How To Test UNRELEASED Music With Meta Ads

In this video, I want to go over a very special type of campaign that you can use to test out songs before you even release them to find out what are your winners, what are your losers, at least in terms of how they perform in meta ads. Specifically, I call this a multi-ong test campaign. And I've used it in the past where when an artist has like an whole album or whole EP ready and they're just going to be batching songs out like every 4 to 8 weeks, we're trying to find like what's the order we drop these songs, what songs do we save for the album, and what songs do we actually release as singles? And not just in the artistic way. Of course, you can do this from like the artistic perspective of like these are the songs that mean the most to me and I want to release them as singles. But when it comes to actually getting numbers and and treating, you know, your music as a business, it's good to have some type of concrete data and this is one way to do it. So, how do you actually build a multi- song test campaign? Uh, the first thing you need to have in mind is you obviously need your music ready ahead of time or at least you need enough of the song ready to be able to make ads out of it, right? It technically doesn't have to be done. Like you could just make a a precourse into a chorus for five songs and that might be enough, right? And then you just make ads with those parts of the songs. Or of course, more ideally, you'd probably just have all the songs ready. So that way once you see how things are doing, you can start dropping stuff. And of course, I mentioned kind of alluded before, you need the ad creatives for each of those songs. So let's say, for example, you have five songs you want to test and you want to see which ones are the winners, which ones are the losers. How would you do this in an ad campaign? So we're in meta ads here. Uh, if you have no idea how meta ads work, um, I'll link you to a video right here that walks through the whole process from start to finish. How to run meta ads to promote your music on streaming platforms. Specifically, Spotify is the biggest one everyone cares about. But the this type of campaign also works for Apple, works for YouTube music, works for any DSP essentially, just Spotify is the biggest one right now. So, we're coming in here. We're going to build a brand new campaign. And it's going to be an engagement campaign like we always do for these type of like Spotify growth or streaming growth campaigns. We want manual engagement. Now, very often I use this advantage campaign budget feature here. Uh in this case, we're not going to use this is one of the rare circumstances where I do not use advantage campaign budget because we are going to be controlling the budget from different adsets. We're going to have different songs. So, this is I'll just name this like multi- song test. Then, we're going to click next. We're going to do some basic kind of housekeeping. Our conversion location is going to be website. We're going to choose the conversion event that applies to the landing page we're using. In this my case, I'm using feature FM. If you're using submit hub links or toneden or smart noise, it'll be view content. If you're using hyped it, it'll be hyped at smart link click. And then we get to this budget section. Right? So, this this ad set that we're building right now is going to be for essentially a song. So, this is going to be I'll just call it song one. Basically, all the ads inside of this ad set are going to be for song one. We're going to target tier one two countries. We're going to run the ad on Instagram. And then we're going to target things that sound similar to the music. And we're going to do that all for, let's say, five bucks a day. I'm recording this video on October 3rd. I don't know when you're going to see it, but let's say we're going to run this test for a week, assuming I was publishing it today. And so therefore, this ad will spend like $35, which should be enough actually to get enough data to conclude like which song's better than the other in terms of performance. You could do a bigger test. You might even be able to do a smaller test as well, but I think actually you could definitely do a smaller test. I I think doing like at least $20 of spend. $5 a day for four days would be kind of like the bare minimum. Doing $5 a day for a week across x amount of songs. If you have three songs, this is like a $100 test. If you have five songs, this is like $175 test. If you're planning on spending thousands over the course of five songs, right? The doing this ahead of time might actually save you a lot of money. You can imagine. Now, I don't want to go through every painstaking detail of this campaign setup because it's in that video, but we are going to use original audience options. I'm going to go in locations and use my preset if I have it in here. Tier one, tier 2 list. I'm going to set the ages 18 to 50. And then we would do define. Oh, whoops. I didn't want that. We would do define further. And then we would put in something here. Right? So, if I'm doing like metal core music, I might do metal core, etc. Right? That's this is a big can of worms. There's a million things you could do here. There's different country things you could do. I don't want to get into that here. I want to get into the functional setup. The last thing that we have to configure on this adset level is our placements. I mentioned before, we're going to do Instagram only. So, for the most part, you could come down here and just turn everything off except for Instagram. Personally, I also turn off the search results and the profile reels. And then we have our placements here. So, this is our ad set. And this adset is going to spend $5 a day for a week. And so, spent $35 total. Now, if I go into the ads, remember this is just going to be song one. I would add my website URL. I would add my videos. Not going through that here. So, imagine I just did that. This would be like song one, verse one, and then I would come and clone this. And this is just a type of video. This might be a performance video of you singing the song. This might be a lyric video. This might be some stock footage that sets the vibe. There might be lyrics. There might not be lyrics. There's a million different things it could be. Um, this is another can of worms that I don't want to get into in this video because that's something that I I talk about