Case Study - Promoting music with PPC (pay-per click) ad platforms
My day job is as the CMO for a digital performance marketing network where i run the internal department. I’ve also spent about 4 years before this running my own business doing essentially the same work, the only difference is the scale of the operation. So as far as pay per click and social media marketing is concerned, I’m very adept. I specialize in Facebook ads but have run on numerous other ad platforms. After seeing the abundant misguided advice being offered up to marketing-curious musicians, in terms of the effectiveness of paid ads, I decided I’d perform a small test and share my findings. I’ve made a considerable time investment in order to perform a study on the music promotion traffic generated primarily through Adwords, by means of promoting Youtube videos on Youtube.
I dont have especially high quality audio assets of my own to promote, being a home studio practitioner, but i took a couple tracks, threw together a video with a still image background and uploaded them Youtube. I think this generally may be the best option for a musician. I used an approach I have used on Bing search ads before, which involves scraping bulk keywords, cleaning the list, running the search metrics on them with Google Bing and Youtube, only keeping the ones with decent search volume, finding more scraping based off those, and repeat yadda yadda. I’ve also utilized alexa.com heavily as a keywords source in order to build my various lists.
Keywords and targeting are virtually the same thing. With a search ad network, keyword bidding is entirely your targeting, the other targeting settings are cosmetic at best. That’s what separates most ad platforms from Facebook, Facebook is the most accurate and extensive in terms of targeting options. Not to say it superior to search. Although ultimately it probably is. But remember with keyword lists its all about cleaning the list too, in most cases. I mean you can bid on keywords like “Facebook login” “gmail login” and get clicks for pennies, but you better be ready with a tricky/clever funnel to back out the tiny fraction of it for which theirs always potential with anything. Otherwise a major tip I have for targeting with any platform but significantly more so with search ads, never NEVER include audience networks or partner networks, content partners etc use only the search engines direct placements.
The metric that’s most significant, reiterating the point the fellow in the comments made, the average duration of the video view, but any social events with YouTube or the respective Twitter and Facebook artist pages are also considerable for gauging traffic quality. As far as fraudulent traffic on Facebook, I check the click logs at work everyday with a range of reports I can assure there is no concern for fraudulent external website clicks based campaigns, but keep in mind, we run absolutely no audience network placement largely for reasons akin to that, so it’s very possible it occurs. This is why I suggest always excluding audience networks, content partners, etc from the placement for an ad platform.
The keywords dataset I built for this study is available at Here. You can view/filter/download it, in its entirety for free, no need to sign up or opt in to anything to do so, I just wanted to make it available in a user friendly format. Let me know if you’d like me to provide anything else.