Building and implementing F2P music
Building and implementing F2P music
Exploring how business models and design systems can be applied
Today’s piece is a guest post from Shokunin.
We think the future of music is Free to Play. DNS.xyz, JUICE, and BlockScience are combining their game development, music industry, and data science skills to build the first implementation of this model. Following the high-level introduction by Dan Fowler, this article explores how the business model can be applied to music (and entertainment in general).
I - How F2P fits the music business
F2P in the context of music
When we introduced the idea of F2P music, one of the main points of feedback was that music is not a game, and fears around the idea that we might be trying to gamify music. We think gamification is manipulative and focused on conditioning the most compliant or addictive users. The last thing music needs is a couple of tacked-on features to increase pre-saves or engagement for one quarter.
I was one of the first product managers to introduce F2P in mobile games, in 2011. Games were similar to the movie industry: you’d work for 2-3 years on a premium product, and ship it once it was finished. The pipeline was designed to ship $60 boxes on the shelves of Walmart. For the sake of the analogy, we could consider albums to be similar products. With the adoption of smartphones, we saw a surge in demand for games that drove the prices down to $0 within 18 months. Customers expected free games. They also reacted well to regular updates (seasonal themes, monthly new levels, etc.). Apple encouraged developers to update their apps, via featuring in various App Store placements. The business started looking more like TV: a pipeline that would ship new content every few weeks, that doesn’t necessarily have a set end date.
So we took inspiration from SAAS products on the web: apps like Dropbox that had large free audiences and viral effects, where a few payers would cover the rest. Much like web apps (and unlike “ship it once” products), we had live data on users. What they liked, where they’d churn. F2P was made for SAAS. It was created to find a way to attract, retain, and involve large audiences of very different users in an app. Using real-time data and rapid iterations to deliver something for everyone. What if instead of just “premium”, we had consumables? Extra lives, costumes, power-ups, and levels that go away after some time.
Music As a Service
We see the shift from albums to singles, as a similar change from paid games to F2P. TikTok and YouTube are pushing this change further. Not only do they provide better data on your audience, but they also have more social features than the music platforms. Artists now experiment with early demos, engage their fans on social, and tweak things, which eventually becomes a track. The EP or Album represents that journey but is no longer the goal. You build over time.
To support this trend, we think creator platforms should work more like Steam’s Early Access. A game like Valheim will ship in a beta state. First, it might be for a small community of hand-picked users. This will then open to a public beta, where people can buy the game (usually at a discount).
Early Access is a great way to signal you want to get feedback from the community. You’re going to ship updates every week, try some ideas, and see if people hate them, before you commit to a full game. People will give feedback, leave reviews, write guides, and also data on what works.
It’s a great way to find your supporters. Release a demo, get feedback, and ship a new version (but let your fans keep the demo to show they were there). Bring fans into the creative process. You don’t want them to design for you, but you want to quickly see what works. Reward them with exclusive skins, and badges that show they were an OG. Some go as far as including them in credits. Once a platform offers this mix of social interactions, small increments, and data, you’ll see some fans become not just listeners, but participants.
Streaming focuses on the wrong metrics
Music platforms have managed to cap Average Revenue Per Paying User (ARPPU) to $10 monthly. In exchange for this, you get 80 years’ worth of every track ever. Imagine getting every single video game or every app ever released on the App Store plus in-app purchases.
When it comes to reach, music is only second to social platforms and vastly outperforms other entertainment. In 2022, 616M people paid for a music subscription. For perspective, this is just 3M users less than if you add Playstation Plus (48M), Xbox Game Pass (30M), Netflix (238M), Disney Plus (146M), and Amazon Prime (157M) together!
You can still buy games for $60, rent a movie on Apple Movies for $10, or buy the Blu-rays. Music is the only industry to have given away the whole thing for $10. In addition, churn is at 30% year-over-year, forcing the streamers to push for high growth (which creates pressure to lower ARPU). These users are usually lost to YouTube, where people will engage with the artist and have fun with other fans.
We believe these metrics indicate too much focus on attracting low-spenders. There is a lack of attention to retention: with time, the small increment in new music means that you’re paying for a product that offers you fewer new things to enjoy. There are diminishing returns to subscribing or coming back daily. Because of this, we believe that a model that caters to different personas, and offers multiple ways to engage (and pay) is more likely to fix the engagement and retention issues.
