TheArtist Manager'sSpotify Ads Playbook
the exact system we use to run Spotify ads for artists and generate millions of streams
this is designed to help you stress less about the results you are getting with your artists
your stress levels
your client's happiness
Disclaimer
There are no guarantees in the world of advertising, and this playbook does not guarantee any kind of results. Part of the advertising process is learning through trial and error. Your results will vary based on your skill, experience level, and the quality of your material. If your ads are not working, there is a multitude of reasons they could not be working. Advertising tactics and settings may also change over time as the technology advances. Please do not make any financially irresponsible decisions with your ads.
The basics
Figure out who your song is for
Try to get specific on the person you think is going to resonate with this song. Pick one song, not the album, and finish this line: a song for people who ______.
Your song and your creative are king
Advertisements cannot fix a bad song or a poorly made ad.
The point of the ad is to convertI need this right now.
Music videos are not always the type of ads that convert. It depends on the artist and the campaign, so the ad will not always be a work of art.
Be patient
Do not make emotional decisions about your ad. Make logical ones.
Run ads for saves, not streams
People who save the song are more likely to stream it again later. More saves and more streams mean the Spotify algorithm recommends you more.
Test one thing at a time
If you test too many variables at once, it makes it harder to make decisions.
Only look at metrics that matter
Cost per result and saves per day. They are going to be your two best friends.
Use your judgment
These are not hard rules, but they are guidelines. Treat each campaign like its own unique situation, using these principles to guide you in your decision making.
The path a listener walks
This is what the user experiences from your ads.
The ad
The video shows up in their feed. It does not post on your main page. (unless you choose to run a post as an ad)
Landing page
Built on SubmitHub. Loads fast on phones, filters bots.
Spotify
The real track link, with the tracking parameters left on the end.
The save
Counted in Spotify for Artists.
* Pixel = some code that tracks when a user does something. This is how ads managers track results.
What do I make for the ad?
There is no single format that wins. These are the three that convert for us most often, and which one works is different for every artist and every song.
Split the music video into sections to see what works.
Performance footage, a phone shot in a bedroom or a rehearsal room, anything that feels authentic instead of feeling like an ad.
Clips that fit the mood of the song with a hook written across them. For example, a visual of a sunset with on screen text that resonates with what the listener is feeling. Often the cheapest to make and often the winner.
Static images almost never convert on these campaigns. And do not assume the polished option wins: one country artist we work with has had his best ad for months built out of video game footage, but another had his best video come from a selfie he shot on his iPhone.
How do I set up the campaign?
Build it in Ads Manager on a desktop. Do not use the boost button in the app.
Each ad set is one round of testing. The campaign holds the budget, and every round, or every type of test you run, gets its own ad set. That way you are reading one test at a time instead of guessing which change moved the number.
Some technical settings
Countries. We run tier 1 and tier 2 together on almost every campaign. Both go in from the start, and you let the results tell you where this artist actually lands.
Tier 1
Tier 2
How much should I spend?
Money buys you information, and information is how you learn and improve the ads.
If you wish to start at $100 a day or more, the biggest thing that changes is the frequency of testing and the amount of creative you would use. You can also test creative far more aggressively. A larger budget lets you run more variants at once instead of one test at a time, so you find out what works in days rather than weeks, and you can keep a winner running while you test the next thing beside it.
Meta will determine how to spend your budget. Your job is to interpret the results.
Recommendations
Never commit a budget that you cannot sustain for a long period of time. The minimum recommended time spent running ads is three months, and the best results come from running them consistently, day in and day out.
How do I test ads?
You are a scientist. You have to form your hypothesis and then test the variable.
Find the best-performing section
Your chorus will not always be the highest-converting section of your song. Split your song into sections that make sense and are 15 to 30 seconds long max. Let the audience tell you what converts for them.
Which hook
The hook is the first 3 seconds of the video. That can be the on screen text, the audio, or a visual moment. Any of the three can be the thing that stops the scroll.
Our favorite hook to start with is for fans of x, y and z, using three recognizable comparison artists. Boring that works beats clever that does not. Never run out now as a hook. If you are a smaller artist, people do not even know who you are yet, so just give them the thing they want immediately.