extensively and it deserves it its own focus, right? This focus on this, how to do the multi-ong test. For the most part, the type of content you would do on social media that does well also will work well as an as an ad in a lot of cases, but it's something you have to experiment with. So, let's say we're going to do three videos per per one. So, we have verse one, verse one, precourse, and chorus. So, we have these three different videos, and this is just under our song adset. So, these three videos are going to get $5 a day for 7 days. And then what we're going to do is we are going to clone this ad set. And of course, this one's now going to be song two. We're going to keep the targeting the same. So, whatever targeting you choose, try to keep it uniform because if you do different targeting between them, you're you're going to get different results. You're kind of you're now confounding variables, right? If you change the targeting and the song, you don't know is it worse because of the target or worse because of the song. So what you want to do is pick targeting that kind of applies for everything and it might mean your targeting is broader than it would otherwise be. But then after you find a a winner or some winners, then maybe you go and do some split tests with audiences to to find how things are doing. But essentially this other ad set, so we have a song one ad set and then a song two ads set. If I zoom in song one, song two, all of these ads under the song two one, you would now come in, rename them, delete the video, re-upload the new videos for song two. Like you might have mentioned that I haven't mentioned anything about the link. Like you might be screaming at the monitor right now and furiously typing a comment, but Andrew, the song's not out. Where the heck are you sending people, right? How are you what are you putting in this website field if the song's not out? What like what is happening? Well, that's the kind of most interesting part about this, I would say, aside from the the general theory of testing songs before they're out. Um, two big options, right? No matter what you're doing, you're going to be sending people to a landing page, something like this, right? It might have less options on it. It might only have one option on it or two or three, or it might have all of them, right? But the song's not out, so you can't send them to the song. So, what you're going to do is you're either going to send them to your profile or to a playlist. So, if I click on Spotify here, it's going to take me to the song. I'm going to go to the profile. You could send them here. This is one option. And then what'll happen is people will come here and they'll start playing music and hopefully they'll like the songs you already have out so much they'll completely forget about the song they heard 15 seconds of on Instagram. Second option, send them to a playlist something like this is every waking moment. Same deal with you. They're going to play it. Hopefully they're going to love what they hear so much they're not going to even know that they were sent to a different song in the first place. And some of you might be thinking, well, what does that actually mean, right? Like, are people going to get mad? Are people going to be upset? Are people going to be yelling at me? And well, I think you're correct for for worrying about that. In reality, it's not really a concern. [laughter] So, uh, you know, if you're thinking that you're misleading people, you're right. You kind of are misleading people, but I have never seen one person complain from this. Remember, you're not doing this extensively. You're not targeting your fans with this either. You're running a a $35 test where you're driving people to your music, people who have never heard of you, you're driving them to a page where even if they do realize that they were misled, and I'll zoom in or go to full screen for this. Even if they do realize that they were sent a different place than what they heard, they're probably not going to think it was for a nefarious reason. They're going to think that you just messed up the ad or something. They ended up in the wrong place somewhere where they shouldn't have. Um because otherwise like how could I have just never seen anyone complain? I have done this for like probably over a dozen times for different artists in different genres. And never once have I seen a single fan DM the artist come comment in the ad and complain etc. So the point of this strategy right is we're testing songs before they come out. We're not spending a ton of money rand if you're planning on dropping like five grand on like an EP or something in ad spend. We're spending like $175 here. So it's a very small amount. You're running this for a week and you're not targeting your fans, you're just targeting strangers. So, it's a very kind of short onoff test and then you're going to get real life data on how things perform, what ads work, what ads don't. And so, even from the perspective of learning how your ads are going to do potentially weeks or months before your song comes out, let's say you have a song that you know you're releasing as a single and the ads suck performance-wise, like they're getting a dollar a conversion. Now, you can go make new ads. So even if you it doesn't change your release play and your release strategy, it helps you change it ahead of time. Now for those of you who might write partials of songs, like there's a whole kind of stack of people who will write chunks of songs and throw them online and see what happens or on TikTok for example, and then they see if a song pops off and they go finish it and release it. This is another way to do that, too. And I know people who have done that, done partial songs. You know, people who have ced a whole album and then used it organize their kind of release cadence and other people who just use it to like test their ad creatives ahead of time to see if they need to go back to the drawing board to make sure that song doesn't flop, at least for the ad campaign. So, if you want to see more about how you can run campaigns like this from start to finish, the bulk of that strategy is in this video right here. I almost pointed the wrong way. [laughter] So, check that out. Uh, let me know in the comments what you think about this strategy if you tried it. And also make sure you subscribe so you don't miss my future videos. Thanks for watching. I'll see you in the next one. Bye.