Michael Walker: Yeah. Alright, so I'm excited to be here today with my new friend Paul Sampson. So Paul is the Founder and CEO of Lickd, which is a platform that's revolutionizing music licensing for content creators by offering pre-cleared popular tracks for legal use in videos. He is a veteran in music licensing. He has experience at companies like Extreme Music and Music Dealers, and he's an advocate for creator empowerment, developing solutions that allow creators to enhance their content with mainstream music while also making sure that the artists are actually fairly compensated.

So today, on the podcast, we'll talk a little bit about their platform Lickd and how it's helping to transform music licensing for creators. So Paul, thank you for taking the time to be on the podcast today.

Paul Sampson: Absolute pleasure. Nice to meet you. Glad to be here.

Michael: Absolutely. So to kick things off, I would love to hear a little bit about your backstory in terms of coming up with Lickd and what the main problem was that you set out to solve with the platform.

Paul: Sure. Okay. So, I have been in sync licensing. I presume you and your audience will know what sync licensing is.

Michael: Yeah. Yep. I'm not a sync licensing expert, but fortunately I get to connect with people who are smarter than I am, like yourself, in the sync world. So we're definitely familiar.

Paul: Okay. For anyone that doesn't know, essentially it means licensing music into picture. So anytime you watch a film, a TV show, or an ad and there's music in it, someone's creatively pitched that song and then negotiated the rights. If it wins the pitch, they make sure that the rights holders, artists, and writers get paid a fair market rate for that usage and then distribute the money accordingly.

I've been doing that since 2005, in New York, LA, and then London. I did it in film, television, and advertising. By about halfway through 2015, I remember reading a report on the music industry that mentioned that sync licensing in total contributed only 2% to global recorded music revenues.

Wow. I thought, “God, there are a lot of us scratching around trying to get a share of that 2%.” I also realized something else: the fastest growing production sector in the world was no longer the sectors that I was licensing music into. It wasn’t film, television, or advertising. It was user-generated content. Think YouTube at that point, right? 10 years ago.

I thought, “Well, how would you go about legally compensating an artist if you wanted to use a major song in a YouTube video?” And what I learned horrified me. What I learned was that if you use a famous song in a YouTube video, not only is there nowhere you can legally acquire all the necessary rights to use it, but YouTube will recognize the song, presume your use is copyright infringement, and that copyright claim would divert any revenues you, as the creator, would earn.

And what that led to was that the fastest growing production sector in the world — UGC — had precluded itself from ever working with mainstream music. The most valuable, highest grossing audience sector of the fastest growing production sector in the world was actively avoiding using mainstream music in their content because it was appropriating their income. Does that make sense?

Michael: Mm-hmm. Yeah.

Paul: And so I thought, that's wild. How do we future-proof this? How do we make sure there’s a solution so that artists and writers can monetize this opportunity? The fastest growing production sector in the world didn’t have a sync solution. That’s a problem.

I also thought, well, hang on a minute. If that scales at the rate UGC is scaling, then it's quickly going to change that 2% of global revenue to 4, 5, 6, maybe 10. So I looked into it and realized what you needed to build was something like Spotify for content creators. That’s the shorthand I used.

But essentially, you needed a platform licensing music at that scale, with the content being that ephemeral. It needs automation. It needs pre-clearance of rights. And it needs a platform where you can audition and license within minutes — not go through a manual licensing process set up for TV, film, or advertising, where there are three-month lead times and you can handle the license over six weeks.

That is a monumental effort, but that’s essentially what we set out to do. The first thing, though: let’s imagine you build the platform, sign the rights holders, they grant you rights, and you match the master recordings to the publishing rights. Whatever is clearable, you put on the platform. Even if you’re lucky enough to get content creators to use it, they still face the problem that YouTube will recognize the music, think they’ve infringed, and divert their revenue.

Then they go, “No, I did it legally!” But who do they tell? How do they get the money back? Horrific. Bad user experience. So the first thing we did was sit down, myself and the developer, and take all the YouTube certifications — the API, the CMS, and Content ID, which is their audio recognition software. We plotted a path through all of those and learned that if we build the software in such a way that rights holders integrate it into their YouTube CMS, this software can inform Content ID of the difference between an infringement and a legally licensed song.

