The Niche Economics of Music Streaming and Creation

Many of us have the inkling that each time we listen to our latest obsession on Spotify, that act generates some infinitesimal amount of revenue for the creators of the music… but for those who are curious to geek out on a fuller picture here is a breakdown of the niche economics of music streaming and creation. 

This overview is based primarily on my dabbling as a dance music producer over the last four years and triangulated through other artists/music industry people and public research. While well intentioned, there is a roughly 100% chance I will be missing things and some numbers will be more off-ish than others. Hopefully you pick up a nugget and maybe smirk one or twice nonetheless. 

The sections are: top-line revenue, costs & players across the music creation ‘value chain’ (there are a lot!), equity ownership (labels!) and economic analysis (artist profitability, moats and motivators). 

Three practical notes: (1) while most full-time artists rely on income from live shows, merchandise and other areas, the focus here is on streaming. (2) Spotify is used for most examples instead of Apple Music as it is the leading platform (and also probably for convenience as I am a Spotify user). (3) There will be run-on sentences and probably too many parentheticals. 

Top Line Revenue:

Generally speaking, each stream (30+ seconds of listening) on Spotify yields ~$0.003 to $0.004… so 1,000 streams is ~$3-4. This rate most depends on whether a listener has Spotify premium or not and which country they are listening in.

Because Spotify’s monthly premium cost ranges from $0.56 in Nigeria (Although not for long! A 44% price hike was announced for the end of the month)… to $11.99 in the US… to $16.48 in Switzerland, the payout rate for a stream will reflect these differences.

Spotify also offers a “Discovery Mode” program to many artists which goes something like: ”Hi Artist! If you sign some of your songs up for this mode, we will make them pop up more often on Spotify radio and autoplay and you will get more listens. For this service, we will just take a 30% cut on radio and autoplay streams. Deal?” The artist will usually say “Ok!” and put some of their popular songs in the program. Typically, what the artist is losing in rate they make up in total stream volume, so the program is (for now!) beneficial to the artist. Discovery Mode is also beneficial to Spotify as it gradualllllly shifts more streaming volume to a lower cost base (cue Syndrome from the Incredibles “if everyone’s Discovery Mode… NO one’s Discovery Mode hehehe”). 

Illustrative Examples

So, if a hypothetical international pop superstar like Taylor Swift has 90M monthly listeners and the typical Spotify monthly listener yields ~2-4 streams (see footnote 1)... say 3… that would be 270M monthly streams. If we apply a rate of $0.003/stream that yields $810K or nearly $10M/year. Spotify owns roughly a third of the music streaming market; so you might estimate total streaming revenue at 3x Spotify earnings or ~$30M/year.

If a dance music star like Fred Again or Steve Aoki has 15M monthly listeners, the same logic would produce an estimate of $135K/month or $1.6M/year from Spotify… and ~$4-5M across streaming platforms.

For a very, very amateur weekend producer like SD Gloria, 20K monthly listeners would produce an estimated $180/month or $2.1K/year… and estimated ~$6K across all platforms (see footnote 2).  

This means for someone to generate an income of $50K / year from streaming alone, they would need an estimated $17K from Spotify or ~150K monthly listeners. To generate that from Spotify alone would require ~450K monthly listeners.

------ 

Costs:

While one could (and probably plenty of people have) spend $20 or so for a music distribution service and upload their beatbox iPhone recording to Spotify with a random selfie as the album cover …. there are often many more costs involved in the creation and distribution of music. These costs vary a lot by whether people own the tools/expertise or have to hire a contractor (classic opex / capex trade-off). They also vary by level of professionalism desired, demands of the genre (do you need live instrumentalists?) and lots of other factors. Below are the most common costs, split up into Opex and Capex. Because I probably like frameworks too much, opex is further split into music creation and distribution/marketing. 

Opex - Music Creation

The cost of creation depends on how much of it you are doing yourself … and I will therefore divide this into “direct” production costs (i.e. for an electronic music artists with just a laptop and a dream) and “outsourced” production costs (i.e. for a pop/folk fusion artist requiring a recording studio, outside producer, other instrumentalists etc.). Now I am done with frameworks… thank you for your patience. 

Direct Production Costs - common services when an artist produces most of the track themselves

Sample Platforms- ‘audio ingredients’ of a track: To have nice sounds to put into a song…. from drum hits to birds chirping to no-copyright vocals…. many producers subscribe to one of two sample websites that allow a certain number of sample downloads each month. These two players are Splice and Loopmasters (with a longtail of smaller competition), with Loopmasters charging $8/month (or a discounted annual plan of ~$50/year) and Splice charging $10/month. The samples on the platform come from a fragmented base of sample producers - many of whom are also artists using sample sales for a bit of extra income.