II - Designing the system
Loops and Personas
In order to get a better understanding of how to address user cohorts and different personas, we like to use two models from the games industry: the 4 Kinds of Fun by Nicole Lazzaro, and Game Thinking by Amy Jo Kim. The former will be explored in detail in a later article.
Amy Jo Kim’s framework can be summarized as, understanding what a user wants to get better at, and building a loop to help them get there. Are they new or experienced? Are they interested in some mechanic or another? It is designed to be agnostic and used outside of gaming. Let’s take a look at the work she did outside of gaming, for Slack:
Here we see Slack represented a typical Game Loop.
A user has a trigger: maybe they have some spare time between meetings, maybe they want to follow up with someone, maybe they’re bored at the bus stop. “I wonder if there are any new messages”.
The user does the core action in the app: for Slack, they read & respond to updates.
They get feedback: They follow up everywhere relevant. The app says “All done!”.
They make progress and get better at mastering the app. For example customize channels, notifications, or integrations.
As the user progresses, they improve their mastery. This is mapped as a “path to mastery” (sometimes called player journey), similar to user journeys you’ll find in UX or sales, but used to understand how people make meta progression.
This funnel’s goal is to get a user to be the best version of themselves, from discovery to mastery. For Slack, their intent was to make it more comfortable for people to either hang out on channels all day and socialize (have a channel for pets at the office, use emojis and gifs for responses), get productive (use apps to say when GitHub and Google Docs are updated) or collaborate (share content, help teammates).
You can’t make something compelling for fans before you’ve “found the fun”, the core actions that people will like. From there, later efforts can focus on bringing in more people, helping them learn the basics, and becoming superfans.
Building for fans
How can we make a compelling F2P music offering? F2P can take many different shapes. For example: Fortnite is F2P, and it's all about skins. Candy Crush is F2P, and it's all about power-up items to let you cheat at the level. Eve Online is F2P, and it's all about buying the most epic starship and lording over other people. In the context of music, the audience and activities that a local metal band would want are very different from the BTS Army.
We start by understanding what kinds of people our fans are. We’re big fans of Nicole Lazzaro’s framework (enough that it warrants its own article in the future). Spotify also has a good starting point with its 16 personality types, which are a music-flavored version of Myers Briggs personality types. This opens up ways to contribute and participate.
Casual fans will:
Play your free songs
Like and leave a comment
Follow your page
Enter low-effort quests or giveaways
Collect a song if it’s free
Creative fans will:
Collect songs
Make playlists
Write thoughtful comments and reviews
Want collectibles, avatars, badges
Contribute ideas
Buy limited edition items (especially if they unlock customizations)
Social fans will:
Collect songs
Participate in discussions
Discuss ideas
Attend your spaces and events
Comment and react regularly
Share your releases and updates
Buy limited edition items (especially if they can be showcased)
Competitive fans will:
Want rankings, leaderboards
Complete quests, objectives
Collect rewards, try to get the best ones
Want badges in recognition of their status
Collect expensive items (especially if they show status)
Artists can experiment with these angles to give aspirations to their fans. We think one way to do this is going to be via quests and challenges. Their cadence would be at the creator’s discretion: daily, for each release, seasonally. Quests could reward newcomers who like and comment or veterans who leave thoughtful reviews. Sharing a link to a review on social platforms. Listening to 10 songs or adding them to playlists. Collecting 5 tracks. Getting a collectible that only unlocks if shared with a friend. Posting reaction videos to your track could be a basic goal, but making remixes of your track would be for advanced users.
This segmentation also allows gating things. For example, you might want to release a song as a 30-second demo that you buy to unlock or perform quests to earn. A limited collector edition of a track could be for people who have been in the community for over 6 months, attended 10 shows, or are willing to pay. A creator can then decide to gate things with time, money, or any other way to contribute.
The platform’s goal is to make it easy for creators to make these ideas happen. Rather than simply shipping a track and promoting it on social outlets, they become game designers for their own fandom. Fans become engaged supporters.
III - Implementation
Our first music release
Enough theory. Let’s put our money where our mouth is! For our first music release on DNS.xyz, we wanted to explore the assumptions in this article and test whether F2P Music held scrutiny.