If we can inform Content ID when a song is used legally, we can lift the copyright claim. And if we can lift the claim, people can come to Lickd, pay to license a song at a fair fee that scales with audience size, use that song, and not be persecuted on YouTube. The software would inform all necessary parties that this was handled legally, that it was paid fairly, and that no action should be taken on the video.

We called that software Vouch, because it would vouch for our customers. Once we solved that problem, we were so excited. By this point, I’d quit my job, raised some money, built the software, it worked, and I started taking meetings with rights holders. I thought they were going to hold me aloft on their shoulders and parade me around the streets like a hero because I’d just built technology to open up a massive new revenue stream for them.

That was incredibly naive. Michael, I’ve said this many times: when it comes to adopting innovation, historically the music industry has only ever been slightly ahead of the Amish. With every major tech disruption, they resisted. And they resisted us at scale.

I remember going into YouTube early on, meeting with one of the music team, and saying, “This is what I plan to do. This is how it’s going to work. This is how I’ll get rights holders on board.” He laughed. Literally laughed in my face. He said, “Yeah, okay. You’re going to get all the distributors in the world to integrate your software, deliver you their music, and then do the work of data-matching all that music? Good luck.” Then he ended the meeting.

Even the guy who gave me my seed funding said, “Paul, I’ll give it to you on one condition.” I asked what. He said, “You stop this pipe dream of landing the major labels and publishers because it will never happen.” I told him, “Well, you better put your money back in your pocket then, because I’m not doing it on that basis. I’m doing it because we need to change this from the very top.” He finally agreed.

Today, there are 14,000 labels and publishers signed to Lickd. They’ve delivered us 14 million songs, 1.5 million of which are cleared and available on the platform for creators around the world to use. Included in that 14,000 are Universal (label and publishing), Warner (label and publishing), BMG (label and publishing), Kobalt, and all the big publishers.

Artists like Justin Bieber, Coldplay, Ariana Grande, Blink-182, Post Malone, The Weeknd — they’re all being licensed by the biggest content creators in the world.

When you think about YouTube today, there are 113 million active YouTube channels. On average, each makes five videos a month, 12 months a year. That’s around 7–8 billion videos published every year. And 84% of them use some sort of music. If you make the best music in the world available to them, just like it is for film, TV, and ad producers, then you’ve suddenly got an amazing new, scalable audience that is motivated and happy to have those resources.

When I had that realization in 2015, by that point, YouTube was 10 years old. All the means of production had already democratized to enable UGC. You didn’t need a cameraman anymore — you had your phone or a holiday camera. You didn’t need an editor with a £1,000/day edit suite and a £50,000 Avid setup — you had iMovie for free on your laptop. All the means of production were on your dining room table. The only thing that hadn’t democratized in line was what I was doing.

That’s what we set out to do. That was the mission. The mission for Lickd was to democratize music for the world’s creators and future-proof digital opportunities for the music industry, making sure artists, writers, and rights holders get paid.

Michael: Cool.

Paul: That was a long answer. That's really awesome.

Michael: I mean, that is so cool. Thank you for sharing it. Congratulations—I love a story like that, especially when there's sort of a face against the odds, and people literally laughing at you at the beginning. Then look what you've been able to do.

That's really amazing. One thing that caught my ear, I thought was really interesting: usually when I hear about licensing—especially well-known pieces of work—I'm thinking of tens of thousands of dollars or hundreds of thousands. Very high licensing fees. I think a lot of creators don't necessarily have a budget like that for their user-generated content.

But what I heard you say is that there's a model that scales as you grow in terms of your audience size. I'd love to hear more about that mechanic. Am I understanding it right—that it basically makes it more accessible for someone starting out to be able to actually leverage some of those placements?

Paul: Absolutely. So to give you an idea, it's slightly more complicated than that, because the model we offer now is a credits package. You get unlimited stock music plus one mainstream song, right? For like $15.99. Or you can get two mainstream songs or four mainstream songs with your package. But the price for each of those scales based on the size of your channel.

