What Artists Should Know About Southworth Media
Southworth Media as in Andrew Southworth? Yes.
If you don’t already subscribe to Andrew’s YouTube channel, you definitely should. It’s basically what I do here but in video format, with new content every week instead of every… whenever I find the time.
When Andrew offered to let me kick the tires on his new service, I of course said yes. He gifted me a Spotify Growth Ads campaign with an ad spend of $1000 for the purpose of this review, to promote my new-at-the-time single “Where Tigers Are Said to Roam.”
This isn’t a playlist pitching service, so we don’t have to worry about fake streams and bots and all that nonsense. There’s zero risk of streaming fraud.
Instead, Southworth Media runs Facebook and Instagram ads on your behalf, on their own ad accounts with their own pixels and smart links and landing pages.
Best of all, they create compelling ads specifically for your track, from scratch. I have to imagine that’s most of what we’re paying for.
To get started, select your budget and provide some basic info:
When your campaign gets approved, they’ll charge your credit card and you’ll get an email asking for read-only Spotify for Artists access, to help them optimize the campaign and provide accurate reporting.
You’re kept in the loop every step of the way. First, a “preparing creatives” email (above), then “your campaign is launching shortly,” and finally, “your campaign is live!”
You can track the minute-by-minute progress of your campaign in two places: your very own custom dashboard, and on Feature.fm.
Both of those links are live as of this writing, so feel free to explore my results for yourself! To see the full campaign, set the date range to January 9 through February 6, 2025.
Things got off to a slow start, with the cost per conversion (a click from the landing page to a streaming service) dipping below $0.40 before rising.
Out of the dozen(!) ads they created, this synthwave-themed one performed the best:
But when I saw the cost per conversion rise to $0.58, I knew I could beat that with the video I’d been using to promote my Vocal Synthwave Retrowave playlist:
I was right! I sent them the video and it sucked up most of the ad spend, getting us all the way down to $0.25 per conversion:
At nearly the halfway point, they could’ve easily just rode it out and left well enough alone. Instead, they made more videos!
Ultimately they were able to beat my video with this one, showcasing the similar artists I listed on the signup form, along with one I’m not sure I recognize. Is that George Micheal? Bono?
Andrew graciously sent me the video without text and I’ve since repurposed it for my playlist ads, with fantastic results!
When all was said and done, we ended up with 4110 conversions at an average cost of $0.24.
The cost per conversion got as low as $0.13 towards the end when they pivoted towards tier 2 countries, mainly Brazil and Mexico. Overall, 80% of the budget was devoted to tier 1 countries.
Here’s a more visually compelling take from Feature.fm, in case you don’t know your two-letter abbreviations:
Next up in my dashboard, we have a breakdown of all the ad sets. Their names tell you pretty much all you need to know.
For example:
HP – IG – Synthwave Group (Brian Visual) – $5 Min
HP = high-paying countries i.e. the ones that pay high royalty rates, mostly tier 1
IG = Instagram
Synthwave Group = targeting synthwave-ish artists/bands
Brian Visual = the video I provided
$5 Min = $5 minimum ad spend to ensure that Meta actually runs the ad rather than relying on the previous winner
I bet you want to see that John Wick ad?
Next up is a gallery of all the ads:
When you click on one of them, it takes you to Meta’s ad sharing page, where you can right-click to download the video. Which is how I’m able to showcase them here!
Let’s see… dudes 45-54. Yep, that’s my audience! Placements were predominantly Instagram Reels and Instagram Feed.
If you’d rather leave the high-level analysis to them, you’re in luck. In addition to the data reported above, they also prepare a Week 2 Report and a Final Report, coupled with a detailed glossary.
Both reports include holistic explanations of data from both Meta and Spotify for Artists.
If you want to read the whole thing, it’s easiest to just go to my dashboard.
Andrew always says that most streams occur after the campaign ends, and that will certainly be the case here:
At almost two months post-release, I’m at 42K streams, but of course I can’t attribute those streams entirely to this campaign.
And that’s just Spotify. A non-negligible percentage of conversions were to Apple Music:
Which generated some indeterminate portion of these results:
All in all, I’m very pleased with the results of my campaign and have already recommended the service to friends.
Could I have beat the average $0.24 cost per conversion myself? Probably not, because I would’ve stuck with the “good enough” performer I’d already created.
Until recently, I was convinced that my little cover art ads were “good enough.” After all, people were hearing the song, and that’s what’s important, right?
Now it’s beyond obvious that the quality of your creatives matters, a lot. I’ve gotten pretty good at making video ads in Canva these past few months, but I simply don’t have the editing chops to put together the concert footage video that ended up being the top performer.
