise your separators, fix your contributor formatting, review your parent/child relationships. Once the data becomes predictable, everything else gets easier.

• The matching problem is a metadata problem: Royalties that don't arrive, statements that can't be processed, ISRCs and ISWCs that don't align — almost all of it traces back to data that was never clean to begin with. Get the foundations right and the revenue follows.

• Automation is only as good as the data feeding it: Platforms can flag anomalies at scale, but humans still need to determine whether a mismatch is intentional, meaningful, or simply an error. Don't outsource the judgement calls.

• Catch it early or pay for it later: Whether it's a duplicate record, a missing mandatory field, or an incorrectly formatted ISRC, errors caught at the point of entry cost nothing. The same errors discovered during a client delivery or licensing request cost considerably more.

• Clean metadata is a commercial asset: Better data means faster identification of repertoire, quicker approvals, more accurate royalty matching, and stronger representation by every downstream partner, agent, and society working on your behalf.

Over to you, publishers, labels, and rights holders...you've heard it straight from the people who live inside this data every day. Inconsistent naming conventions, broken parent/child relationships, royalties that fall through the cracks because nobody standardised the identifiers — these aren't niche technical problems. They're commercial liabilities.

Synchtank's Catalog Management tools are built to get ahead of exactly this: field-level validation at the point of entry, duplicate detection before records multiply, bulk update tools that let you correct a composer credit across thousands of tracks in a single operation, and integrations that let you verify your data against upstream and downstream sources without manually cross-referencing external systems. Pair that with Synchtank's Core Platform for end-to-end rights management and monetisation, and you have the one source of truth Rachael, Janet, and Paul are describing.

Consider this your sign to tighten things up.