How many listeners does my community radio station have?

By James Cridland
Posted 29 January 2017, 6.06am est

I've just done some work to calculate an estimate of listening figures for every radio station in the UK - yes, including community radio. I look at over 200 stations in RAJAR and compare their weekly audience figure with their Facebook/Twitter/TuneIn figures. I then use the same multiplier - which is relatively statistically accurate - to calculate figures for stations that aren't in RAJAR.

Here's one example for the station in Exmouth, Bay FM. Search for your station, and click the 'listening figures' link in your page for details.

You'll appreciate, I'm sure, how many times I put the phrase "estimate" all over these pages. I'm still not sure I believe them. But take a look and see how your station has done!

Feedback is very welcome.


1 year, 8 months ago

It's an interesting way of looking at it, would be good if we could see the underlying algorithm and also the estimates it produces for stations that are in RAJAR so we can see the disparity for each station.

I would have thought the BBC/commercial stations would have been pushing RAJAR for a breakdown of All Internet Stations / All Community for example.

1 year, 8 months ago

The algorithm is, for stations with all three social media networks:

((total number of tunein followers multiplied-by tunein multiplier) + (total number of tunein followers multiplied-by tunein multiplier) + (total number of facebook followers multiplied-by facebook multiplier) + (total number of twitter followers multiplied-by twitter multiplier)) / 4

... as you can see, TuneIn is weighted twice as highly as the others, due to its comparative correlation.

You can calculate the multipliers by visiting the comparison pages that we link to, and filtering for stations with a maximum of 50,000 weekly reach.

The algorithm comes straight from comparisons of RAJAR vs TuneIn (etc), and so the estimates it produces for RAJAR'd stations would be similar to the actual figure. That's, kind of, the point.

Being very clear: I believe these figures are too high. Much too high. It's interesting to note, however, that those stations who have run surveys appear to say they're in the same ballpark.

1 year, 8 months ago

Rinse FM's estimate would put them ahead of many London stations including BBC Radio London. I wouldn't be surprised if they subscribed to Rajar that their Rajar would be in the region of around 500k, which would make it the UK's largest CR station.

1 year, 8 months ago

I'd wager that you'd need to downplay younger stations' data as they're happier to follow/fan etc and perhaps up-weight older ones. So perhaps Rinse is marked lower and Angel marked higher.

But having had a look through, I think it's all a little high, though in the right sort of proportions when comparing stations.

1 year, 8 months ago

Population of Falmouth and Penryn: 28,801 (2001)

Estimate of listeners to Falmouth/Penryn CR station: 17,746

Maybe Beaumont has something after all...

1 year, 8 months ago

@matt I think you're right. Rinse does do very well, but will benefit from the young demo. And I also do think they're a little high. For that Falmouth station, my guesstimate would be a 25% reach - 7,000 - might be vaguely possible but on the high side.

The figures work for commercial radio, but I wonder whether the community services are just too small to be accurately mirrored in this way.

1 year, 8 months ago

Resonance's estimate of 96k may not be far off to estimate how many people are listening to CR in the capital, while Reprezent at 50k which is youth skewed may be off.

1 year, 8 months ago

I'm surprised that nobody has included data from app downloads as surely this offers a more sensible figure of listeners than the number of Facebook likes. After all Facebook interaction could be no more than liking and sharing news articles without ever listening?

1 year, 8 months ago

I'm surprised that nobody has

...normally ends "because that's not possible".

The Fun Kids Radio Android App Android app says it's had "1,000 - 5,000" downloads. On iOS, you get no indication at all of downloads.

We pull in ranking data for each.

After all Facebook interaction could be no more than liking and sharing news articles without ever listening?

You're right. However, there is a correlation between the number of Facebook likes and the number of listeners.

You can now see the estimates for surveyed stations too - Here are BBC Radio Newcastle's surveyed figures and here is our estimate. Just add "?estimate" to the end of any listening-figure URL to see those.

Two things to bear in mind: 1. the algorithm is tuned to smaller stations; 2. Fun Kids - if you try to compare that one - is only surveyed in London (approx 10% of the population) but is available nationwide.

1 year, 8 months ago

I've withdrawn this service, following some email correspondence with a number of different radio stations.

1 year, 8 months ago

Shame - it was a good effort - but just had a few too many holes in the way it was applied - for example a tiny internet only radio station with maybe a hundred listeners - was given an estimated audience of 40,000+ where even the local ILR doesn't have that many listening - mainly because of it's sterling efforts buying likes on facebook and twitter etc.

Tunein stats are a more interesting indicator but even these are distorted by what other online platforms stations are using and whether the programmes being broadcast are of more global interest outside their station's TSA.

1 year, 8 months ago

Trying to estimate a listening figure I can see causing issues with some stations if such a prominent website such as this has a figure that's significantly different to what they may be quoting themselves.

Perhaps instead there is something that could be drawn from these metrics as some sort of "audience engagement ranking".

There may well be a pretty strong correlation that can be drawn between social media and listening figures for some community radio stations, say some of those in smaller towns, with a fairly middle of the road format playing the local card.

As others have highlighted there is strong evidence that this won't work as well for community of interest stations, or stations that have a range of specialist interest programmes, with some listeners who may only listen for an hour a week.

Surely if the want for "listening figures" is for commercial sales reasons, a well thought out audience engagement metric could be just as useful.

I bet most community radio stations have had the "but you're not in RAJAR" conversation more than once. It could be powerful if they took the line of offering access to an highly engaged audience, rather than just the most bums on seats.

Login or register to comment
It only takes a second with your Google or Facebook account.

- follow us on @minfodiscuss