Music Radio's Poison Pill

While discussing the Bridge Ratings' streaming data for his station the other day, the station's program director and I raised the question of what traditional radio's purpose was in 2016.

With so many options available to consumers, studies have consistently shown in recent years that traditional radio's role is to play the hits, the songs listeners want to hear again and again.

Yet, radio's role has always been to play 'the hits'.  It's just that today we are fortunate enough to have the tech that actually shows what the hits are.

Radio has proven to be the curator of choice.
— Dave Van Dyke, President Bridge Ratings

We see it time and again in our radio usage field studies: today consumers choose radio for the most popular music and choose streaming platforms like Spotify, Apple Music, even Pandora when they desire to control what they listen to and when. Radio has proven to be the curator-of-choice. It is clear that traditional radio and music streaming together have been expanding the consumption of music in the last two years.

Then why are so many radio stations not reflecting as close as they can to 100% of their listeners' favorite music?

The information is readily available through on-demand streaming research which Bridge Ratings provides to stations around the country on a weekly basis.

As I've written in this space previously, on-demand streaming is a much better indicator of true music appeal and consumption that any other music research radio relies on.

The least effective method of managing music rotations is a poison pill by programming by consensus, i.e. watching the published airplay charts available and programming based on how songs rise and fall on those charts.

Read my blog "Stop Programming By Consensus" for more detail.

This method causes slow station death as listeners tune out more than they tune in because they are seeking the quick satisfaction of hearing their favorites.

Those industry publications are printing charts based on hundreds of similarly formatted radio stations the results of which is a chart of averages of airplay by programmers who are each watching each other's playlists.

Add the music industry's release strategies and we end up with a chart that really is the furthest thing from reliable data on radio listeners' favorite music.

Using published airplay charts to determine music popularity causes slow station death as listeners tune out more than they tune in.
— Dave Van Dyke, President Bridge Ratings

Case in point.

We recently did an analysis of four key radio music formats: CHR, Urban, Country and Alternative to see the correlation between published airplay charts and true music consumption reflected in on-demand music streaming charts.

Radio finds itself in a position to be led by the available technology - on-demand streaming data.

High correlation means that the two charts - published airplay vs. on-demand streaming of identical lists of songs - are fairly well aligned or similar.

These results may surprise you.

Keep in mind the definition of the word "correlation". Correlation measures the strength of association between two sets of variables. In this case, we are checking the correlation between published airplay chart rankers and on-demand streaming rankers.

Here are the correlations:

How to read: 65% of the top 100 Top 40 (CHR) songs on published radio airplay charts correlate in rank with on-demand streaming charts. 35% of the top 100 Alternative Music songs on published radio airplay charts correlate in rank with on-demand streaming charts.

  • CHR/Top 40 - 65%
  • Country - 60%
  • Urban - 44%
  • Alternative - 35%

The CHR/Top 40 correlation is surprisingly high. The airplay and streaming charts are fairly highly correlated at 65%. This may be attributed to the fact that Top 40 hits are more universally consumed or that Top 40 programmers are more unified in their song choices and rotations.

The same can be said for Country with a 60% correlation. 60% of the top 100 songs on published industry music charts hold rank positions with a margin of error +/- 10%. The remaining 40% hold weekly ranks that do not represent the true consumption by fans of the format as determined by the on-demand streaming research Bridge Ratings analyzes each week.

But it gets more interesting.  

Of the radio music formats dissected for this report, Urban is next with 44% correlation to the published charts top 100 songs by rank.

There are many examples of songs that rank considerably higher on station streaming charts compared to the published charts available.

And we often find songs on published charts with lower rankings for longer periods of time when compared to on-demand streaming data especially when examined by station and market.

The Alternative charts we've studied only correlate with 35% of the songs ranked by airplay. The format certainly offers a unique mix of hit singles and hit album cuts and that variety may be feeding the significant variance between actual airplay and actual consumption determined through on-demand streaming analysis for both Alternative and Urban radio.

For Alternative radio, Twenty-One Pilots is creating all sorts of challenges for radio programmers these days. With the huge success of their current album, we're seeing a number of 21 Pilots' songs rising into the top 75 most-consumed songs by that format's primary listeners. Some of these songs are currents released in the past 45 weeks but many are gold and radio programmers using old school format playbooks are struggling to properly reflect the popularity of this group.

So they don't play many of the most-highly listened to songs by this group. Or if they play some of them, they are not being heard nearly enough by the target listener.

Coldplay is another example. And there are many others.

Traditional radio doesn't seem to have truly migrated to using on-demand streaming data to program its stations. For the first time, radio is no longer able to dictate its musical approach to listeners. Rather, radio finds itself in a position to be led by the technology - on-demand streaming data.

And that reversal of roles is causing consternation among those whose responsibility it is to reflect the tastes of today's music consumer as it applies to radio's role as the source of hit music.

Through viral sharing and social media recommendations, music fans and radio are now on divergent paths.

