Business Interviews

Q&A: Charles Caldas, Merlin

By | Published on Friday 6 February 2015

Charles Caldas

The continued rise of the streaming sector was undoubtedly the most important trend for the record industry in 2014, and to a great extent for the wider music business too. As streaming income grows at an incredible pace, while CD sales continue to decline and, in some markets, download revenues have also peaked, that services like Pandora and Spotify are key to the future of music is now undeniable. Though there’s been plenty of debate about how the streaming music business is evolving, and what the streaming sector could and should look like.

Charles Caldas runs Merlin, the agency that negotiates and administrates digital deals on behalf of a sizable slice of the independent record industry; and he has negotiated deals for his member labels and distributors with most of the key players in the streaming sector. As well as having access to its own stats and data, Merlin also conducts an annual survey of its members, published each summer, to gauge their opinions and thoughts on trends in the digital domain.

Tapping into both that research and his day-to-day dealings with the streaming business, Caldas spoke to CMU Business Editor Chris Cooke about the current state of play, the big debates, and his own organisation’s recent deal with Pandora and its new office Stateside.

CC: Looking at your membership survey from back in the summer, and from speaking more informally to labels of all kinds, a trend in the last couple of years seems to be that the shift from physical to digital, and downloading to streaming, is happening faster in the indie sector than in the record industry overall. Why do you think that is the case?
CCal: I’m not sure it’s all happening faster, for example the drop in download sales amongst our members seems less pronounced than in industry-wide figures. But I definitely think the independents are adapting quickly and creatively to the shifts in market dynamics, and their success over the past year – both in the overall growth of the business and at a mainstream chart level – underlines that.

It’s possible that indie labels are primarily selling to the kind of consumer who is naturally shifting faster from one model to the next. But I think the rate of growth the indies are seeing in streaming is also due to the nature of those platforms and the ways the indies use them. In physical, indies were often at a disadvantage because of barriers to entry, to getting records into high street record stores. Many of those barriers were removed by downloading, though most download stores were built to mirror the record shop experience, just with a lot more shelf space.

But with streaming platforms, there are multiple entry points, via social media, email, artist’s own channels, as well as any channels controlled by the digital service provider. Many indie labels have capitalised on this in clever ways – Domino’s ‘Best Of Indie Radio’ approach comes to mind, but they are not alone – and I think that is helping those labels see the benefits of the streaming sector quicker than the majors.

CC: Do you think that the indies performing better in the streaming space will be a long-term trend. You mentioned that indie music consumers are perhaps more likely to be early adopters. As streaming goes truly mainstream, do you think market share stats more akin to the CD-era might reappear?
CCal: I truly believe that there is a fundamental shift happening in consumer behaviour. That once you liberate consumers from tightly controlled storefronts, and give labels direct access to consumers in the way these platforms allow, our sector will continue to grow, and thrive.

The early adopter theory is disproven for me by the fact that some of our strongest territories in terms of market share are also the most developed streaming markets. And in certain countries, the Nordics, Netherlands, the UK, a significant portion of the growth in the wider streaming market is now coming from mobile bundling. Those consumers aren’t early adopters in the digital music space, and yet in those territories on those platforms we are seeing growth inline with the wider market.

So I think we are seeing a more fundamental shift in consumer behaviour. And the fact that 2013 saw both the rapid rise of streaming music and the best chart performance by the indies in the UK since the days of the big pop independent labels like Jive is no coincidence.

CC: You have talked a lot about the ‘market share’ of the three majors and the indie community – and the Merlin membership – in recent years, and the way those stats are calculated. Why do market share figures matter, and what do you think is wrong with some of the stats that are out there?
CCal: Each player’s market share is incredibly important when you are negotiating first deals with new digital service providers, because start-ups have a finite amount of money for advances, and equity to offer to rights owners, and the simplest way to divvy that up is by market share. Except if new DSPs take self-declared market share figures from each player, they’ll get to 111%.

It seems to me that when those first conversations take place, the majors present market share figures based on the record industry at large, even though these are start-ups operating in the streaming music market. And, as we’ve shown, if you take just the streaming music market, the indies come out with a bigger share. We see it every month in our reporting from partners, whether that is Rdio, Deezer, Google Play or Spotify. The majors can also skew the figures by including sales of the indie labels they distribute through their label services businesses, even though the long-term rights to distribute that content are often far from assured.

Nielsen, who many rely on for their record industry data, do not publish market share stats for just the streaming market, which means start-ups have to rely on self-declared figures which are likely to be misleading. Independently produced streaming market share figures would help a great deal.

