‘We have lagged behind other industries in identifying opportunities to collaborate.’

June 21, 2022

The following MBW op/ed comes from Mark Douglas (pictured), Chief Information Officer (CIO) of UK music licensing company PPL.

This quarter, I want to take a closer look at some key developments that have been taking place around the systems that manage the flow of royalty income from a licensee to record labels and artists. As the CIO at PPL, I will focus on the recordings side of things, but there are many parallels in the musical works side of our industry.

Before getting into some of the specifics, it is worth looking more generally at the sort of systems that are used for this. In some respects, they are quite unusual. Unlike many businesses, very few of our systems are about repetitive transaction processing, such as you would see in, say, a manufacturing or retail environment.

So much of what we do is about data management, trying to turn incomplete or low quality data into something that is good enough – and complete enough – to act as the basis for allocating and paying out hundreds of millions of pounds.

This is true for the repertoire databases that we all manage, but also the matching engines that link the reported music usage data back to that repertoire database. Not many industries have to put so much focus onto this sort of data management, nor do it at such scale.

These type of systems have some very interesting characteristics. Firstly, the option available to more traditional businesses, of selecting and configuring a world leading ERP package, is not a path that is open to us.

As a result, most, if not all, of the required functionality has to be custom built. Such custom build projects are typically costly, and generally come with a lot of risk around time and cost overrun.

“Not many industries have to put so much focus on data management, or do it at such scale.”

Secondly, some of the core functionality is difficult to fully automate. Developing software that can systematically and reliably determine whether the performer line-up on a particular recording is correct is difficult. Especially when you consider that the logic has to work not just for contracted, featured main artists, but also for the large number of non-featured session musicians.

Determining whether the stated rights owner for a recording is correct is equally difficult.

Using modern technology and approaches, such as Artificial Intelligence, can assist in the job, but getting it to make accurate, definitive determinations, at scale, is not possible.

I know this only too well – we use these kind of tools at PPL and they certainly help.

We have tools to help us group different versions of recordings into clusters so that we can look across them for inconsistencies in performer line-ups or in rights ownership, but they only take us so far.

Beyond some low hanging fruit that we can automate, we need to engage human beings to make the ultimate determination.

And that gets us to the heart of the matter. The systems and databases that we use have required large upfront investments. But more than that, their ongoing operation is costly.

Some of this is pure technology cost, but the lion’s share is the payroll cost for the large amount of human effort that is expended on a daily basis, managing the repertoire databases and matching systems that underpin everything we do.

For a long time, every Collective Management Organisation (‘CMO’) having their own set of systems and databases was the norm. It was as if being totally self-sufficient was a badge of honour, rather than the result of a properly considered assessment of the economics and risks.

To any rational observer, we have lagged behind other industries in identifying opportunities to collaborate and share back-end infrastructure, but things have been changing in recent years.

Much of that change has come about quietly, but it is starting to make a very real difference. By shedding a light on some of those success stories, I can hopefully encourage even more focus on them, and drive even greater adoption.

First up is VRDB, a project that I suspect many of you have never heard about. Initiated at the start of 2014, this SCAPR (the international trade body for Performer CMOs) initiative had nine of its largest CMO members work on getting to the heart of the collaboration challenge.

At its core, VRDB has established a central repertoire database to hold a single and shared view of the truth on performer line-ups. It achieves this by having each member CMO upload the recordings that are commissioned in their country, or where they have at least one performer member included in the line-up.

VRDB takes these multiple uploads and merges them into a single, shared repertoire database. Where one CMO’s upload suggests that additional performers should be included in a given recording’s line-up, this is notified to the CMO in the country of commissioning for approval.

This alone is a game changer – it centralises the effort to determine who is and is not on a line-up and makes the answer available to all. Indeed, a requirement of using VRDB is that the CMOs must synchronise these amendments to line-ups back down to their local database.

By doing this, VRDB removes the need for every CMO to independently assess what the line-up should be. It also removes the annual swarm of claim files that were traditionally sent between CMOs. Importantly, it removes cost. It also improves data quality and therefore it improves the speed and accuracy of payments to artists.

“This alone is a gamechanger. It centralises the effort to determine who is and isn’t on a lineup and makes the answer available to all.”

Unsurprisingly, it was not an easy project, and it has taken time to get to the point where it is now delivering real value.