Mixing and Mastering- Mixing is the process of balancing all the parts in a track (e.g. making sure the drums are neither too loud nor too soft, making the vocal fit with the instruments) and mastering is the process of making sure the overall song is appropriately loud. Highly technical and “mission critical” for a song to sound polished, mixing and mastering are often outsourced to audio engineers by new and experienced artists alike. 

Mixing solutions are provided by freelance audio engineers largely discovered via good old word of mouth or platforms like Fivrr. Prices per track range from ~$30 (typically provided from a lower cost of living locale) to $5,000+ for renowned engineers. 

Mastering is often bundled with mixing, or can be bought individually through a freelance audio engineer. Unlike mixing, mastering has also seen a rise in AI solutions- like Landr, which charges $7 for an online AI master. 

Let’s call a budget cost for a mix/master $50/track.

Outsourced Production - common services when the artist is primarily a vocalist and does not produce the track themselves

Recording, Other Artists and Outsourced Production- This is where genre plays a large role in the creation cost of a track. 

For example, imagine you are a small (very small) electronic music producer. There is no need to hire other artists or rent a professional recording space because generally the track inputs are coming from samples that you got from Splice or Loopmaster. There certainly also is no need to hire a professional producer to build and arrange the track because, generally speaking (see footnote 3), you are (trying to be!) the producer. Therefore, for many electronic artists this section is largely irrelevant.

If, however, you are an aspiring pop/folk fusion star this could easily be your largest expense line. To bring your song idea to radio-ready life you would likely need a whole gamut of solutions: A rented recording studio to lay down your vocals and guitar line, a producer to build out an instrumental and arrange your masterpiece, a fiddle player to add fiddle flare and so on. 

A recording studio can cost anywhere from $50 to $200+ / hour … with $100/hour being typical. An efficient 3-hour session would bring this to $300.  

The recording segment is location-based (so far, you cannot benefit from someone else’s expensive microphone without singing into the microphone) and fragmented, with very few multi-location players. The most (and potentially only) quasi-‘scaled’ player, Pirate.com studios, operates 35 locations across England, Ireland, Germany and the US representing 700+ studios (i.e. the actual room artists rent to record). The vast majority of the market consists of small independent studios ranging from your neighbor's home studio rented on Facebook marketplace to the iconic Abbey Road.  

Let’s shift to outsourced production and other musicians. A producer will typically commit 15+ hours to building out a song for their customer (i.e. the folk/pop fusion vocalist) including revisions. This is sometimes done virtually especially with producers in a low cost locale and rates therefore could range from ~$10/hour to $500+/hour for top tier producers (someone like Calvin Harris would probably charge a whole lot more). $300 ($20/hour * 15 hours) can be used for illustration. A professional musician might charge a high rate/hour but only be needed for a few hours (often requiring the rented studio space): for that fiddle flair one can assume a 1 hour, $100/hour session. 

The producing and ‘other artists’ segments are peer to peer- with a handful of players serving as marketplaces. While Fiverr and Upwork do not specialize in music services, they are a major option especially for lower-cost jobs. Soundbetter (owned by Spotify) is a music focused marketplace leaning towards more credentialed and consequently more costly producers and artists. A second music-focused marketplace is Airgigs, fairly similar to Soundbetter. There is a long tail of other, music-specialized players, sometimes with more niche leanings (i.e. ‘BeatStars’ is more hip-hop oriented).

For our illustrative, budget friendly-ish pop/folk recording - the total comes to $700/track. 

Opex - Distribution and Marketing

Once you have your masterpiece, these are the required costs of distributing it to streaming platforms and optional costs of marketing it.

Distribution services: To upload a song on Spotify, artists not using a large record label must use a distribution service like DistroKid, TuneCore or CDBaby (the three leaders of the independent music distribution industry).

Distrokid currently charges $20-80/year for its plans depending mostly on how many artist profiles are needed.

Playlist Marketing- Once a track is complete, website platforms exist to connect artists with relevant playlists (mostly Spotify playlists) that might feature their tracks to introduce the music to new listeners and increase the odds the track is supported by the ever-elusive black box that is the Spotify algorithm. 

