Marketing Monday Written by Jon for March 23rd

---

I owe you some good ol' fashioned marketing stuff.

My assumption is that most of you have found yourselves on this email list via some combination of bad luck and an interest in music marketing. I hate to disappoint people, and having spent the past couple of weeks philosophizing about the morality of Spotify and rambling on about the great questions of life in our tech-drenched age, I figure I should probably get back to my lane.

So let's talk about Meta ads.

While you could be forgiven for thinking that Meta ads are only tangentially related to my main job of complaining about life and technology, they are actually the main thing I do. Practically, this involves setting up campaigns to run on Facebook and Instagram – most often to drive music streams, although also sometimes to sell tickets or merch items. Here's a quick walk-through on how this works if you're unfamiliar.

I'm actually midway through reviewing every campaign our agency has run in the past year or so. It's been... interesting.

At some point soon I'll probably share the full dataset, but for now, I'll give you some of the early highlights. And, more specifically, I want to give you some clear benchmarks on what makes for a good streaming campaign.

If you're running ads for your music, here are a few numbers you should know.

👋 Quick ad break...​ ​Want to learn how to find your fans? Join our Music Marketing Club! Here's what we've got coming up next month for members: Music Ads 101: How to run ads that get your music heard Plus there's a full course library, recording of past Q&As, one-on-one support, and connection with other awesome artists. ​Check out the Club here.​

Before we begin, keep in mind that the following numbers are:

• Only applicable to Meta campaigns for music streams – so, not TikTok or Google ads, and not ads for tickets, etc.
• Almost entirely focused on Spotify results (rather than other DSPs).
• Taken from a little bit less than half of last year's agency dataset, so they'll likely change some by the time the whole review is done.
• Not actual reality, just arrogant little symbols that gesture vaguely in its general direction.

Okay! With those few caveats, here's what we're looking at so far:

Average Cost Per Result: $0.57

This is the average cost we've paid in the past 12 months to get someone over to a streaming service (nearly always Spotify). That's more expensive than it was the previous year; in 2024, our average CPR was $0.52. Looks like Meta ads really are getting a little more expensive.

It's also worth noting that the CPR varies widely based on the target of a campaign.

• Single-country campaigns are nearly always above $0.70
• Tier 1 campaigns are usually $0.40 – $0.70
• Tier 2 campaigns are usually $0.20 – $0.40

A general rule of thumb is that narrowing your target increases your cost to reach it. For example: If you'd like to target only Instagram, or only women, or restrict your age range to only people between the ages of 20-30, you'll almost certainly pay more.

But yeah, to recapitulate the headline: If you're under $0.57, in general you're getting a good deal on traffic, and if you're above that mark, things are more expensive than average.

Average Lowest One-Day Cost Per Result: $0.26

Over the course of a full campaign, of course, your cost per result per day will vary pretty dramatically; some days impressions will be more expensive, some days less. But our average low point, across the life of a campaign, is $0.26 per result.

Interestingly, this is actually a cent lower than it was in 2024 (when it was $0.27).

But even more interestingly...

Average Time to Lowest One-Day CPR: 48 Days

In other words, it takes nearly seven weeks for an average campaign to reach its most cost-efficient point. This number is up 16 days from the year prior.

And this is one major reason why I don't like front-loading an ads budget (i.e., spending half your money in the first couple days of the campaign).

Sure, there's a valid argument to be made that sending more early traffic to streaming services helps in terms of the streaming algorithms – but on the Meta side of things, if you spend all your money in the first few days, you're pretty clearly spending inefficiently, which is a kinder way of saying "you're wasting your money."

Now, to be fair, my dataset does include a couple of campaigns that are significantly inflating this number. One campaign, for example, reached its lowest-ever CPR 359 days after its start date – not because it really took that long to get optimized, but because we kept turning it off and on.

But still: The data unequivocally shows that most Meta campaigns take at least a couple of weeks to get optimized, and many take longer. So if you're only running ads for a week, you're likely missing their best period of performance.

Average Saves / Conversions: 68%

I've saved my favorite number for last.

We've found that saves, much more than streams, playlist adds, or any other metric, have the highest correlation to repeat listens. And that's why this, in my opinion, is the most important metric in evaluating the success of a streaming ads campaign: The percentage of people who, upon clicking from an ad over to Spotify, end up saving the song.

You could have an incredible CPR on a campaign – say $0.10 – but if none of the people you're sending to Spotify actually like the music, then it's still wasted money, like pouring a ton of water into a bucket with no bottom. Conversely, you could have a higher CPR – say $0.70 – but if every click leads to a save, your numbers (and your fanbase) will stack exponentially over time.

A high saves / conversion rate is a great sign that a campaign is working. The number will vary based on other promotion you do, but honestly, anything around 68% is very solid.

Let's pause there for now.

Like I said at the top of this newsletter, I'm planning to share the full dataset with you at some point soon (and I'm planning to make a YouTube video about it, too). But for now, I hope these benchmarks help as you evaluate your own Meta ad campaigns.

Wishing you, as always, good luck.

– Jon

Jon Anderson Founder @ Two Story Media Surprisingly Bad @ Scrabble

🫆 100% Human Guarantee I wrote this, not AI. Today's random photo: Last week we had our first thunderstorm of the season... and found out that our dog is absolutely terrified of thunder. This was the aftermath of her attempt to claw through our front door. If you have advice, I'm all ears.

P.S. – News thing! Starting two weeks from today (the week of April 6th), I'll be running a live class called Music Ads 101. I ran this last fall, and it was great.

More info to come next week, but if you're interested in learning how to run ads yourself, this is for you. You can join by joining our Music Marketing Club; club members get access to all of my live trainings.

P.P.S. – I got a lot of responses to last week's email about my personal approach to technology, and literally all of your responses were kind and thoughtful. Thanks so much for reading and engaging with that. Also, if you're wrestling with tech and work and life, know that you're not alone.

(On a semi-related note: Three different people wrote in to tell me that I have "an easy way with words." I'm flattered by the sentiment, but I also find it funny; it took me more than five hours to write last week's 2,208-word email, which comes to like seven words a minute. If that's easy, I'd hate to know what hard looks like.)