If you’re in the same boat and running Facebook and Instagram ads anyway, it’s possible you could save money by enlisting a service like Southworth Media. Especially since you’re given the option to continue the campaign, presumably indefinitely.
If you’d like to try a Spotify Growth Ads campaign, or any campaign Southworth Media offers, you can get 5% off using discount code PassivePromotion here.
Be sure to share your thoughts on results in the comments!
great summary. 2 questions
1) So the conversion event here, was a Spotify stream? ie the Meta user seeing the ad, clicking on the FFM link, making it Spotify (and a bit of apple music) and listening to at least 30 seconds?
2) Do you not think its worthwhile to keep ads under 15 seconds so they can run on Stories?
Great questions Shah!
The conversion event is a click from the landing page to a streaming service. Spotify and Apple Music are the only two services on the landing page.
The ads are edited such that everything of substance is done by the 15-second mark, and the rest is just the music playing out. They’re also formatted so that they work when cropped to square!
Hi Brian!
It’s been a minute and glad to see you are still doing well with the reporting.
The only thing I am curious about is the use of various concert video and the john wick clips when promoting the song. So Meta never sent a copyright strike to Andrew or you for using that?
I was referencing Meta’s ad policies and I’ve seen accounts get shut down immediately for that very same thing. Could you shed light on that?
The only platform I’ve ever gotten a copyright strike on is YouTube, for my Synthwave Top 10 podcast. It didn’t matter that I had permission!
I’ve never heard of a Meta account being shut down that reason, but Andrew would know better. I’ll poke him and see!
Any fresh news on this mate?
I wonder how Andrew gets away with this?
For my part, heck if it works then it works!!!!
I’m about to try them. I’m waiting for my next song to be finished and released. I want to test them with new material.
Fortunately I never had an issue, but since then they’ve made it clear that they don’t use copyrighted content. I’ve gotten plenty of updates from friends and readers that have run campaigns, and none of the ads I’ve seen are Wick-level anymore.
Ok so the conversion is just the user making it to Spotify/Apple, not necessarily a stream? I misread their email to you saying “Your campaign generated 4757 conversions. These conversions have contributed to thousands of streams”. The campaign could have generated 2 conversions and that second sentence would still hold true.
So youre saying if the ad is longer than 15 seconds it can still appear on stories, but will cut off at the 15 second mark? I thought that if the content was longer than 15 seconds, it would not be shown on stories. Either way, youve made me realize that if i have a video thats more compelling when longer than 15 seconds, the better strategy is to run the full length video and everywhere other than stories, and then edit down to a 15 seconds version that will only play on stories
I’d say the odds of two conversions generating thousands of streams falls into the “monkeys might fly out of my ass” realm of probability. As there’s no way to track the user past the click to Spotify or Apple Music, that’s the best we can do.
That said, there’s no easy way to translate conversions into streams. The first 30 seconds of the track could be terrible enough (or just so completely different from the ad) that nobody makes it past that point, which doesn’t qualify as a stream. But at least to these safe countries, conversions are a solid indicator that the campaign is getting people to your track.
You can see from the screenshot that there were 275 conversions on Stories, so there you go. I used to create separate 15-second versions of my ads as you describe, but now I just go full Southworth and let the axe fall where it may. 😉
Get this… I just saw my ad on Instagram Stories and it played the whole 21 seconds. Now I’m confused, but not complaining!
cheat code unlocked.
speaking of cheating wanted to ask you something about fb ads. was going to email you this directly but maybe others can benefit
I reupped with feature fm for smart links, and then was looking at the user data obtained from when I was previously described. The vast majority of the users in the database had totally bogus looking emails (eg [email protected]) , ie these were fake accounts/bots.
What i dont understand is how to ads end up getting consumed by bots? . Presumably, it costs effort or money to operate a bot, so I dont understand why someone would waste that money/effort on a bot consuming a musician’s Meta ad. The only explanation I can think of is that Meta operates these bots to exploit advertisers. Meta has former Mossad operatives in its executive ranks, so I expect them to have zero ethics, but if fb was actually operating these bots i feel like it wouldve been exposed by now.
And if I understand correctly, while bots consuming ads is an accepted part of the Meta ads experience, the 2nd step in the feature.fm process (ie where the user clicks ‘save on spotify’ or whatever prompt button theyre shown after initially clicking the ad) should prevent bots, right? since completing this step requires: clicking the prompt button; having a Spotify account; and being logged into that Spotify account.