The dominant provider, Submithub, is a marketplace where artist pay a few dollars to submit their song to a playlist and the playlist curator is paid half of the fee to listen to at least 60 seconds of the artists’ song and write a few sentences of feedback if they reject it (and ~7 times out of 10 they do reject it). An artist might spend an average of $3/playlist submitting their track to 10 playlists to support a release: $30/track in total. Submithub also offers artists the chance to send their music to Instagram and TikTok influencers who only charge if they use a song. The cost for influencer placement has no cap and can be well over $100.

Cover art and supporting video- Every track requires a corresponding cover image, and Spotify encourages the use of background videos to boost engagement. Much like mixing and mastering, this is also a freelance heavy landscape with a wide range in costs. Many artists opt for a DIY solution but a similar cost range to mix/master can be applied at $50/track.

In all, the cost of doing business for direct production with this full suite of services starts at ~$70/year and $130/track. For outsourced production, costs could run $20/year (only Distrokid would be needed) but a hefty $780/track. 

Capex

These are the tools required for someone to make their own music. If our pop/folk fusion star hires a producer and rents a recording studio, they might not need to purchase anything. If a different artist only pays for distribution and marketing, they will likely need almost all of them: a classic trade-off between asset heavy and asset light models!

DAW (digital audio workstation)- This is software producers use to make the music. From a vocal recording to a digital piano to drum samples... all the elements would be arranged in the DAW to create the end track. The big three options are Ableton, FL Studio and Logic. 

While each offers different versions, the standard cost is ~$500 for Ableton and ~$200 for FL Studios and Logic. There are a host of other DAW options, including a few free ones, but the vast majority of producers use one of the big 3. Just like the iPhone, every few years a latest-and-greatest version will come around for an upgrade fee. 

Plug-ins (software)- aftermarket digital instruments and tools that are added into the DAW: First, a touch of background. Each DAW comes with a stock set of: 

1. Samples (i.e. the drum sounds or background birds that the sample platforms offer) 
2. Effects (i.e. echo, reverb and any other sound modifier)
3. Digital instruments (i.e. a piano or synth you play with your keyboard or "click in" notes)

For each of these three categories, there is a market of aftermarket offerings promising to do what stock features cannot. This goes like “I want my drums to be punchier and can’t figure out how to do it with Ableton’s stock effects or samples…. but my fellow producer friend told me about 'XYZ Killer Drums plug in' so I’ll go buy that.” While aftermarket samples primarily flow from the two main subscription services mentioned earlier, aftermarket audio effects and digital instruments are largely purchased outright by a wide range of providers and can be even more expensive than the DAW’s themselves . 

To illustrate this, here is a selection of some of the most popular audio effects and digital instruments and their price. While they are often discounted, it is clear that the aftermarket plug-in spend can quickly surpass the cost of the DAW itself.

Audio Effects

• Waves- Autotune: ~$200
• Soothe 2- Mixing tool: ~$200
• Valhalla Reverb- Various reverb tools: ~$50
• iZotope Neutron- Mixing suite: ~$400
• FabFilter Pro-Q- Filter tool: ~$200
• Gullfloss- Mastering: ~$200

Digital Instruments

• Xfer Serum- Synthesizer: ~$190
• Native Instruments Kontakt- Multi-instrument: ~$200
• Native Instruments Massive- Synthesize: ~$200

There is a bona fide ocean of other plug-ins, spread often through word-of-mouth (or forum message). 

Plug-in spend varies a lot, but once discounts factor in, $500 should get most producers equipped to a standard level. 

Hardware- From studio monitor speakers (sorry roommates) to microphones to powerful production-ready laptops (DAW software tends to take a lot of processing power) to physical instruments 🎸, there is no shortage of musical ‘hardware’ available for the creation process (just walk in a Garage Center!). This is not the place to begin the slightest tear down of that massive segment- so let’s assume a $150 purchase of a cheap but decent mic and audio interface and move on.

Ownership:

For the average amateur artist, the ownership picture is very simple. They bear all costs and have all the upside if a song does well on streaming services. Things become a bit more complicated as labels enter the picture. 

A label acts like a venture capital firm. Our aspiring pop/folk fusion star could send a demo pitch to the label who loves it and says “we will pay for this to get professionally produced and use our industry contacts to promote it and in return we will take 70% of all money earned from the song… and maybe also a portion of shows and merch income”. The artist will think: “wow this is great because I would have trouble paying for an ok producer and renting recording space, and this label will cover all that and help me get the song out there!”. Then perhaps they will negotiate, especially if the artist has multiple labels interested. After they sign an agreement, the label will hope the song succeeds amazingly to offset the other songs they have funded which are doing somewhat less well. 

