14,770 Spotify Streams for $443 with Meta Ads

In this video, I'm going to walk you through a music marketing ad campaign that my agency is running to promote this song on streaming platforms using meta ads. And the cost per conversion is down to 14 cents, which if you ever run a campaign like this, you know is actually an insane result. [music] And the way these campaigns work is we have ads that are running on Instagram, just kind of cool vibe footage or performance footage from an artist with a song under it, different parts of the song, etc. people click on it, land on a landing page, and then from there they click to their preferred streaming platform with Spotify being the dominant platform nowadays. Uh, and the song itself has 60,000 streams. We started about halfway through. Um, you can see this drop and then this inevitable increase here. Um, they did start a playlisting campaign in conjunction with the start of our campaign, but that doesn't impact the saves. So, in this video, I I walk you through um the entire campaign, how to analyze it, how to make sure a campaign is actually performing as good as you think it is, and actually some some cool tricks in Spotify artist to learn how to interpret campaigns, which actually led to me finding out that this insanely good campaign was was actually an AI song, which I did not see coming. Um, and you'll see in the video at what point that I that I learned that [music] >> if if you've ever looked in the comments of any one of my ad tutorials, you've definitely seen people like, "Hey, I ran this and I didn't get the streams that I was expected to get. I get way less streams than conversions." And that's what I mean by you want to make sure your campaigns are not too good to be true. So, we're going to go back into the Spotify. I'm going to show you how this campaign is is is set up. But what I want to kind of instill into you is a good process you can do every time you run a campaign to make sure that you're not just burning everything down by accident. So, uh, this is what I do every time I run a new campaign, especially if it's looking really good, but even if it's not, what I do is I use this thing called breakdowns. And I'm going to zoom in so you can see this better. Breakdown. And then I go through ages, countries, placements, and creative related media. to look for any red flags. So, if we go to ages, what I'm looking for here is I don't want to see anything odd, but also I don't want to see a bunch of 65 plus. 65 plus on metads for campaigns like this is very unreliable. Um, but even more so if your music doesn't make sense for it, right? I've had people where they're promoting like a hip-hop song and it's like all 65 plus and they don't get any streams. So in this case for this song um this age distribution does make sense that it would skew older. So it's skewing mostly to 45 to 54 but really the broad demographic is kind of like uh 35 to 54 which makes sense. But we we we do have a good spread across everything. Meta actually tried all the ages. So this this checks the sniff test. It doesn't seem suspicious. Next we go to countries. And here we're just looking for any countries that we're not supposed to be targeting. one, like if there's a country that you you didn't expect to be in here, but also if there's any red flags, like for example, is like 90% of the budget going to one country that's like really small. For example, it'd be really weird if 99% of the ad spend was going to the Czech Republic cuz Republic's a very small country. So like why would that ever be happening? It makes sense that the Brazil and Mexico and South actually South Africa is a little bit of a weird one. um because it's not a huge country, but it makes sense why you'd have like Brazil, Mexico, and then even US because they're massive countries. Brazil and Mexico are also cheap countries. So, it's very typical to see a lot of results there. Um but South Africa is a little odd. So, it's actually a really good example here like, well, what we want to do is we want to make sure we're getting what we expect from South Africa. So, we can follow that regular rule of I want to have at least a quarter of these conversions as saves, but specifically in South Africa. So, what I can do is I can flip over into the song stats here in Spotify artists and I can look at just South Africa and uh I did a video recently which I'll link to here where I go through like the the most important metrics you should understand how to look at in Spotify artist and this is one of them. Being able to filter by countries and look analyze stats is is very useful. Um so like the campaign had six about 600 saves or sorry 600 conversions. So, we're looking for at least 150 saves here. And we have it. If we look at last 12 months, you can see that whatever campaign the artist was running prior to ours, they must not have been targeting South Africa. So, we can see that pretty much all the South Africa conversions are coming from us. And we see more than 150 saves. So, we know the campaign's working and it's actually working above average, right? Because this means either more than half the people are saving it or we're having a better conversion to listener ratio. So, going back here, that passes the test. We know that we're getting results that translate to streams. And you could check the other countries, but just in the interest of time, I'm not. Now, we'll go down to um placements, and same deal here. We're looking for anything unexpected or anything odd. It'd be kind of weird if if all of your results are coming from Instagram feed desktop because most people use Instagram or mobile. or if all the results are coming to Facebook reels, that's fine as long as it's the numbers on Spotify look good. But if it's not looking good, then that would be a red flag. So, I I look through that as well. And the last thing we look for is creative related media. This one is one of like the gachas where some meta will like recommend other media in your ad account to use. So, sometimes it grabs footage of like the wrong song. So, um, and