So Im totally perplexed as to how that previous Feature.Fm campaign data consists of 1000+ fake profiles (and maybe 5 possibly real ones).
Please make it make sense
I’m sure others would benefit from this discussion, and that many can answer your question better than I can!
I’m looking at my Feature.fm audience now (395 fans) and the emails all look legit. In fact, I recognize many of them.
You might investigate whether any particular segments of your audience are to blame, or if the whole lot is useless. In which case, I’d launch a new pixel and start over, because you’ve trained Meta that your customers are bots. I’d start with just the US and expand to other first world countries once you know things are working properly.
I’ve had it happen before. I was getting $0.04 conversions from Ukraine, before the Russian invasion.
I really don’t understand the motivation beyond disrupting American businesses. I doubt the major labels are out there paying foreign agents to deploy bot farms to take down indies! And I highly doubt Meta is responsible, because it threatens confidence in the ads product.
Oh, one other point: the bot doesn’t need to have a Spotify account or be logged in. It just has to click a button on the landing page. That’s very, very easy to program.
Yeah Ive looked thru the email list, mostly all Russian domains. So thats what this is? A Russian interference campaign? I was a national surrogate for Bernie Sanders’ 2020 presidential election campaign, but it seems like a stretch that those are connected.
Im not sure how any particular segments of my audience could be to blame, but “youve trained Meta that your customers are bots” seems like this is a much bigger problem then i realized. Just to clarify this was from 2020. Ive been running meta ads again since the end of 2023. I resubscribed to feature.fm about a month ago, and i think the results have been dismal. 10000 ‘total visits’ resulted in only 239 ‘clicks to service’ and 3 email addresses. In the presave campaign they have to manually input their email address.
I assumed the 2020 results were from an Action campaign which prompts the user user to ‘follow’ which results in their: 1) following me on Spotify, 2) following my playlist if thats what I lead them to, 3) email address. The shortcoming of this approach is they have to be signed into Spotify when clicking the link. This is the FFM campaign Im going to use now. and i assumed this was the FFM campaign that led to all those fake russian emails.
How would I know if there’s a lingering effect of that 2020 russian interference campaign? I guess I could only direct users to apple music instead of spotify, since apple requires paid users. but that would leave me w no spotify strategy except for playlisting
Man this sucks
Yeah, it does suck. I doubt that your personal political preferences were to blame. I’m not seeing where you mentioned your country targeting, but that’s most likely it.
I don’t think I’ve tried an Action campaign like that but it must be expensive! I’d guess it would be close to $1 for a first world user. An email and follow is a big ask, especially when all those permissions are required.
I’d start by using a new pixel like I described earlier, with a simple landing page directing to Spotify and Apple Music, US only. Unfortunately your ad account itself might be more prone to go in the wrong direction now. You could open another one or just start slow and make sure the clicks are generating streams every step of the way.
I don’t see how directing users to Apple Music instead of Spotify would help. It’s the users Meta is finding for you, not where they’re clicking through to.
Hope that helps!
goddamn this sucks. im not sure what countries were targetted in that campaign, but the fake russian emails were all had America listed as their country (i think when they registered for the spotify account). So I was motivated to do that Action campaign after reading this https://www.reddit.com/r/musicmarketing/comments/1hn11up/another_meta_whatever_strategy_for_a_playlist/
I think he got it to under $1 but tier3. he also was using a very small budget, tho he also said a small spend $10/day i think, was the optimal amount
So my goal is to replicate what he did. ie real human fans.
is there any way to tell if my account is sullied like youre suggesting? based on the ratio of 1000 fake profiles to 5 real ones, it certainly seems like facebook was deliberately presenting the ad to bots. im hoping that wouldve been washed away over these last 18 months of ads, but the lack of results suggests you might be right
facebook providing me an audience comprised of nothing but bots seems like grounds for a lawsuit. you think theres a lawyer whod take that case on a contingency basis?
It looks like the campaign described in that Reddit post is what I’ve been doing for years. You don’t need to ask for the follow and email from the landing page. Too much friction. Just direct to Spotify and Apple Music and the streams and follows will happen naturally.
I don’t think you have a legal case. Sure, it would be nice of Facebook could do better in the bot department. I mean, I get a half dozen phishing messages a day and have for years.
As for testing your account, I’d add one more thing to my previous instructions. Run your ads on Instagram only: Feeds, Explore, Stories, Reels. That’s all I usually do anyway.
That should eventually teach Meta to find you real listeners!
