How we got 80,000 streams with $100 and a playlist

[00:00:00]We got 80,000 streams on this song using only $100 in meta ads and putting the song on our playlist. The playlist we put the song on only accounts for roughly 10% of the total streams for the song. And we actually ended the ad campaign early about after 1 week, but clearly the people who did hear the song really liked it because now we've gotten over 50,000 streams in the past 28 days alone thanks to all of Spotify's algorithmic playlists. If you want as many streams as this, this, and this song, well, this video is for you. We're going to look at the ad campaign setup settings so that you can replicate it yourself, and we'll talk about what the plan is for the song now that we've seen really good results despite a lackluster ad campaign. Let's start by looking at the Spotify for artist data for this song. So, we'll look at the last 12-month graph here. You can see that the streams for the song have grown substantially with time. So, the song has only been out for about 2 months, and if we look at the save graph, we can see that the saves at the beginning of the campaign came from the ad campaign that we ran.

[00:01:05]The saves after the campaign was over, and the campaign ended about April 9th, all are coming from the algorithmic playlists. The streams during our campaign were also at their smallest, and it wasn't until after the campaign was over that the song started getting boosted by Discover Weekly and Radio.

[00:01:23]Here we can look at the playlists for the last 12 months, and we can see here our playlist that we put the song on accounted for 10% of the total streams for this song. Radio and Discover Weekly are the biggest drivers in streams so far, and in the past 28 days these playlists accounted for 40,000 streams alone. If we take a look at the locations, so where are the streams coming from? You can see that the majority of our streams are coming from the US, Germany, Canada, then Brazil and Taiwan. So, the top three countries are all countries that pay a decent amount per stream, so they are desirable to have at the top. And countries that pay lower are not really dominating this particular song. If we go back to the 28-day view, we can see these spikes here. These are the Discover Weekly spikes. So, on Mondays, the streams jump because the Discover Weekly algorithm is refreshed and everybody goes to go listen to it. We haven't had a Discover Weekly boost the past two Mondays, so it'll be interesting to see if we continue to get boosts or if the big algorithmic push period is done. Here in the Meta Ads Manager, we can see when the campaign was running. So, it started March 28th, we were getting 20 to 40 conversions per day, which if you look at the saves for the first 28 days the song was out, this makes sense. We're looking for at least 20 to 50% of those conversions turning into saves, which we do see here. We can see we only spent 100 total dollars before the campaign was turned off, and that is because the cost per conversion was 52 cents. We like to keep our cost below 30 cents, and if an ad campaign with the simple artwork visuals that I will show you soon do not perform at 30 cents or below, we will typically turn off that campaign at $100 and then just see what happens with the song. In this particular case, uh we have easily made all of this money back considering the algorithmic boosts.

[00:03:15]80,000 streams is going to net somewhere in the neighborhood of 250 to $300, so we've already doubled or tripled the investment into this song with this initial small campaign. Obviously, the playlisting helps, too, and there's no way of knowing how much this campaign affected the song versus how much the playlisting affected the song. All we know is that we did both, and so there's some combination uh that's triggering the algorithm for this particular song.

[00:03:41]As you don't know, a Meta Ad campaign is when you pay Meta to show your video on Instagram and or Facebook that has a link in it so that people can click the link, go to a landing page, and then click a link to a music service where they will then listen to your song. All of those steps have to happen in order for you to have a successful ad campaign, and the idea is that these ads will trigger algorithms just like it did for this song here that we were talking about, and then make your money back in ideally 8 to 12 months. But, of course, in this particular case, we did probably make all of our money back within 2 months. Now, let's take a look at the settings for this campaign. So, it's a typical auction engagement campaign with a spending limit at $100 using the ad set budget. We used website conversion location, maximize conversions, same data set we use for every other campaign, same conversion event we use for every campaign, Hypeddit smart link click.

[00:04:36]We were doing $15 per day starting on March 27th.

[00:04:40]Again, here it says Advantage Plus is on. We have Advantage Plus audience off.

[00:04:43]To be totally clear, because I know it's confusing, it says Advantage Plus is on, but that's just a bug in my ad account.

[00:04:49]Advantage Plus is off, and yours should be off, also. We target all of the countries that have listed in the uh spreadsheet in the description of this video. Age is 18 to 50. Targeting for interests is Spotify narrowed by electronic dance music. Advertiser and payer is just our label.

[00:05:06]And we did not do Advantage Plus placements, as usual. We just did the Instagram feeds, stories, and reels placements. Every single one of the actual ads that we ran looks the exact same. So, if I click all of these different ads, you see the same visual.

[00:05:23]The difference is what part of the song is playing.