then it like I believe it even sends I don't know if it sends them to the wrong link, but a lot of times when I've when Meta has turned on related media without my knowledge, it it ends up being like none of it actually works. So, it just ends up burning money. So, if you come here and you turn on the breakdown related media and you're seeing all your money going to related media, that's a problem and it's probably causing issues with your campaign. And so, all of that data tells me that there's no red flags here. So, like I can trust this campaign that's 14 cents, but also I verified it myself on Spotify for artists. I'm not taking Meta's word for anything, right? One of my biggest rules of thumb ever is don't trust Meta cuz like well like they're always going to try to overount. They're always going to try to to make things look really really good. So, you don't want to by default have to trust them. You want to be put yourself in a position where you can independently verify how your campaigns are going without having to trust them. And if you do that, then you you can start you start to trust these ads a lot more because you're not you don't have to have blind faith that it's it's working, you know. So, moving on from this, now we can actually start looking at the campaign itself. So, I can show you how it's structured. Uh if I come into the campaign, uh I'll click edit here and you're going to see the campaign's actually quite simple. Um it is just a campaign with four adsets. We have two batches of countries. We have HP, which is effectively tier one. We call it HP because it's higher paying countries. I have like a a separate list of tier one, which then I have another list that's basically tier one, but also includes some other higher paying countries that you wouldn't normally consider tier one. So, don't worry about the nuance of that. It's effectively tier one. So, these are more expensive markets. These include the more expensive markets, but also the cheaper markets. And then what we have here is we've tried um two different batches of visuals. So we have four V1 visuals, four V2 visuals, and then we're trying EDM and house, and then EDM and house. Now, in in this case, the person on my team structured in a way where the EDM audience just gets the V1 visuals, and the house audience just gets the V2 visuals. Um, that's not always the case. And it's not always the case that we only do eight videos. Sometimes we do more. Um, it's just that this one worked immediately, so it's when it's working like but it's getting 14 cents. My rule of thumb is don't touch it, right? Like could we theoretically squeeze from 14 to 12 cents? Yes, but we're probably not going to. So, when something that's this good, it's like don't touch it, especially when the budget's fairly small. $500 in the grand scheme of things. Um, if you're hiring an agency to do an ad is is is the smallest offering that we we have. Um, on that note, if if you if you um you know, this is just the ad spend, it costs more, but if you have kind of a thousand and up in terms of a budget, um you can hire my agency to run your ads for you. I have a link in the in the description for that. If you're running ads yourself, 500 is actually quite a common budget. 500 to a,000, I'd say, is the most common budget range for someone running ads themselves on a song. And if someone's outsourcing it, then it's going to be like, you know, one to 1,000 plus, but like two 2 to 3,000 is probably more the average because it's more worth it to outsource it then because the money involved is so much higher. But now that you know the structure, if I go into the campaign, we can start to look at what was actually spent. So, um, immediately the the EDM audience did better than the house audience did. And this is actually another kind of lesson learned that I'll bring up is the house one technically was cheaper, but meta didn't like do anything with it. And sometimes you can force it, but a lot of times when you force it, it just ends up doing worse um because you forced it. And so at at one point, and I think if I go in here, you can see that um what happened was on August 25th, that's when we added we did the tier one um or the HP audiences. And then the next day we added the tier two. And we we did that to give the tier one a head start. Um usually we would actually wait a little bit longer before adding the tier 2 because we want to see if we can get a good enough result with tier one. Um in hindsight, this campaign, if I was personally running this one, I might have stuck with tier one exclusively for longer. Um and I might have even forced a bigger chunk of the budget to go to tier one to balance the countries out. But it's a hard call because because we went to tier 2 early, it allowed way more of the budget to go to this tier 2 EDM, which brought the budget the cost down from 28 cents a conversion to 13, which is a huge difference because the person who ran this made that decision early, even though I think I would have waited, um, it actually made it so that the campaign got way more conversions overall and is most definitely in a much better shape because of it. It's just a lot more the budget went to Brazil and Mexico. So the these are kind of like the more judgment calls where there's not necessarily a right or wrong answer. Um and a lot of it's happening on the fly. And also of course there's conversations with the artist and what their goals are and and that changes factors as well. So it again the way I might run things for my song is not necessarily the way that every campaign needs to be to be run. But you can see that the difference here like people ask like well why do you why do you want to have uh cheaper markets here? Why don't you just do all tier one? It's like, well, cuz tier one, this is the cost, right? Getting half as much results for the same amount of money, it's not always worth that tradeoff unless you have some way to capitalize on it. Now when we looked at the breakdown of countries when I was back here if we go back to breakdowns the over