I may try this out. I’m pretty proficient with Facebook Ads, but it takes way too much time, time I would rather spend composing/writing/recording. But the fact that I know how to use Facebook Ads gives me the confidence to know that an agency I engage will also know what they are doing. And both you and Andrew knows what you are doing!
Thanks for what you do Brian.
Joe
My pleasure Joe! I totally hear you on the money versus time thing. I devote more time than I otherwise would for promotion so I’ll have something to write about!
Hi Brian,
The results look reasonable and the plan follows indepreneur’s best practises for ads of this kind. I like the fact they will make the creative for you and keep testing throughout.
I am surprised at the targeting though – why would countries outside “green light” countries be targeted. I see little to be gained in long term customer value for targeting south american countries, for example.
What are your thoughts here?
Hey Tim!
I’ve always been frustrated with Indepreneur’s outright ban on “red light” countries. In fact, every time I’d share a new post to the group, they would disable comments because a “red light” country would be mentioned, once even just in the comments! It really rubbed me the wrong way and I didn’t bother renewing.
That said, I typically only promote to “high paying” countries, but throwing in a little Brazil/Mexico is totally legit. You’re striking a balance between nailing down your ideal fan and generating cheap streams. Of course I have actual fans in Brazil and Mexico, but it’s safe to say that they rarely buy merch, which is Indepreneur’s primary focus. If it wasn’t clear already, they consider Brazil and Mexico “red light” countries.
There’s no actual tension here because you can just tell Southworth Media your preferences. If you want 100% of your ad spend devoted to the US, you can absolutely do that!
Hey Tim,
If you’re trying to trigger Spotify algorithmic playlists, the country selection doesn’t really matter. Meaning, as long as people are streaming it doesn’t matter where they’re coming from it will help you reach that goal. Often when you hit the algorithmic playlists you’ll get most of the traction in Tier 1 countries, even if you used Tier 2 countries to trigger it in the first place.
That plan changes entirely if your primary objective is touring, merch sales, email list growth etc. Personally I run my own Spotify ads to Tier 1/2 because I don’t care where people are from as long as they like the music and are listening to it with good engagement. But when I do email list growth or merch sale campaigns it’s most often USA only or at least Tier 1 only.
For my agency we start every campaign with what we call HP countries, which is basically Tier 1 (which is similar to Indepreneurs ‘green light’ with a few extras thrown in). Then if we can’t get good enough results with those we add in the full Tier 1 and Tier 2 list.
Sometime’s clients come to us and say ‘we want USA only’, or ‘we want Tier 1 only’, or even ‘we only want India, Pakistan and other south Asian markets’. It’s their money so we’re happy to do whatever the client wants. That being said, we will ask *why* the client wants to do it, because a lot of the time their reasoning is rooted in false assumptions so we want to be sure they understand all the facts before they tank their own campaign.
In other words, if you don’t have a good reason to target USA only you’re just doubling or tripling your cost per conversion for no good reason.
Hope that makes sense!
The Man from “My .04 cents” Podcast weights in? This clinches it. Props ta’ Andrew to the South, North and Woolworth! The Tier 2 target audience. More bang for your buck…and you’ve mentioned this target as a guest in another podcast…I believe it was the “One More Time” guest appearance with those dudes from Atlanta? Hi-Five from Iowa City, Iowa and to the boys @ .04
I recognize some of the stock footage being used for the creatives; however, was a bit taken back to see what appeared to be copyrighted footage (john wick, concert footage), being used in some of your ad creatives. Are they licensing this footage, what are the terms of the licenses, and do you risk getting your own ability to run meta ads because a third-party is running copyrighted video/images on your ad account?
I doubt there’s much risk. Meta can always reject the ads. Generally Southworth Media runs the ads on their own accounts, which was the case for my campaign. That said, I’m using the same ad now without any issues!
Meta’s advertising Terms of Use do not allow posting content that violates someone else’s intellectual property rights. Just because you haven’t been caught, doesn’t mean there’s no risk. Since you are now re-purposing their ad on your own account, I’d be weary to risk losing your ability to advertise via Meta.
Similarly, I still find it questionable that an entity is willing to take your money to do things that are specifically against Meta’s Terms of Use.
I get where you’re coming from, but I mean, we all do that sort of thing all the time. If I want to share a record I’m into, I’ll post the cover art. AMVs are 100% copyright violation and seem to be welcome on every platform. I use other bands’ logos in my free CD campaigns. A good percentage of memes (most?) use others’ intellectual property.
But sure, Meta could shut down my account for any number of reasons. I’ve never heard of anyone getting in trouble, but I suppose that if the copyright owner complained, it could happen.