CC: Although we’ve heard a lot this year about iTunes income peaking, one of the features of your membership survey was that download revenues were still rising across the board. Do you think that’s because the two kinds of digital business models don’t necessarily compete? Do different models satisfy different consumers, or does each customer want to pick and choose?
CCal: The notion that there is a silver bullet that will ‘fix’ the record industry, a model that will dominate for the foreseeable future, is a fallacy. Formats have always shifted over time, and will continue to do so, some attracting certain market segments, other having wider appeal.

Actually, market segments have never been mutually exclusive. Yes, we used to distinguish between the supermarket shopper and the HMV shopper when selling CDs, but plenty of people crossed over, maybe relying on the former most often, but the latter at Christmas. Or you had the avid Rough Trade customer, who would still pick up a present from Woolworths because of the discount.

Streaming is not only about monetising behaviour that previously had a value of zero – though it is by turning former pirates into subscribers – but it is also allowing more discovery and experimentation, which can lead to consumers spending in other ways. I believe subscribers will continue to buy physical, especially vinyl, and may support certain artists by buying downloads or other products.

And those relying on freemium music services are also still consumers of other music products.

CC: Do you think streaming music services can reach the mass market audience most of the services need to succeed – ‘mainstreaming’ as Spotify boss Daniel Ek recently put it?
CCal: Yes I do. We are at the nascent stages of a massive format shift, and are just starting to understand the power of music bundled with phone plans and devices and so on. It may take a while to get the bundling approach right, but it really has the potential to take subscription-based services mainstream. As I said, in some countries where it has been done right, that is already happening. The Vodafone partnership has started to deliver for Spotify in the UK, and the Orange-Deezer tie-up in France also demonstrates the potential.

When it comes to the growth of streaming, you have to look at things on a territory by territory basis. The size of the CD and download markets is usually key. In the Nordics, where we know streaming is already dominant, legit downloading never really took off, the market went from CDs to piracy to Spotify. In the UK, and especially the US, iTunes became a really big market as CDs went into decline, so the shift over to streaming is slower.

But the bundles really help with the shift, because they simplify the process for more mainstream consumers. The success of Netflix and its like also plays an important role here, it is easier to sell subscription music services to mainstream consumers if they are already buying subscription movie and TV services.

CC: The indie sector does seem to be generally pretty optimistic about the streaming market, would you agree? If so, why?
CCal: Yes, in the main indie labels are optimistic. Because people are seeing subscription services deliver incremental revenue, despite the tension between the various business models. And that’s before you grasp the potential of the tools within the streaming platforms to help artists and labels find and understand their audiences.

CC: Though, of course, there have been some very vocal critics this year of the streaming business, some from labels, but most from the artist community. Obviously the Taylor Swift initiated debate was loudest, and that focused in particular on Spotify’s freemium model. What would you say to those who are against the free level of services like Spotify?
CCal: The first thing you have to look at is the entirety of the value chain across all forms of consumption; so if you are going to criticise free access to music on the web, you have to look at YouTube and SoundCloud as well. I understand different people will take different views on how each of these work, and everyone’s opinions are equally valid because we are still at an evolutionary stage, and are all trying to understand what works as it happens. But freemium on services like Spotify can’t be evaluated without considering YouTube and SoundCloud.

In his response to the Taylor Swift debate, Daniel Ek re-ran the line that without the freemium entry-point taking people from a low value to a high value service is practically impossible. Whatever you think about that, there is plenty of proof that his approach does work; people sign up for free, live with the service, understand how it works, and what the benefits are in paying for it. I can understand why, for an artist, streaming is a challenging addition to the traditional album campaign system, because the new model is more about monetising consumer behaviour over a much longer time period.

But the reality is, this process of moving consumers from low value to high value services is key to the wider industry, and that is what Spotify freemium is about. Yes, the income is less, considerably less than Spotify premium, though still more than YouTube, and obviously significantly more than SoundCloud until it fully works out how to monetise. And more importantly, with Spotify we are seeing that movement from low value to high value; and at an impressive growth rate too.

CC: Our survey of the artist community for The Great Escape earlier this year showed a lot of confusion about the way digital services pay royalties. Do you think the labels and digital service providers could do something to address this confusion, or is it inevitable when you have licensing deals with so many variables?
CCal: Transparency is important. And in that domain, Apple was always pretty open about its iTunes revenue model, there was more or less a 70/30 split in the rights owners’ favour, which was a similar split to physical.

The streaming services actually adopted a pretty similar model too, at the top level. Though it gets confusing because in the streaming domain, total revenue is linked to subscriber numbers not overall usage. Which means we have to stop measuring the market in terms of units, and start thinking about it in terms of number of customers, and what each customer is worth. Which means asking ‘how many streams equals a download?’ gets you nowhere.

