Streaming Research & Radio: The Perfect Match




DATE: JUNE 9, 2019

As you know, we've been analyzing the research potential for on-demand music streaming since 2014 and we have analyzed billions of streams of music preference and matched that information with radio formats and radio listening.

We have streaming clients who swear by music streaming data as the best way to capture true music consumption.

We know that studying on-demand music streaming behavior of radio format P1s (primary core listeners) has proven they have different streaming behavior than the overall universe of streamies, a group which includes both light users and non-users of radio.

After four years what have we learned?

Based on our analysis, the use of streaming research to align a radio station's music exposure with true popularity is much more efficient and more reliable than any current forms of music research we studied.

In fact, the variety of methodologies of other forms of music research can cause confusion about a song's health which often leads to incorrect assumptions due to conflicting information.

Analysis paralysis is the state of over-analyzing available information so that a decision or action is never taken.

In a four year analysis project, Bridge Ratings compared music streaming data with other forms of industry standard music research* to determine how well each correlated and succeeded over the long term to predict music consumption and purchase.

As most of you already know, the on-demand streaming research we provide our clients is format-based which is a more reliable measure of radio's use of this data. We looked at a) how well the data predicts high consumption within the first two weeks of a song's release and b) how well the data correlates to highly-consumed songs over the long term.

Based on our analysis the use of streaming research to align a radio station’s music is much more efficient and reliable than other forms of music research analyze.
— Dave Van Dyke, President Bridge Ratings

Our analysis looked at billions of genre specific music streams*, call-out research studies, auditorium music tests and MScore reports.

Streaming Correlation Update 06.2018.png

The most-effective type of on-demand music streaming data is that which is sourced from P1 listeners (heavy users) of a radio format, followed by market-level data and National streaming data. Auditorium testing with 51% correlation correctly predicts a new song's popularity half the time over the first thirty days yet does improve after. Internet-based Auditorium Tests (AMT) does provide more confidence in popularity prediction.

The reliability of on-demand streaming remains stable when looking at streaming performance of new music over the first two weeks' of release.

Streaming Correlation First 2 weeks of release Update 06.2018.png

As a predictive tool  standard industry music testing has shown in our analysis to be less dependable within the first weeks of a song's release.

Why is this form of music preference better?

Because it is the only true method to reflect actual consumption with higher reliability over sales since frequency of play cannot be determined with sales metrics.

There are also tools to help programmers measure a song's consumption passion and burn. We often see songs which sit at the top of the streaming popularity chart (similar to radio's cume metric) experience reduced frequency of plays over time. 

All of this insight brings higher-level intelligence to programmers of music who wish to align playlists with true listening behavior. No other research currently arms radio professionals with such a powerful tool.

If you have any questions about this analysis, don't hesitate to contact me.






*Source material for this analysis: 1 billion on-demand music streams, 520 weekly call-out music tests, 100 auditorium tests and 1000 MScore song reports. from 2014-2018 YTD. Radio formats studied include: Urban, CHR, Adult Contemporary, Rock, Country.