The obvious challenge of getting multiple CMOs to agree on a common set of requirements turned out to be not such an issue. Indeed, after initiating the project in 2014, a live system was launched in mid-2016.

The hard bit has been making the fundamental business process changes back in the local CMOs to embrace a fundamentally different way of working, followed by the large volume of data cleansing that then has to take place.

The member CMOs have been plugging away at this over the past few years and here, in 2022, a great number have now abandoned the old, claims-based way of working and use VRDB as their primary means of identifying the correct performer line-up on a recording.

The second project I want to highlight is RDx. Conceived by IFPI and WIN, the Repertoire Data Exchange is a data hub that sits between record companies and CMOs. At the heart of RDx were three main aims.

The first of these was to make it easy for CMOs to gain access to authoritative data about recordings and their ownership, while at the same time providing record companies with feedback about how their data has been ingested, thereby gaining confidence that royalty distributions are based on this authoritative data.

The second aim was to reduce costs and complexity for record companies by providing them with a single channel through which they can provide their data. This allows them to retire the plethora of legacy data feeds and formats that have evolved over the previous decades, and ensures that everyone gets exactly the same data.

The third aim was to promptly surface competing claims of rights ownership in a recording. RDx does this both at the point of upload to RDx and also when that data is then loaded into a CMO’s local database. RDx notifies all parties involved in the conflict. It requires the conflict to be resolved and for corrected data to be re-uploaded.

RDx is a big step forward in sharing and collaboration. And just like VRDB, it is also a big step forward in ensuring that authoritative data is being used consistently in the payment of royalties.

But it is arguably only a stepping stone as it is not, in and of itself, a repertoire database. It is a data hub and pipeline to local CMO databases.

Whether it evolves beyond this remains to be seen. The potential benefits could be large, but would require agreement and focus to realise. As the VRDB experience shows, changing the very heart of how a CMO operates, with its complex set of tightly integrated systems, is difficult and takes time.

In many ways, this is a secondary concern. If RDx remains exactly as it is today, it drives out error, complexity and cost, and for that reason, its adoption by record companies and CMOs must remain the focus.

The pipeline of both is looking very healthy and those that have onboarded are already enjoying the benefits.

The final area that I want to look at is how we are sharing existing capability or combining data and IT system investments. I touched earlier on how complex and challenging it can be to build the IT systems we need.

One consequence of this is that unless you have the necessary scale or resources, it can be difficult to do it well. One solution to this, and it is one we have been pushing at PPL for several years, is not to attempt to build in-country systems, but to leverage existing ones.

“The rest of the world has. embraced sharing capabilities without fear or losing relevance or competitive edge.”

As I touched on earlier, many industries see it as completely normal to outsource certain back-office activities to existing, established players.

Be it motor manufacturers sharing core component suppliers, oil and gas companies sharing exploration and refining capability, or hotel and airline groups sharing booking platforms, the rest of the world has embraced sharing capabilities without fear of losing relevance or competitive edge.

Whilst we have been slow to embrace this in our own industry, I have really noticed things start to change in recent times. There are still some large in-country systems investments taking place, but I have seen a real uptick in CMOs looking to offload some of the heavy lifting to those that have the IT systems, databases and scale to do it well.

The ones that have done this – and PPL now provides back-office services to six countries, including Portugal, Ireland and Switzerland – have gained access to best-in-class capabilities at a tiny fraction of what it would cost to recreate locally.

In a similar vein, I have been encouraged by projects such as Soundsys. Led by IFPI, SoundSys has seen India, Indonesia, Singapore and Thailand pool their investment budgets to build a single system that they then share.

Two other countries are in the process of onboarding to Soundsys, with a third in the pipeline. Shared development and collaboration around existing systems is, like outsourcing, a good way to reduce cost and risk around the back-office IT systems.

These approaches can free up time and focus to concentrate on vital CMO processes, rather than on managing IT projects.

I have talked much about sharing in this article, as I strongly believe it to be the way forward. Our focus must be on maximising the amount of money that flows back to performers and recording rights holders and not on building yet more IT systems or databases that already exist.

Whether this sharing takes the form of back office services, pooling of investment budgets or in larger industry initiatives such as VRDB and RDx, this must be our continued focus.

There has been some really encouraging progress in recent times, but we must not be distracted. We need to build on the green shoots of collaboration and sharing that we have worked hard to achieve.