I think the challenge, more so than understanding how the DSP/label revenue model works, is working out what this shift in thinking means for the label/artist relationship. The challenge is working out how artist deals should look as streaming becomes a key revenue stream.

CC: When some high profile artists discuss Spotify, the fact labels – including your members via your deal – have a stake in the company is used against them. What would you say to those artists?
CCal: Merlin exists partly to enable and assist digital services in getting a significant amount of repertoire, but also to ensure the independent record companies we represent get appropriate value for their catalogue. And we are not for profit, so the value we secure is for the benefit of the labels.

With the services where we have equity, it’s one of the mechanisms we use to ensure we are competitive with the majors. If we license a service where the majors have equity and we don’t, and that service becomes a success, then our labels’ content played a role in creating value that only the majors see the benefit of. By taking equity on behalf of our members there is commercial balance.

We don’t gamble our members’ rights by giving hugely preferential rates in return for equity, doing so would do our members a disservice. But allowing our members’ repertoire to generate value only for their competitors – ie the majors – would also be counter-productive.

None of the digital service providers in which we have equity have as yet been sold or floated. But if and when they do, as with any form of monetisation, we will endeavour to distribute the monies in line with usage on the service in question. The system will work in a similar way to when we have won damages for our members from piracy services, such as with the LimeWire settlement. Though, actually, with a service like Spotify, we’ll benefit from very accurate data on past content usage.

CC: One of the big developments at Merlin this year was the deal with Pandora, meaning your labels can now license the American personalised radio service direct, rather than via the SoundExchange system set up by the compulsory license that exists for that kind of streaming Stateside. Why did you do that deal?
CCal: The compulsory licensing system in the US means labels have to license services like Pandora via SoundExchange if that’s what the DSP chooses. The system also ensures that all rights owners are paid on an equivalent basis. The only reason for doing a direct deal, therefore, is if we can find an arrangement that is attractive to Pandora – so they choose to take it – but also offers a better deal to our labels than licensing via SoundExchange.

So the aim was to find a commercial arrangement that benefits both Pandora and our labels, so that at the end of the deal, our members have done better overall. This is possible because there is so much usage of our labels’ content on a service like Pandora, that just the tiniest of adjustments can make a positive difference overall.

I should stress that our deal only affects the label’s share of income via Pandora. Recording artists are also due a cut of this money, and that arrangement is unaffected and will continue to be handled by SoundExchange. Though if our labels’ content is used more, then the artists who appear in those recordings will do better too.

CC: The other big news from you guys was the opening of a US office for the first time. What motivated that?
CCal: Part of it was a matter of scale, around a third of our membership is American and probably about half of our revenue comes from the US market. And it’s the fastest growing part of our membership too. Me and my team were travelling to the US every few weeks, and it became clear as this year went on that we needed someone to deal with matters there on a day-to-day basis. Having a direct relationship with Pandora is part of it, though all of our active on-demand streaming partners have also shown substantial growth in the American market this year.

CC: Despite the digital boom, piracy was still referenced as a ‘threat’ by labels in your annual survey. I know this isn’t directly in Merlin’s remit, but are their any approaches to further combatting piracy that you sense the indie sector – which always resisted the more draconian anti-piracy initiatives – do actively support?
CCal: We have successfully sued infringing companies on behalf of our members, including Limewire and Grooveshark, and I think where the indie labels see companies turning piracy into a business, they feel we should collectively act.

And indies, like the majors, are concerned that it is still so easy to find free unlicensed music on the internet. And I think there is a common position between the majors and indies on the legislative side here, that if they won’t do it voluntarily, the likes of Google need to be pressured to stop linking to illegal content in their search results. And I think most labels look to the industry’s trade associations to lobby on this issue.

But day to day, I think the indies are much more focused on what’s next in the monetised space, and are still hoping to building a digital music market that renders piracy pointless.

CC: Another issue I’ve been hearing from indie labels this year is the data challenges thrown up by the booming streaming sector, ie how to process DSP data in a uniform and efficient way, to work out what labels and their artists are due. Is this something your members are talking about? Can Merlin play a role in tackling this challenge?
CCal: It is a challenge. The amount of data labels have to process just to generate royalty statements is becoming very large indeed. We are trying to help. Wherever possible our labels have direct access to analytics tools to track usage, and we are working on technical solutions to allow usage reporting to become simpler, and more standardised.

Technology will help us overcome these issues in time. The challenge then will be to work out, of all this data, which provides important market information that can genuinely inform business decisions, and which data actually isn’t very useful at all.

This interview appeared in the December 2014 edition of the CMU Trends Report. Buy our reports from the CMU Shop or get every edition by signing up for CMU Premium