Tag: MPAC

The Math of Matching – Part 1

Posted on | by Hans Peter Roth

Streaming music leaves rights owners – particularly the publisher, society, composer, author groups – with two major challenges. First, the Digital Service Providers (DSP’s) like Apple, Spotify, Amazon, Tidal, Deezer, Pandora, etc, can send anywhere from 1.5 to 5 million different tracks in a monthly usage report. The massive volume of tracks used and reported every month has quickly become overwhelming. Second, The descriptive metadata is flawed at best and missing at worst. The information we got from the “liner notes” in vinyl and CD’s are seldom available online. The names of composer/authors are often missing or wrong, even on major releases. 

This is why it is so important for digital rights administrators to be able to both handle vast amounts of data (without “cutting off the long tail”) and triangulate the flawed/ambiguous metadata accurately in order to match all tracks to their corresponding works. Indeed, the ability of a digital rights administrator to master both aspects are key if the rights owners are not to end up losing more when paying less for services.

Below, I will look at the challenges of handling the massive amounts of usage data reported by DSPs and how that affects rights owners collections. In a later post I will focus on the prevalence of flawed metadata and the effect it has on payouts.

CUTTING VOLUMES MEANs LOSING Money

In the year 2000, at the peak of CD sales, 942,500,000 CD’s were shipped in the United States. As an example, let’s assume that each release sold 1,000 copies on average. If each release had 10 tracks we would be dealing with 9,425,000 unique tracks in the year 2000 alone.

As high as that number may seem, it is only what roughly represents one or two monthly usage reports from one DSP for one product tier! In short, what was once years  if not decades of usage data is now reported every month in the digital music industry! Indeed, I have previously calculated that the online music data volume in the Nordic region rose from index 100 in the year 2000 to index 112,500 in 2013 – and that was seven years ago!

So how did the industry cope with this increase?

Well, when 90-95% of value can be found in the first 20-40% of used tracks (cf. above between 300K and 1,600K tracks), it was, and is, “easy” to simply not analyse the remaining 1,200K – 2,400K tracks for the last 5-10% of value. Particularly since it simply became too expensive and/or time consuming for the rights owners/administrators to analyse the data in the tail. Hence it became an industry standard solution to “Cut off the Long Tail” when processing data, and focus only on the lucrative part in the “hits” section.

While this solution saved you 60-80% of the workload, this also meant that rights owners said goodbye to 5-10% of revenues in an otherwise equal scenario! Even then, we know that not all the remaining 20-40% of data was or is always processed.

Leave No Bit Behind – MPAC to the rescue

When we set out on Muserk’s voyage to become a leading global digital rights administrator, we set the goal that our back end had to be able to cope with all the data volumes.We were going to leave no stone unturned – leave no bit behind. Not only do we believe that everyone should get paid correctly, we also believe that our customers and their rights owners have the right to know when their work was matched but did not generate (enough) money to warrant a distribution. As our Head of Technology, Collin White, put it; “we have to carry the zeroes as well as the heroes”.

So that is what we did. Today, our proprietary scaled cloud infrastructure, MPAC ( Muserk Primary Automation Cortex), enables us to handle a multitude of workflows simultaneously. This enables Muserk to process tens of millions of reporting lines in less than an hour. To draw on our example of CD’s in the year 2000, Muserk processes a year’s worth of data in less than an hour or more. That’s two decades of data in less than a day! At Muserk, we are able to look at every single reporting line, every single track, every single bit and match it to our customers repertoire. And, from first-hand experience, we can see the difference that makes to our customers and their earnings.

Money Spent Is Actually Money Earned!

“So how does this affect my collections?”, you may ask. Well, in an industry where the average commission rate is typically between 10% and 20%, finding the 5-10% of value in the long tail makes a huge difference when calculating the actual value of a service. In fact, you can very easily end up with less money by paying a lower service fee.

For example, rights administrator A charges 11% commission, but cuts off the tail and only finds 90% of value. This leaves you with about 80% of your copyright value, and with no visibility of the long tail. Thus, you have no idea which works represent 10% of the value of your rights, how they are used, by whom they are used, and where your usage is coming from. Conversely, rights administrator B charges 19%, but does not cut off the long tail. Now you are getting 81% of your copyright value plus full granularity on all usage! In short, the true value of a digital rights administrator is more than its price tag. That is true when you look at volumes alone, but even more so when you include accuracy.

More to come

In my next blog post I will look at how the ability to accurately triangulate the flawed metadata of the online industry further augments the example on price and value above.  Stay tuned for more! In the meantime, feel free to check out some of the other Muserk blog posts.

Automated Copyright Administration – Where Technology Meets Process

Posted on | by Colin Moriarty

In the last two decades, we have watched the music industry explode with innovation. Today, all that is required to listen to nearly any piece of musical content is an internet connection. Likewise, the basic tools necessary for creating said content now come pre-packaged with most computers. The average songwriter harnesses the power to make their work available to the world overnight. These simple facts are often taken for granted, as is the value we assign to music and those who are responsible for bringing it to our devices. In the wake of an unprecedented rush of content, the industry is tasked with making sure that all songwriters are accounted for in a timely, efficient, and accurate manner. Failing to do so will only further validate the feelings of distrust and skepticism that many artists hold towards the music business.

An unfortunate consequence of the modern royalty collection ecosystem is a prioritization of the high-earning works. Our proprietary matching technology called MMatch® enables Muserk to treat long-tail works with the same weight as top-earners. Through a combination of MMatch® and standardized processes, Muserk is capable of handling a massive volume of data while simultaneously minimizing the opportunity for human error. This value-agnostic approach allows us to treat all royalties with the same priority.

Identifying usage has surpassed the scale at which humans alone can reasonably achieve adequate results. Every month tens of millions of sound recordings are streamed on music services in the United States alone. In the beginning, to even reach a starting point where a person could begin analyzing potential matches, they were left to rely upon the only common composition-level data point that exists in most usage data- an ISWC. This means that songs lacking an International Standard Musical Work Code could not receive the attention they deserve, since matching on title or writer alone produces disastrous results. Besides the limitations on accuracy that are inherent with such a process, scaling to a global marketplace could simply not be achieved. With these obstacles in mind, Muserk began developing the MMatch® technology, which is capable of evaluating the relationship between text-based data points such as titles, writers, and artists. Muserk’s data pipelines have evolved to match the work once done by hand and have been enhanced with the capability to handle a wider range of data points. 

A common criticism of employing artificial intelligence to overcome obstacles in royalty collection is that one cannot be entirely sure that a link has been accurately identified. At Muserk, we recognize the truth in this sentiment by acknowledging the critical stages of human analysis that occur prior to pushing newly discovered data to the DSPs. In any industry, technology is meant to aid one in the ability to perform their job. Just like doctors do not rely on heart monitors alone to save the lives of their patients, a rights administrator cannot solely use any piece of software to confidently collect on behalf of their rights holders. While we cannot completely eliminate human interaction in the rights collection process, our data pipelines help us reduce our input to near zero. 

We are evolving our process with every iteration by continuously targeting our biggest bottlenecks and identifying how information can help us make better decisions. With a desire to do more with less we are motivated to continue to reduce the workload required by humans to collect royalties. As the methods through which music enters the marketplace continue to evolve, Muserk will remain an instrumental player in shaping the narrative of modern rights management.