If an artist has no incremental costs to produce their music (i.e. because they make it all on their laptop with no recording or other specialized help), the label offering to cover the costs is less compelling and therefore the main value proposition is to help with promotion.

As with venture capital, part of the value of the label can also be the ability for an artist to associate themselves with a well-regarded label. The artist might think: “hmmm maybe I could probably promote this song independently and not give up a big stake of the proceeds, but having one of my songs signed to FANCY LABEL RECORDS will probably be helpful to my career later and look cool and maybe I will get to meet my hero EDM producer who is with the label.”

Economic Analysis

Artist and track level profitability

So when we round up all these costs- what is the overall picture for our beloved hypothetical artists? Well, as we see in this SUPER illustrative cost summary, to break even on a per track basis with streaming revenue, our small amateur electronic artist would need to generate $130, whereas a rising pop/folk fusion vocalist would require $780. Applying the $0.003/stream estimate gets us to 43K and 260K streams respectively to generate margin on the track. They would need more streams to cover their annual costs and generate a reasonable return on their capital investments (i.e. pay for Ableton).

For most aspiring, small independent artists, this is really hard! How hard? Well. Spotify has a cool calculator which shows you how many artists have varying levels of listenership. You can ask the calculator “hey calculator how many artists have 10 or more monthly listeners?” and it will say “about 4M”. Lots of fun. Anyways, the chart below is what the calculator said alongside some very illustrative, rough, high-level, exemplary (+ add caveating word of choice) circles (see footnote 4) showing at what # monthly listeners an artist would cover costs or earn a living (caveat, it depends where they live…. caveat, it depends what sort of music they make… caveat, caveat, caveat).

Ok, without doubt, a portion of the 4M artists in the far left bar are likely not particularly active, but even if only 2M are active… it seems just ~10-20% of streaming artists are covering their expenses… let alone generating a meaningful return on all that plug-in investment. 

Right-to-Win:

So what gives an artist a right to win, skiing down and right on the slope of a bar chart above? As a very (very) small, amateur artist I clearly do not have the answer. But I can guess! Here are two guesses that work together.  

Guess 1: Differentiated and Compelling Product: 

A track generally succeeds because it makes people feel a certain way and they want more of that feeling so they save the song and even add it to a playlist and if a lot of people do that the Spotify algorithm says “hmmm this is maybe a cool song for XYZ demographic so I will show it to them so that they will like Spotify and be 0.000001% less likely to cancel their membership”. 

How do you make a song that makes people feel a certain way that they want more of? As Mason Ramsey (young, Walmart-yodeling sensation, for the uninitiated) demonstrated, natural talent plays a meaningful role. This is not only talent in developing beautiful sounds, be it as a singer or synth maker…. but also in having a personal taste that resonates with some broader audience (see footnote 5). Thankfully for myself and other “less-than-mason-ramsey-level” talented artists, it seems experience helps too. Experience means you know how to get the song in your head into a track more quickly. It means you know how to get an acceptable mix and master. It means you have a process that helps you (finally!) finish the track and release it. If you are a cool artist (i.e. not me) it means you have developed unique samples and ways to process sound that achieves an end outcome few others can. Like with any industry, it means you messed up a bunch already and are probably less inclined to mess up in the exact same way again. 

Guess 2: Adequate Distribution: 

This means that you have a reliable way of getting the track to enough people that the Spotify algorithm has sufficient data to decide what to do with it. In my case, this means over the last four years I have collected a list of fitting playlist curators to send new music to. 

Also- it is easier to click on a new song from an artist you have heard of (and like). So, as an artist is exposed to more people and gets more followers there is more natural momentum for new tracks.

Returning briefly to labels, their job is to entirely take care of the “Guess 2” part AND mostly take care of the “experience” half of the “Guess 1” part. They have to choose artists who bring the “talent” half of “Guess 1”. 

Barriers and Moats:

What are the barriers to entry for an artist putting music on Spotify? An annual $20 distribution service subscription. Barrier, not barriers. While moats for highly profitable artists/labels certainly exist (one could argue the moat held by the likes of Taylor Swift is comparable to Meta or CostCo)… the barriers for people to actually enter the music industry are for sure lower than before the streaming era. A record store had room on its shelves for a few thousand artists (equivalent to only Spotify artists with >5M+ monthly listeners). So practically all the people who aren’t turning a profit (and most of those who are!) would not have a clear way to share their music broadly prior to streaming. 