We were approached by Kane Mayfield, a respected rapper known for his releases on multiple blockchains and his pioneer work in AI music videos. His latest release, New Jack City, was his best video yet. He didn’t want to release on a regular social platform but didn’t want to use a regular web3 platform either as most of them aren’t good at serving videos. The track would be playable for free, but affordable to buy a collector edition so that as many new collectors as possible would join.
Web3 music tends to sell at price points ranging from $100-1,000+, which causes tracks to be collected by a select few competitive investor types. The opposite of our goal. We wanted to experiment with simple quests to encourage people to collect the track while keeping a low price point.
We landed on the idea of:
A song that is free to listen to on the website
A $10 price point to collect one edition of the song
A leaderboard with a tiered ranking system for rewards
The top 50 collectors would get the mp4, mp3, lyrics and a cute 3D avatar character
The top 20 would also get an exclusive remix of the song, unreleased anywhere else
The top collector would get their face and name in the music video, which would be updated for the occasion
Effectively, there was one quest: collect the song once, or multiple times, and you will climb the leaderboard. It was the only quest for this experiment, and it only focused on paid collections. This put the focus on social fans who wanted to be present, and on competitive fans who wanted to get the top prize.
What we’ve learned
In spite of the lower price point, this was Kane’s most successful release to date: 500 editions were sold to 100 collectors, nearly 5x the collectors, and 2x the revenue of his previous best. It accounted for 25% of music NFT sales that month, despite selling for $10 a copy.
The leaderboard launched 1 day into the event. Its introduction doubled sales velocity, even though sales had slowed down post-launch. Its inclusion was very helpful: he was able to use the updates in his social media activity (“new top collector! Oh someone dropped out of the top 20”), and participants were able to regularly check on each other. This rapid loop helped people be engaged and even gave them something to tweet about.
People used the multiple editions as a pseudo Patreon. These people were there simply to enjoy the fun during the 4 days of the release. The small perks were a bonus. Casual and random fans saw the song on social platforms and came to listen to the music. Whales were the bigger drivers. A few bidding wars happened in the top 5 users. We saw the power of social comparison, similar to the behavior in mobile games. He was happy with the result: “I’ve sold more copies with you in 6 hours, than I did in 6 years of streaming”.
The event did rely a lot on Kane’s activity on social platforms, promoting his release everywhere. It also only catered to collectors. There was no way for people to contribute or participate for free (besides leaving comments and retweeting). There was also no way for them to earn anything via effort.
What’s next
For further releases, we want to expand on the notion of the game loop. We’d like to have aspiring content for casual, social, and creative users. This means quests tailored to them, as well as small rewards that fit their personality types.
Ideas range from showing “I was here” badges and avatars on their profiles, showing your edition number on albums, creator collectibles (avatars, banners, emojis), badges for event participation, unlockable behind-the-scenes content, gated community posts, simple quests for newcomers (follow, comment and collect this free item), direct to fan thank you messages and award showcases.
We’re going to need to figure out with creators what kind of tools they’ll need, so they can offer this content without too much overhead. What data they like to get in real-time. How to manage which posts or rewards are unlockable, and based on which criteria. It’s going to look like the relationship between a game engine developer, and a game designer.
It’s also going to need a dialogue with the fans, to understand how to make the experience of listening, participating, and discovering fun. This needs a blend of social features and a player, which only TikTok and YouTube have managed to deliver so far. It hasn’t happened in music yet. This will probably take the form of us dogfooding our platform to get fan participation.
IV - What’s next
F2P Music isn’t going to be for everyone. It won’t make sense to re-release the Beatles back catalog and build a community for super fans. This model is going to be a fit for artists who want to connect with their fans and engage them day to day. It’s a way to help them build their own fandom. It comes with the cognitive overhead of figuring out how to engage the audience, make interesting quests, and deliver some rewards and content at a regular cadence. Because of this, we think it will need a combination of experimental and counter-culture artists who want to move on from the status quo, with communities who are very engaged on separate platforms.
In our next post, we will be exploring how F2P Music can be designed to engage the different kinds of personas who engage with a creator, particularly the 4 Types of Fun, and how this approach can lead to better data and feedback loops.
 | A guest post by
|