[00:05:30]So, in this campaign, the focus of it is finding the part of the song that works the best while using simple visuals that still makes it obvious that what we are promoting is a song, so that we can see if the song can stand on its own two feet when being promoted without flashy visuals or anything that is super eye-catching. There is some disconnect between uh how well the campaign did on Meta and how well it did on Spotify, but we'll talk more about that later. For now, let's look at the ad settings. So, here we just use our Facebook and Instagram pages, manual upload, single image or video, multi-advertiser ads is on.

[00:06:03]Our landing page just has Spotify on it.

[00:06:06]So, if we click that, it'll take us to Spotify and start playing the song automatically. No browser add-ons. The ad creative is just these videos that we showed you. We had a bunch of different text options. The The song kind of has like a classic Alan Walker feel to it.

[00:06:20]So, we just said, "If you love EDM, check out this song. If you love Alan Walker, check out this song. If you love K391, check out this song." So, different artists that we are mentioning that are similar to the Alan Walker style of music. Um and then also just some feelings like, "If you're lonely, this song is for you." Call to action was listen now. And then we had these enhancements on.

[00:06:41]I don't really pay attention to these. I just keep the ones that are on on and I typically don't turn any of them off unless they affect the sound of the video. For this particular campaign, we did do the creative testing tool. So, each one of these ads was an AB test with itself so that the campaign would spend a decent amount on each ad before giving up and just choosing one ad to shove all of its budget into. Now, the big question for this particular song is, "Do we try to push more money to it to trigger even more algorithms?" And it's a tough question, but before I go into what we are going to do for this particular song, I want to talk about the three reasons why you might spend more. Reason number one is the cost per conversion is under 20 cents. If I get under 20 cents with the simple artwork visuals that we use and I see the benchmark numbers being hit like 20 to 50% of conversions turning into saves and then at least a one-to-one ratio of streams, but typically it's a lot more than that, then I will run more money to an ad campaign. The second reason is if by the end of the first ad campaign or by the end of let's say this first $100, these streams are still rising. They have not hit a plateau. You can also look at active streams for this, but typically I like to see all streams together rising throughout a campaign. If that happens, I will spend more on it. The third reason is if a song gets more streams after the promotion is over, then I will run more money to it. So, this particular song is meeting criteria number three. It's getting more streams after the campaign is over than when the campaign is on. The problem is that the cost is much higher than my benchmark cost. Ideally, I'm under 30 cents per conversion, but to run more ad budget to a song, I want to be closer to 20 or below. But, I am at 52 cents with this song. So, I could spend another few hundred dollars testing different visuals to see if there is something that really clicks with Meta. In my experience, other types of visuals besides the artwork visuals do perform better if you make them well.

[00:08:46]The problem is that if the artwork visuals don't perform very well, it is a sign that even if you can get lower costs for other types of visuals, that it might not be worth promoting anyways.

[00:08:58]And so, the downside of putting more money into the campaign now is that we spend a couple more hundred dollars, the song does not take off in any meaningful way, and then we've just nuked the song's profitability. Now, obviously, a few hundred dollars is not life-changing money for everyone, but we still like to be somewhat conservative, and it's nice to be able to pay out to the artist who promoted the song. And we don't want to ruin their ability to get paid by promoting something that we don't think should receive more promotion. Another thing that tends to happen is sometimes you will see an ad campaign that doesn't perform well, and then it spikes the algorithm, and you're getting tens of thousands of streams. Then all of that dies down after a few months. So, what we're going to do for this particular song is not put any more money into it right now. We'll obviously keep it on our playlist, but we're going to watch to see where the streams settle after these big boosts are over. If it continues to get big boosts over the next month, we will probably find it worthwhile to add more creatives and spend more money on the song to see what happens. If the streams die down and go back to a much more normal stream count, then we'll probably just let it be and save our budget for this artist's next song. And to answer the question of why did this song trigger the algorithm like crazy, I really have no idea. It's possible that maybe it's hard to get people to get into the song, but once they do find the song, they really like it and repeat listen, and that can trigger Spotify's algorithms. But also, we've released multiple songs with this artist, and then every single song that we do release kind of gets pushed into the same audience that we always push our songs into, which makes it easy for Spotify to push our song via their algorithm to a bunch more people. So, we have really, really good algorithm pushes even with somewhat mediocre campaigns, and that's just because Spotify knows what to do with our songs.

[00:10:57]But also, I think the song is an absolute banger, and I listen to it a bunch, so I'm not surprised that the people who do discover it really do like it. And that's just how some songs are.

[00:11:07]Not every song will be the song that grows your audience from an advertising perspective, but it doesn't mean that it's still some people's favorite song.

[00:11:17]I hope this video helped you to promote your music better. If you want us to get these types of results for you, submit a demo to my record label using the link in the description of this video.