 For all the flack the streaming industry gets, this is a cool thing. Transition to the last point.

Motivation and Conclusion:

Ok here are the main points:

• Spotify generally pays $3 per 1000 streams (+ a million caveats)
• There are a boatload of costs involved with creating music (especially if you are not doing it all yourself on your computer in the basement… ahh the dream!)
• Making the numbers work is hard (super hard!) and most Spotify artists are unprofitable.
• But! These artists get to have their music out in the world and that would not have happened before streaming

So what to make of this? I would propose the following choose-your-own-adventure of a conclusion (see footnote 6).

•  If you are an artist (especially one that requires a recording studio rental and outside producer!) viewing music streaming as a largely-passive cash printing machine that will set you up with generational wealth… I am sorry. Your vision is technically not impossible, but you will probably lose money and be bummed.

• If you are an artist viewing music creation as a personal past time… like amateur painting or pottery or carpentry…. and you see the whole “uploading to Spotify thing” as “well I put so much effort in, I like the song, might as well share it with the world and see it offsets my Spotify subscription!”. If that is your view, you will probably lose money but not be bummed because it will feel like spending money on any other hobby. 

So here’s what I might takeaway from all this:

For amateur artists: if you don’t love the music creation process as one would love painting or horseback riding …. you probably win by doing something else? Yeah, probably. 

For non-small artist readers, IDK, I hope you enjoyed it? Here is a possible application. There is probably a lot of money to be made selling shovels into the gray space between people digging sandcastles (pure hobbyists) and construction workers (pure work). This is because people are willing to spend a lot more on a “hobby” if they can mentally justify it as an investment in their business/career. If you have no way to make money from an expense it is a purely indulgent purchase and that is ok but we are oriented to limit those. If you DO have some way to make money from an expense- even if it is objectively a bit far fetched- you can reroute that expense as an ‘investment’ and now that spend is in another less limited mental bucket. I would venture to say a lot of the cost bucket industries above are the biggest beneficiaries of the shift to streaming and consequent eradication of barriers to entry.

Wow you made it! Thanks! If you have a perspective or comment, feel free to leave a note. Also if you have a delightful corner of niche economics you’d like to share, let’s talk… perhaps the 3rd write up will make its way in 2026. 

Footnotes.

1. This is industry standard but might be very low for someone like Taylor Swift given I am probably one of her less-active monthly listeners and still usually send more that 2 streams/month her way (Style is such an infectious tune) Also: while the ratio of streams to monthly listeners is generally 2-4, this can vary a lot by artist. A niche punk rock band with 100K diehard monthly listeners could generate more streams than a background chill beats producer with 200k monthly listeners from large playlists and Spotify radio.
2. Though generally the smaller the artist (and artists like SD Gloria are really quite small) the less reliably listenership will be proportionate to overall streaming service market share (i.e. if Spotify’s playlists/algorithms disproportionately impact listenership so while Apple Music stream count is dramatically lower than Spotify)
3. There are rare cases where an electronic music artist will out-source the production of their music to ‘ghost producers’ and act more as a creative director. 
4. You may be thinking “Wait I thought for a little house artist it only takes $200/year to put out a track and pay annual expenses. Isn’t that like 66K streams? And isn’t Spotify a third of total streams so it is only ~22k streams on Spotify? And isn’t that like 1.8K streams/month? And if each monthly listener contributes 3 streams isn’t that just 600 monthly listeners? Why is the green circle so far right?!” ….. To which I would say 1. Very astute. 2. See footnote #2 on why Spotify would probably be more than 22k streams for small artists. 3. With a few exceptions (shout out ‘Running Up That Hill’!) people tend to listen to newly released (or newly discovered) music then get interested in other songs after a couple weeks. So, to keep a stable (or growing!) monthly listener base most artists need to put out a handful of tracks a year, like 4 to 6. 
5. For example, a perfectly executed idea that involves random, dissonant combinations of middle school bell tones being triggered for 45 minutes might proceed to dominate its global market of 72 listeners. 
6. I think most amateur artists probably fluctuate between these two sentiments with a resting state somewhere in between… at least I do! Although, no aspirations for generational streaming wealth, just maybe moving from budget grocery money to slightly nicer grocery money.

References.

The Process - Loud and Clear

Spotify raises monthly premium subscription fees in Nigeria by 44% - Nairametrics

How Much Does Spotify Pay Per Stream in 2024