If you took just the mainstream music away—going back three years when we didn’t offer stock music and only offered mainstream—a mainstream song for an entry-level content creator would cost $8 (seven euros, six pounds). That same song on the same day might cost the biggest content creators in the world a thousand times that. And anywhere in between along that spectrum. Does that make sense?

When you create an account with Lickd, you OAuth interlink through your YouTube or Google account. It ingests all your channel analytics. It works out what we call your AVV—your average video viewership—and that falls into a sliding rate card of bandwidth.

So zero to 50,000 views per video would be six pounds, seven euros, $8. Fifty to a hundred thousand views per video would be ten pounds, eleven euros, $12, and so on. Each time you log in, it recalculates.

What we've noticed is our customers grow with us, and many of them credit the growth of their channel to using mainstream music because it engages people better. They get more comments, people watch for longer.

Imagine you're watching a 15-minute workout video on YouTube. It's hard enough to finish a 15-minute workout. It's even harder if it's scored by stock techno music. But if that workout is scored by five of your favorite tracks, you're going to finish it.

YouTube looks for metric changes. If suddenly people are watching videos longer, completing them more often, commenting, sharing, and watching more, it drives you into the algorithm. It starts promoting your content, which drives more views, which drives more revenue. You're in the opposite of a vicious circle—a victorious circle.

What we've noticed is people start with us at $8 and end up at $100. At that point you wonder if they’ll churn because they’re paying ten or twelve times more than a year ago. Invariably they don’t. They say, “We got here because it increased the production value of our content. We’re making better content. It sounds like we want it to. People enjoy it and resonate with it because they recognize the music.”

So it's been a win-win for everyone in the value chain. Rights holders get more revenue. Artists and writers get paid more than before. Creators get to use the same music available to other production sectors. And the audience gets better content.

Michael: Cool. I like it. That just feels like the best way to set it up: usage-based, with value paid out to people based on how it’s being used. That’s really cool that you cracked the code on how to make it accessible so people can feature those songs and grow at the earlier stages.

As someone forward-thinking enough to have an idea like this, have people laugh at you, and then see it through to change the industry—it feels like we’re having a tidal wave moment with generative AI. It’s changing a lot in terms of creative fields. Just a couple of days ago, at the time of recording this, OpenAI dropped Codex, which we’ve been using for development, and that’s amazing. It feels like every day something new is happening.

Paul: Yeah.

Michael: In particular with licensing and music sync, there’s a large use case when you’re not trying to use a recognized song, but to generate with AI. Probably your platform is even more suited in that world because people come to you because they care about recognition from mainstream artists. That’s the resonance they’re looking for. I’d love your perspective on where things are at, where they’re headed, and how you view this next wave.

Paul: It's really interesting. I think you somewhat answered the question in your question.

Michael: Yeah, I talk too much there.

Paul: I’m not saying that. I mean you touched on something really interesting. You said when people are looking not to use mainstream music. Well, when are they not looking to use mainstream? There are times. That’s why we offer stock music as part of the bundle.

Stock music does one thing very well and one thing very badly. What it does very well is provide a toolkit of sounds not provided by mainstream music. So if I need my video to sound like Pirates of the Caribbean, or to sound mysterious, or like Psycho, I need a toolkit. I’m not going to get that from Ed Sheeran or Jess Glynne. I go in and type “horror,” “dramedy,” “comedy,” “fantasy,” “adventure,” “drones.” That’s your toolkit.

The production costs of creating stock music catalogs are going to be driven incredibly far down, because people will soon be able to generate those sounds without paying monthly subscriptions. Similarly, the companies in that space won’t need to spend $3,000 on each new track to keep their catalog fresh. What they should do is pivot to incorporate GenAI, then use it to augment the tracks they already own.

In terms of using it instead of mainstream music and the threat that holds, I’d say this: if we don’t democratize access to mainstream music and make it affordable—understanding this is a volume game based on low fees and high volume—then we will lose out. If we cling to monetizing scarcity, the industry will lose big revenue opportunities to GenAI.