time of running um the campaign actually did quite good in the US 15 cents um and even in the UK it was actually quite expensive in Sweden was pretty good um Portugal pretty good New Zealand pretty good a lot of the tier ones pretty good some of them really bad but a lot of them pretty good If if this were my like my band's campaign, I what I would probably do and I might actually recommend this to this artist u because this is a good campaign. It's probably worth extending if they have the budget to justify it is I if if they're US-based actually because that does change it. then I might recommend um that we if they were to extend it for like another 500 we do half of the budget forced into the US because then that's going to help future you know having more people in your home country or your touring region is going to help when you're trying to play shows going to it's going to help if you're trying to do merch all the things right because you it's cheaper to ship in your home market you're going to be touring in your home market if they're if they don't care about that stuff then then they might just say nah let's just let's just go for more conversions for cheaper which I would actually agree with in that Um, but 15 cents in the US actually very good. And another thing is the US pays about two to three times higher per stream than Brazil and Mexico. So that's another factor. If the US is only 10% more expensive per conversion, but it pays two to three times more per stream. Economically, it actually even makes sense to to switch to it. Um, but it might not make sense from a raw like volume if your goal is just bulk amount of streams, if you know what I mean. All right. So now that we've filtered by source of streams, this tells a much more compelling story. Now, so now that we've filtered by source of streams, this gives us a much clearer picture. The blue is the active and the green is other listeners playlist. So it's not the entirely the case that this this campaign that we're running is the full picture, which is why some of these filtering tools are so useful. When we look at saves, that gives us a pretty clear indication because thirdparty playlisting doesn't give you any saves. It's like less than 5% almost all the time. And um so there's probably only saves coming from active. But this does mean that a lot big amount of this bulk increase is just because they they coordinated a playlisting campaign in conjunction with the active coming from the ads. And we don't see any algorithmic happening [snorts] yet either, which is actually very strange. So, this song I don't know if it is a human release or or an AI release actually, which is actually a good question. All right. And I just checked the song is actually a Suno track and I didn't I didn't check this out before. I just saw an awesome campaign. So, I'm I'm keeping this video um because this is this is kind of taken a lot of turns that I think kind of are hopefully helpful for you in terms of analyzing ad campaign performance. But the reason why I thought of that is is um I did a video recently explaining how um AI music has been throttled in the algorithm since like March. And the fact that this song has this many active streams, this good save rates, this good analytics, and has zero pretty much zero algorithmic traction tells was a hint to me that this might be AI because Spotify doesn't push out AI music in their algorithm anymore. It's it's basically just whatever you can drive with active or third party playlisting. So that changed the story entirely, right? the artist would not want to do anything for touring because they're not a touring artist. They're an AI artist or artist, you know. Um, and it's just it's a completely different ballgame, right? I don't know their motivations for for running this campaign, but it's it's it changes the dynamic. Certainly, it also shows you that uh it's getting 14 cents a conversion and it's an ad generated song, which is uh wild, honestly. Now, the next thing I want to show you in this roller coaster of a campaign is the the visuals. And I don't want to show them because they give away the song and then everyone's going to look it up and then it's it's a whole mess. So, I want to keep things anonymous, but I'm using Flow Stage in as visuals. And what that means is um I'm using the built-in EDM flow stage visuals. If you've never heard of Flow Stage, I do have a a video right here where I walk through the entire platform, but basically it's using stock footage of different like EDM looking shows and DJ sets and stuff um as part of their video library and it's just doing really well. And there's some hooks on it like I think one just says like here's a new banger for you on the ad and it's it's very simple and then it has different parts of the song with different hooks in each video. So very simple ads. almost all of the budget went into this one singular ad here, this V2 or V1 part two. Um, part of it might be that because these visuals were Yeah, these visuals were tested here in the tier one and Meta had concluded that this V1 part two is the overall winner. Um, and again, ignore small sample sizes like this, but this V1 part two is the overall winner. And so when we reused the same visuals here, Meta was just like, "Yeah, whatever. we're just going to settle on V1 part two. And you can see that there was a little bit of budget spent on these. So, it wasn't like they were turned off. Um, it just meta ignored them entirely. And then they throw this delivery error because if you didn't get a conversion on an ad in 14 days, they just turn off the ad automatically, which is a somewhat new thing they've been doing. Now, if you want to see how you can run ads like this for yourself from start to finish, check out this video right here to see the entire process. And make sure you subscribe so you don't miss my future uploads. And if you're looking for someone to hire to run these ads for you, I have a link down in the description for my agency, Southworth Media. Anyways, thanks for watching. I'll see you next video.