BY MARK FRIESER
“Music is the pleasure the human mind experiences from counting without being aware that it is counting.”
As the guy who conceived, built and managed the US Ringtone Charts with Billboard in 2004 (along with a team of amazing tech and BD people at Billboard and at my own company, Consect – the systems we built became the foundation of Nielsen Mobile and its RingScan service), new measurement standards and charts are incredibly important and interesting to me, especially when they impact our business, like the Billboard/Clio Top TV Commercials (Music) Chart and the THR Top TV Songs Chart.
Both of these charts are wonderful, and their existence is a testament to the importance of music in media and ads as a primary way of building audience and sales for music and musicians and enhancing the stories and messages of visual and interactive media. I salute both efforts.
With that said, looking at these charts from the dual perspective of someone working in music licensing/composition for media/ads and someone who’s developed industry measurement standards, while I think they’re a great start, I think there’s room for further improvement to better give people in the music and media business actionable, reliable data.
In this post, part one of two, I’ll take a look at the Top TV Commercials Chart, and what I believe would add to the chart’s statistical probity and utility.
First, let’s take a look at the two main issues I have with the current measurement methodology and sources.
1. Reliance on Shazam tags.
I think Shazam is great, but using their tags as the primary metric for measuring engagement through discovery queries is overly simplistic and doesn’t provide enough of a sample to be reliable. Though it’s a good start, at the end of the day, all you’re getting is a sampling of Shazam users.
Why are there no search engine queries? Surely Google and others would offer up their data as long as it was safely handled.
The lack of this data significantly skews the results as many people will search online and on their phone who don’t use Shazam.This is a significant flaw.
And why no data from other ad music search services like adsense, adtunes or iSpot? All these data points are significant, and if they were included, you’d have a more accurate result.
And what about social media mentions/shares data? Knowing this number also would greatly add to the accuracy of the results.
2. Using overall downloads and streams as data points to correlate to Shazam tags.
Obviously, you can look at streaming and downloads from one date to another, or one time to another, and roughly correlate it to Shazam tags at particular time points of a campaign.
But how much can you really? How much of the lift is correlatable to the ad? What was the streaming/dl/Shazam tag from the previous month or averaged from the previous year?
If you don’t have that data, you’re not really getting to the context of the real lift a song gets from being in a commercial.
And what data sources? What streaming is included? Which services? Is YT included? I would assume so, but I’m unsure as the data sounces are pretty opaque – and what those data soruces are makes a big difference. We need a clearer picture to get better accuracy.
From my perspective, there are two things I’d like to see in this measurement beyond simply an accurate charting of popularity.
Instead of providing data based on weighted numbers from unknown sources at unknown times, what I would really like to know is the percentage of people that streamed a song, saved it or downloaded it directly from the app after Shazaming it, as well as how many people Shazamed/downloaded/streamed a track in the first 24-48-72-96 hours after a campaign goes live, when most of the consumer interaction takes place.
Knowing the percentage of people who Shazamed a song then clicked a link from Shazam directly and subsequently opening a stream or download at what times is far more interesting to me than gross numbers. That shows real engagement and is real, actionable data for the industry.
And the good thing is, you can get that data from Shazam if they wish to provide it – they have it all.
2. Trending and Lift.
It’s great to say that there were X downloads, Y streams and Z Shazam tags during a month, but to get an idea of context, you need to know what came before that figure.
What was the song doing the previous month in DL/streams/Shazams?
The previous year?
Month-to-month since release?
Getting this historic view of what sales/streams/tags a song received before its use in a commercial, then comparing those numbers to the numbers tallied after the song’s usage will not only give you an accurate idea of the lift a song received from being featured in a commercial, it will also provide agencies and brands with an idea of what markets they’re hitting with their campaigns and how much engagement their campaigns received.
Make no mistake – I think this chart is good, and it’s a great tool to present ad music to consumers, but to provide real value, it needs to provide more context and get a better sampling of data.
Next week – my take on what’s needed for a truer sampling of the popularity of music heard on TV.