My facebook feed is full up of ads for courses that teach you how to use facebook advertising to get more listens on Spotify. Facebook ads are apparently so powerful that everyone is terrified they’re going to overthrow democracy, so maybe there’s a chance they might finally find an audience for our band.

I decided to do John Gold’s Spotify Growth Engine course. Wasn’t expensive, was recommended by a few people, sure what have I got to lose? Breezed through it pretty quick, and set up an ad.

(This wasn’t my first facebook ad campaign, btw – I’d run over a hundred facebook ads at different stages over the last couple of years. Some of the them got pretty decent clickthrough rates, or, in the case of the stuff I’ve done with my kids, led to some youtube views, but none ever made the slightest difference on Spotify)

The basic way any of these things work is you make an ad (video or an image with your music in the background), and then when people click “Listen now” they get brought through to a gateway page, and then they have to click on another link on the gateway page to be brought through to Spotify. The point of the gateway is to stop bots clicking on your link and then not listening to your tune – costing you money with no benefit to you. You can get Facebook to optimise the ad for clicks on the gateway page, choose your audience and away you go. You might run a few different ads at the same time (say with a different image, or different music clips playing in the background) and see which performs best.

I should point out here that I never click on ads for music on facebook, but I guess I’m hardly your typical music consumer so ¯\_(ツ)_/¯

So here’s the first ad I created. Aimed at fans of They Might Be Giants, Cake (the band not the stuff) and Weezer

Took me a whole day to put together 7 seconds of claymation using a bunch of rocks I dragged from the garden into the kitchen as the set, and I think it came out pretty damn well \o/

Set my budget to a fiver a day and BOOM people started clicking on it like crazy! I was super excited about all the clickthroughs, but … no action on Spotify.

Waited a few days – still nothing.

Lots of people were clicking ‘Like’ on the ad, which I thought was encouraging. You can see who likes your posts so I clicked through on a few … and every single one of these people’s feeds was jam-packed with religious stuff :/ Mostly languages I don’t understand, but lots of Christian and Islamic imagery. One guy had a picture of some bishop-y looking fellas as his profile pic, I’m guessing one of the Orthodox churches. WTF?

Took me another day to figure out what was going on … and then I realised that the first word of the lyrics is “God”. Could people really be clicking on the link because they the word “God”. I snipped that word out, and sure enough the clickthroughs dropped catastrophically

… to the extent that it was costing me about EUR1.25 for every person who clicked through on the gateway page, and half of them didn’t seem to listen to the music anyway, so I was spending EUR2.50 to get a single listen on Spotify that pays about $0.004.

Not exactly sustainable. Tried a few different ad variations and a few different audiences, talked to some other people, got the cost down to around EUR1.25 per clickthrough, eventually got sick of throwing my money away, and gave up.

Did nothing for a few weeks. And then …

I had the idea that maybe it’s the genre that’s the problem, so I did an ad aimed at my solo piano stuff (yes, that’s my actual bedroom)

… and straight away my cost per clickthrough was about EUR0.12. A 90% reduction on what it was for the Stoat stuff. Yikes. I can’t see that there’s any much difference in quality between them, so maybe it is the genre, right?

So I let that run for a month or so, and then I took a notion to try the band stuff again, with this ad

I aimed it at people aged 18-35 who said they were interested in “indie”. Yeah, I know I’m a lot older than 35, but I thought seeing as my 13-year-old designed the cover maybe it’d appeal to younger people. Set the budget, started running the ad. Cost per clickthough-on-the-gateway page – EUR1.38. Bah. Worse than before.

… so I changed the age of the audience to 18-30, and the cost-per-clickthrough instantly dropped to EUR0.24. Yep, more than an 80% drop compared to 18-35. And then I took a notion and changed the age to 35-65, and it dropped by half again to EUR0.12

Ok! Result! So it’s not that nobody likes our music, it’s bloody young people (shakes fist), and particularly 30-35 year olds. Weird.

It’s not the happy end of the story though – the Spotify listens seem to have collapsed since I switched from 18-30 to 35-65, and even the solo piano stuff costs me like 5 times as much as I’m being paid per stream.

None of this makes any economic sense

… but sure if I hadn’t any sense I wouldn’t be doing this in the first place, I suppose, so maybe I’ll keep fiddling about for another while until the lads notice that I’m spending all the money in our kitty on Facebook ads …

— Cormac