CRS 2014: Finding hits before they’re hits


CRSThe panel, “Cracking the Code: Using Data to Discover the Hits,” revealed new research on predicting a hit song 2/20 at the Nashville Convention Center. The data was collected by MusicMetric, the world’s largest music data analytics firm, providing data on online consumer behavior for the entertainment industry, and by Stone Door Media Lab, headed by industry vet Jeff Green.

Green, along with Mark Tindle of MusicMetric, revealed their finding of data collected on consumer activity on social and online media, including Facebook, Twitter, BitTorrent and more. Nick Fishbaugh of Shazam also participated in the panel, the results of which could give radio programmers and record promoters a competitive edge in discovering and predicting new music.

“A funny thing happened on our way to determining the role social and online media have on making Country hits. We discovered a predictive model,” said Tindle, SVP/Sales and Business Development at MusicMetric. “The reality of being able to reliably predict a hit song within the first four weeks of its release has a major impact to the business of music. Key social and online media sites such as Facebook, Shazam, Soundcloud, VEVO and even BitTorrent are showing ever-increasing roles in that success.”

The Stone Door Media Lab participated heavily in the research by analyzing correlations, particularly between radio airplay, sales and Shazam.

“There is no foolproof method for determining probabilities for hits, but we’re seeing that data can provide a competitive edge in identifying a new single’s potential many weeks earlier in the artist development process,” said Green. “These statistics prove that what many thought were hunches are really true. We plan to dig into the numbers further to help provide additional insights.”

Here are some important takeaways:

•           Emerging Social and Online media outlets have fragmented the once linear exposure channel for music, making it harder to identify hits.

•           By tracking 81 Country radio charting singles over 28 social and online metrics (including radio plays); Musicmetric was able to determine milestones in the journey of a hit and find the significant online drivers of that success.

•           While more time and development is needed to perfect the model, MusicMetric found conclusive evidence that by achieving certain thresholds in online and social media consumer activity, it is possible to predict hit songs (top 10 and even #1s) in as little as 4 weeks from the track’s debut on the Country Aircheck/Mediabase chart.

•           Key online drivers for Country genre success include: An increase in Facebook Fans, SoundCloud plays, Shazam tags, VEVO plays, Shazam tags per Radio Spin and BitTorrent file sharing.

•           The Stone Door Media Lab found that Shazam and sales-per-spin ratios are powerful new tools to predict airplay success at an early stage. The data also supports a link between being No. 1 most added and reaching No. 1.

•           The average peak on the Country Aircheck/Mediabase chart of a No. 1 Most Added single is 6, while the average peak of a No. 2 Most Added single is 12. These songs far outperform the average peak of all Country singles, which is 20.

RBR-TVBR observation: Being able to quantify social media activity on emerging and new songs gives the PD and MD a great tool for bringing up ratings. The sooner they get this data and move it to the playlist, the more listeners will come to realize that particular station—be it Country, CHR, Alternative, etc.—is the place to turn the dial to for music discovery—not the competitor in the market and not Pandora.