But one thing GenAI music will never be able to do is evoke nostalgia. It can’t resonate with your lifetime of memories. Now, you might say that in 20 years there will be hits made by GenAI that people feel nostalgic about. But I’ve played around with the big GenAI apps, and I just don’t know. I don’t understand the commercial model.

Michael: Have you used them?

Paul: Yeah.

Michael: I have too, and I’m with you. They’re not there yet in terms of being on par with the top mainstream artists. I will say they’re mind-blowing in how quickly they’ve developed. If we have this conversation a year from now, we’ll see where the quality is. But if you’re asking if they’re on par right now with the top music in the world—I don’t think so. Though it kind of blends together since artists are starting to use those tools now.

Paul: I think there's a really good use case for Gen AI, and it's in music production. If it helps you produce music quicker and better and with fewer people, right? Mm-hmm. That's amazing. We should be welcoming that.

But if you're gonna create and release music and then tell me I've gotta get emotionally connected to it, I'm gonna struggle, right? I'll give you a perfect example. I signed up to the two big ones, right? There probably are more now, I know, but I signed up to a big one. I put in a couple of prompts. I augmented them, I iterated them, and then I played them to people and they were like, “God, it's amazing what Gen AI can do.” But no one stood around dancing, right? No one said, “You've gotta send that to me.” And after about a day, I haven't touched them since. So as a B2C product, I don't know what the model is.

Right now, I don't know about you, but when I fall in love with a song, the first thing I wanna know is who produced it, who wrote it, who's this artist I haven't heard of? What was it that happened in their life that made them come up with that lyric? What diction did they have? What heartbreak did they go through? Where were they born? Where were they raised? Where's this place they're telling me about, right? You'll never get that from AI. Never. It would always be a robot that did it, right? And so it's always gonna lack that emotional connection with its audience at scale.

Will it be able to fill some dance floors? Yes. Will it really grab the hearts of consumers globally? I think there will be some inevitably that do, but will it do it at scale? I don't know. I think most people love first the music, and then fall in love with music when they understand the artists and the motivations, and that human being's ability to resonate with you and connect with you and describe something that you went through as if they're describing your life. And that's why you feel connected to them.

The moment I hear a robot did it, I'll be like, “Oh, well this is just synthetic.” Mm-hmm. So I think there'll be a place for it, but I don't think it will ever A) be able to create nostalgia, or B) be able to connect emotionally in the way that we think about music and fall in love with music and artists and festivals and live performances. Right.

Michael: Makes sense. Yeah. So what I'm hearing you say is that these are incredible tools when they're used as tools, and when they're separated from the human or from the artist itself, it sort of loses that emotional connection. It kind of gives it more depth versus when it's just like a robot did it on its own, there's not a whole lot to sink your teeth into. It's like, okay, cool, a robot did it. Sort of like watching two robots play a game of chess against each other. It's like, okay, that's impressive that they can play chess against each other, but it's just not entertaining. It's not really fun to watch.

Paul: Exactly right. Exactly right. So look, there's a place for it. I think ultimately the bigger consumer play will be that there'll be more hobbyist music producers, perhaps, and AI will help them not need a studio, right? Mm-hmm. In the same way that I've helped YouTubers or content creators not need a business affairs team, right? Because they were never gonna have one, and not need to go through a six to eight week clearance process, because we've streamlined things.

This can help people that love music but don't necessarily have musical ability or access to a studio or funds suddenly start to be able to create some music, and there'll be a home for that music somewhere. But it'll come from a human. The input will be the human. So yeah, I think that's where the market is. I just don't know that we're all gonna end up paying 15 pounds a month to have the ability to create music on the fly.

Like, someone sent me one and I was like, “It's not, it's quite good, it's quite catchy. When did you make it?” And they said, “Oh, it just took about eight seconds, I did it on one of those apps.” Immediately I lost all feeling for the song. I was like, “Oh, well I could have done it.” And I feel like if I could have done it, there's not much value in it.

Michael: Yeah, it's super interesting. Even—I don't know. Regardless of what I say or what either of us says, five, 10, 20 years from now, there's gonna be some way that I’d look like a fool. But I do feel like I probably align most closely with Ray Kurzweil from his line of thinking in the book The Singularity Is Near, in that we appear to be on an exponential growth curve when it comes to information technology, and music is an important part of that.