Which visual
Raw often beats polished, because too polished looks like an ad. Static images almost never convert for Spotify ads. One country artist we work with has had his best ad for months built out of video game footage, but another had his best video come from a selfie he shot on his iPhone.
Name every ad something that makes it easy to know exactly what you are testing. That way you can open Ads Manager and immediately see what is not working.
What numbers should I look at?
Cost per result tells you whether Meta found the right people. Saves tell you whether those people actually wanted the song. You need both, and you read them in that order.
Number one. Cost per result
Cost per result=what it costs to get one person to click through the landing page to Spotify
The goal is to get the right kind of person to convert, for the best possible cost. It is not to get the lowest cost.
* Sometimes artists come to us and mention that they are getting extremely low costs per result, but they forgot to check their Spotify for Artists. When they do, they realize their campaign setup was wrong and was not actually generating the results they needed.
Scale this upMeta found the right people. Our goal is to get campaigns to $0.20 to $0.30 per conversion.
It dependsWhat to do here depends on whether you are generating saves. This will require you to make judgment calls based on your circumstances.
Change somethingSomething probably needs to change here, even if you are getting saves.
* If you are getting saves per day at this cost, figure out how to take the findings from your winning ads and bring the cost down so your ads become more efficient.
These will vary campaign by campaign and song by song.
How many saves per day should I be getting?
You can find this in your Spotify for Artists account under your song stats.
If the cost per result is fine and saves still are not coming, you reached the wrong crowd.
This is a normal healthy day for most artists on a starter budget.
Scale it by duplicating the winner into a new ad set, never by editing the live one.
Outside of cost per conversion and saves, check streams per listener. If you have a streams per listener of 2 or higher, you are doing a good job, although this will not always be the case.
How do I know what to do during the campaign?
Running ads comes down to testing and making decisions. Here is a framework on how to make decisions with your campaigns. Please note that this method changes slightly when you spend above $100 a day.
1 First, what day is it?
This is the number of days into a test, not the total number of days the campaign has been running.
This is for budgets under $100 a day.
2 Before you change anything, one question
3 Then, which way is cost moving?
If saves are coming in on Spotify for Artists and it is reaching your goal, duplicate and scale the winning ad, then keep testing more hooks and more creatives so you can build a library of winners.
First, check the ads themselves to make sure they are actually quality, then run the three checks below.
4 If cost is climbing and not going down, check these three
We have had campaigns fail on one track for months, move to another song from the same artist, and start converting in week one. It is a step in the process, not a verdict on the music.
5 Two sizes of change
That last line is the important one. Every major change restarts the clock, so a change never costs only the change. It costs another week or two of not knowing.
6 I have a winning ad. Now what do I do?
Build a library of winners.
Advertising is a lot like mining for gold. If you found your first gold nugget, you would keep digging to see if there are any others. A winning ad is that first nugget. Scale it, keep it running, and test new hooks and new creative against it. Anything that beats it or holds its cost goes into your library. Anything that loses gets turned off.
- Keep running hook and creative tests. That is how you keep finding new winners. Put each test in its own ad set so the winner keeps spending while you learn.
- Room in the budget? Scale it. As new winners prove themselves, raise the budget behind them.
- No room? Hold the budget and keep testing. Keep spend where it is, and keep testing new ideas against your current winners.
This library is one of the most important things you will build. Every new release and every new campaign starts from what already worked, and it keeps paying off over the entire course of a career.
Before you launch
The checklist
- One song picked, and the "who this song is for" sentence written.
- Landing page built, with the pixel on it.
- Same pixel inside the campaign.
- Ad account and pixel active.
- Spotify for Artists access in hand.
- Built in Ads Manager on desktop, not boosted in the app.
- Stories and Reels placements only. Advantage+ audience off.
- A daily budget you can hold for three months.
- Round 1 live: one visual, four to eight song sections.
- Every ad named after what it is testing.
- Calendar reminder to look in 7 days, and not before.
How do I know this is going to work?
Real accounts, run on exactly the system in this playbook.
Don't want to run this yourself?
We are a team of musicians who actually care and run a tight system. We run this exact process for artists every day.
Book a free consultation to get help and start improving your ad campaigns today.
Book a free consultation