The ideal size of a station’s active music database is an age-old debate between programmers. Those that figure out that sweet spot number that’s just big enough to keep TSL high, but small and mass appeal enough to reach a broad audience and drive cume, typically find ratings success provided the rights songs are chosen. But, what is that right number and why are so many programmers drawn in by the urge to lean toward big databases? 

I’ll tackle the latter question first. There are a multitude of reasons why programmers fall into the trap of the mega-database. A simple one is that many programmers add more songs than they rest or delete, hence over time their databases gradually grow out of control if they don’t take the time to cull out the weaker, less popular tunes to get back on track. Another reason is, nobody pats you on the back for playing a song everybody else is playing. Regardless of personality type, everyone enjoys hearing how smart or cool they are for exclusively jumping on a tune and it also tends to come with more love from artists, promoters and labels. Plus, many of us are in this business because we have a deep affection for music, which makes it hard to pass on great songs that never really caught on. I can’t count the amount of times over the years I’ve had to say to a fellow programmer, ‘We can’t rewrite the history of music based on what we think should’ve happened.’ Music is a cruel and unforgiving business that doesn’t necessarily reward the most talented or hardest working, because it’s forced to follow the whims and constantly changing tastes of the masses. 

So how big should a database be? That depends on a whole host of factors, format, demographic, market size and competitive situation within that market. Some of my fellow consultants and radio friends who I respect lean toward the larger end with 1,250 or more, while others are much more conservative in the 200 range. I tend to fall somewhere in the middle with 250 to 500 song databases for most new music based stations, and 500 to 750 for recurrent/gold based stations. Granted, I typically keep another 100 to 200 on hold for refreshes. 

Itunes and Spotify have given us some great insights into answering the elusive question: How many songs does the average person know? The average Spotify playlist has just 170 songs in it. Yet, according to TuneUp the average iTunes library has an astonishing 3,000 songs in it. That’s significantly more than I would’ve expected, however the challenge for programmers isn’t to answer the ‘how many songs does the average listener know’ question. Our challenge is to find the number of consensus songs within a specific demographic and a specific format. That’s why traditionally stations that air on the side of caution and go with smaller databases tend to do better. It’s simple math, if you pick the 200 most popular songs within a format you’re much more likely to have the majority of your target listeners know and sing along with 90% of those songs. 

However, I would argue that with the easy availability of the world’s entire catalog of music, today’s terrestrial radio databases can push the limits a little more. After all, there are currently 50 million songs available to stream on Spotify. 

What are your thoughts on database size. Comment below or email me at andy@radiostationconsultant.com.

Photo by Expect Best for www.pexels.com.