I really do think that artificial intelligence is sort of misunderstood in the terminology. We have “artificial intelligence” and it feels like fake or not real. And when I think of Microsoft Sam, when it first came out, you know, “I am a robot. It talks like this,” versus like, you know, modern AI now and it's starting to articulate. It's still not quite there, right? It's still not quite there. But I feel like we're on the verge of a superintelligence that changes what it means for our economics. We're just gonna have this age of abundance and the ability to create.

I'm aligned with your line of thinking in that I think this type of intelligence is something that I hope humanity comes along for the ride with, where we find a way to synergize with it. But I'm probably a weirdo that really believes in that. We're building a form of intelligence that is going to—either humanity is gonna come along for the ride and get Neuralinks installed and be able to use these tools, or we’ll have a relationship to it similar to how an ant has a relationship to us as humans. It's just hard to even fathom or understand. Like, if an ant were to try to imagine what it's like to be human, it just doesn't compute.

That was probably way down a rabbit hole and a bit off track versus what we were talking about.

Paul: Listen, I think you are right. I think being asked to answer questions like these, as you put it to me, will make one or all of us look stupid in the longer term. In terms of my industry, I don't see it affecting sync licensing in the next 20 years dramatically, other than I think it will change the means of production for stock music or people that can't get access to mainstream music. I think it will democratize that space completely.

Michael: What I like about your company and what I feel it's well suited to thrive in, in this new world of generative AI, is the model you've created with making it accessible. Making it accessible to use these tools and for it to scale with you. Because there's always gonna be a need for—like, we're gonna want to soundtrack our music with the music that we enjoy listening to, and that people enjoy listening to, that resonates with them.

I think the only question I would have is around personalization. Probably at some point in the future it'll be like you're watching a YouTube video and then it knows who you are, so it's gonna soundtrack it based on your personal preferences and what you like. And so it just dynamically generates the soundtrack.

Paul: Yeah, I think it would need another input, which is: what does the video need? Absolutely.

Michael: That would be a pretty complex equation.

Paul: I don't think it's that complex, really. But like you say, look, what does the video need? And of those, what are my favorites, right? What you like and what the video needs are two separate things. The video might need drones and space and a climax and a slow build. And your favorite song might be 180 BPM consistently for five minutes. But of my slow build songs, which I’ve liked on Spotify, which ones will fit this piece of the video first so that I'm most content with the finished product? That would be really, really cool.

So yeah, selection and supervision, absolutely. But there's a place for it. There are other places for us to work on in licensing before AI comes to the fore. By the time this comes out, we'll have launched a new product. If you think about Lickd as I described it, it's essentially the first consumer-facing mainstream music licensing platform in the world, right?

What we're launching is a B2B version of that. I don't know if you've seen all the lawsuits in the space at the moment around copyright infringements by brands on TikTok and Instagram.

Michael: Mm, I know it's a bit of the Wild West right now.

Paul: It is. And there’s confusion in the space, because the social space just keeps evolving, and the lines keep getting blurred year after year. What’s content? What’s paid? What’s promotion? Who’s an influencer? Is an influencer a brand? Is a brand an influencer?

You and I, Michael, can use whatever song we want on our Instagram and TikTok. But if we get paid by a brand, we no longer have access to that music. It suddenly requires a license.

Because of the presence of brands on socials, and the graying of that space with branded content and influencer marketing—where brands say, “It’s not just going on my channel, but it’s going out on everyday people’s channels”—they think they can use what they like. Well, no. You’re paying them. This is branded content. This is a commercial endeavor. It requires permission. It requires a license.

I have some empathy with the brands—not a whole lot, but some—because if you try and get a license to use a song, you’ll go to a label. If you’re lucky, they’ll get back to you. If you’re really lucky, they’ll end the quote process by saying, “By the way, you also need to clear the publishing. We have no idea who owns that publishing, so good luck. Go and contact the seven multinational companies you don’t know yet. If they all respond in time, come back and pay us an equal share of what they quoted.”