Bottom line: published airplay charts - in general - do not correlate with actual consumption of music as represented by on-demand streaming charts. Radio listeners have very good taste in music and through viral sharing and social media recommendations, music fans and radio are now on divergent paths.

Radio station programmers who understand this reality will reflect today's true music consumption behavior and will properly reflect consumer tastes.

Dave Van Dyke

Methodology: Correlations were calculated by running correlation analyses comparing an average of four of each music format's top 100 published lists during the month of September 2016 with Bridge Ratings' on-demand streaming rankers for each music format. Song rankings outside of the standard deviation of +/- 10% contributed to the overall correlation variances noted in this analysis.

The Benefits of Using Streaming Data to Pick Music

As on-demand music streaming has become THE manner in which most folks consume music these days, Bridge Ratings has found a growing group of radio station clients that rely on the weekly data we provide.

The reason our on-demand streaming clients align their music rotations closely with our on-demand music streaming data is because it works.

80% of our clients with at least six months using our data have seen the following improvements in audience:

a. Increased daily listening occasions
b. More minutes spent listening with each occasion
c. Increasing daily and weekly cume audience.
d. Resulting in increases in ratings reflected by Nielsen.

Don't pay attention to the myths about streaming research. Actual field experience shows these facts to be true: 

1. On-demand music streaming reflects actual music consumption because users are self-selecting the songs they listen to and listening to those songs frequently and for weeks at a time.

2. Streaming on-demand music is more akin to listening to the radio than any other type of music research because streaming listening behavior mirrors radio listening. Call out research and auditorium testing rely on short hooks of songs played to a small sample of possible station listeners. This is not the way people listen to music.

Believe us: people don't listen to their favorite songs this way and based on our research, fatigue and boredom impact these types of research scores.

3. Streaming data reflects the total spectrum of consumption: positive, negative and neutral. By focusing in on station format, demographics, local market data and listener streaming data, the Bridge Ratings streaming reports are much more granular, i.e. station-specific than broad-based, national rankers of streaming music.

Songs that are not popular over time do not maintain rank status. These songs either don't rank in the top 100 each week or they simply die on the vine as mid-charters. Saying that on-demand streaming research doesn't reflect songs people don't like is inaccurate.

4. Airplay charts only occasionally reflect actual music consumption.

More often than not published airplay charts are built on consensus, i.e. hundreds of station playlists are merged together to present an overall ranking of how much airplay songs are receiving on those stations. Programmers often depend on these airplay charts to align their music categories. This approach simply does not reflect market and station specific listener behavior. Adding to this, record labels are always interested in seeing their new releases climb these consensus charts so record promo folks can encourage other stations to play the record, whether that record is appropriate or not.

Use of market-specific, format-specific and listeners specific on-demand streaming data is a far more effective way to reflect true listening consumption.

5. Streaming data is a true reflection of actual consumption. The data is sensitive to the popularity of a song and any associated fatigue. When people slow down or stop their streaming of a song, its rank falls and either settles at a spot on the chart where it lives for some months (Walk the Moon's "Shut Up & Dance"), or it fades into ranking below 100.

6. With billions of streams nationally, and thousands at the market and station level, on-demand streaming produces a significantly better "sample" than any other music research. These large samples, even in small markets, reduce the margin of error so significantly that the data is consistent week to week providing a higher degree of data trust than any music research available.

7. In our experience of working with on-demand streaming data for more than two years with our client stations using the data weekly, streaming data more often than not is ahead of the curve.

Not only does streaming data pick up popularity immediately, it often is far ahead of record labels in determining successful hits and follow-up hits.

One example of this phenomenon is the Chris Stapleton song "Tennessee Whiskey" which immediately flashed to the top of the Country radio streaming charts right after Chris received his armful of Country Music Awards in 2015.

While the televised awards show was still on, we saw "Tennessee Whiskey" climb to top 10 status. The song has remained in the top 30 since, yet gets little radio airplay.


From our feedback from our station programmers and music directors, many had not been serviced with the song by Stapleton's label so they couldn't play it. Because so many stations stayed off the song, it never saw any significant chart action on the published airplay charts, yet the song remains hot with Country listeners.

For programmers that rely on these published charts they are missing one of the most popular Country music songs of the year.

Undiscovered songs listeners are consuming at high levels but which radio has ignored is another benefit of on-demand streaming research.

There are numerous examples of this phenomenon.  Also in the Country format, this week and last saw the new Florida Georgia Line song "Smooth" break out in the Country top 10 in Nationally and in many of our markets. It doesn't appear on any airplay chart yet.

While there may be "limitations" in most types of research, on-demand streaming eliminates most research negatives and if you are thinking about using streaming data or are using it. consider these real-world examples of on-demand streaming use cases as confirmation the data works to properly align radio playlists with that of its listeners.

As one of our staunch supporting station owners proclaimed when the ratings book was released after using our data for six months or more, "This sh*t works!".