That’s the six-to-eight-week process we spoke about. Meanwhile, people are putting out content three to four times a day. So even for organic branded content, there has to be a much quicker, more streamlined licensing solution.

This is what we built—we’re calling it Lickd for Brands. The same way you’ve got Spotify and Spotify for Business, there’s Lickd, and now Lickd for Brands.

Lickd for Brands offers brands a platform that houses major label music—because major label music is not available to brands on Instagram, TikTok, YouTube Shorts, and so on. It enables them to ask permission to use a song and notify every owner of every percent of that song simultaneously. The moment you click submit on your request, every rights holder in the world with a stake in that song is notified in real time. They all have the information about your brand and your intended use, and can click yes or no in a portal on the back end of the platform—after discussing with management or the artist.

That means we can take a six-to-eight-week clearance process, which has led brands to either avoid mainstream music or just willfully infringe out of frustration, and bring it down to 24–48 hours. At the moment, we have the top six publishers on the publishing side, Universal on the label side, BMG on the label side, and by the time this goes out, Warner Music as well.

Even in private beta, we’ve seen L’Oréal, The Body Shop, Ray-Ban, Meta, and a couple of eyewear brands already using the platform successfully—licensing mainstream, major label music quickly, simply, cost-effectively, and legally.

That has legs, because artists are getting infringed and lawsuits are being filed left, right, and center. Waiting for infringement to scale and then slamming down the gavel is not a long-term monetization strategy. Ultimately, without a solution that’s scalable and streamlined, it’s the artists and music creators who lose out.

I’m excited about this product because it uses all the scalable technology we’ve built, the rights-holder partnerships we’ve cultivated, the data on ownership we’ve collated, and brings it all together for the further benefit of artists and creators by providing access to another revenue stream.

It’s really important people get on board with that. These platforms aren’t going anywhere. They’re only going to eat more promotional and advertising space, and need more music. If we can’t get it there, then AI will get there first. Stock music solutions will get there first.

If you look at someone like Epidemic Sound, they’ve built an incredible business. I’m impressed with them, and I know the people that created it. But if you want to know how a stock music company with just 30–35,000 tracks you’ve never heard of gets a $1.4 billion valuation, that’s how: because the music industry sits on its hands. The music industry fails to innovate, fails to accept technical solutions and opportunities.

The industry is 50,000 fragmented rights owners. They can never act as one. All innovation comes from third parties, and they need to work hand in hand with those third parties—saying, “Help us solve the problems of the future, and we will democratize our assets to you.” Otherwise, they’re in trouble.

Michael: Wow. Super, super cool. Paul, thank you for everything you’ve built—especially the foresight it must have taken early on, before you had those major players on board. It takes that kind of drive, what Steve Jobs called the square pegs in the round holes.

So thank you, on behalf of the whole music industry, for being one of those square pegs. Ultimately, it sounds like one of your main purposes is to make it easier to do the right thing. Because if it’s hard for people to license mainstream music, they’ll turn to routes that are easier but not as regulated or fairly compensating. This is super exciting.

For anyone listening or watching right now who wants to explore the platform more, what’s the best place to go?

Paul: The core platform is lickd.co. The new one is brands.lickd.co.

Look—it’s important. I said before, our mission is to democratize music for the world’s creators. In 2016, I meant YouTube creators, but I always wrote “creators” with a small C. Today, brands are creators, and they offer artists and musicians an enormous untapped revenue stream. In the not-too-distant future, people building in the metaverse will be creators, and that’s another untapped revenue stream.

The digital landscape will always evolve. Every time there’s a new fissure, stock music gets there on day one and capitalizes. AI could get there on day one as well. The mainstream sector needs to get there at least on the afternoon of day two—not in year three, which is how it acts now.

That’s where we see our role. There are good people at the majors who want to innovate, but there are lots of layers of bureaucracy. So if you’re listening and you want to innovate, and you like our mission, come on board.

Michael: Awesome. We’ll have the links in the show notes for easy access. Paul, I really appreciate you taking the time to be on the podcast today and share about your mission and how you’re helping revolutionize the music industry.

Paul: You’re very welcome, Michael. It was a pleasure speaking with you. Thanks for your time.