Skip navigation links
Member resources
Help Centre
News and features

How we collect music usage data 

Ever wondered how we know what music is being used across the UK, on TV, on radio, in pubs, shops, hotels and restaurants?

The first point to note is that we don’t know the details of every play of every work on every stereo in every premises and every broadcaster we licence. Whilst we aim to make our royalty payments as accurate as possible, to collect and process information for each individual play of every work across the UK would, in some cases, cost more than we can actually collect in royalty revenues for those plays.

So how do we know what music is being played?

Although we use a 'pay-per-play' or census method wherever it is feasible, such as when music is broadcast on national radio and terrestrial TV, we also employ other methods to calculate royalties when this is not practical or cost-effective. All are based on the principles of offering the best balance of fairness, accuracy, cost effectiveness and transparency for our members.


This is the preferred basis for calculating royalty payments. It’s often referred to as pay-per-play and is where we collect, process and make a payment for every single performance.

We often use this method, for example, census is the method used for the majority of:

  • terrestrial TV channels - all BBC, ITV, Channel 4 and Channel 5 TV and 88% of Sky revenue
  • online usage such as music streaming sites – iTunes and YouTube
  • radio broadcasts – all BBC radio and 93% of commercial radio revenue
  • major concerts – 100% of all pop and classical concerts

This list is not exhaustive and we aim to use census data collection as much as possible. We have moved lots of usage to census over recent years, but unfortunately it is not always feasible to distribute on a census basis either due to difficulty in obtaining a full dataset, or the cost and resources that would be needed to process and calculate census royalty payments relative to the value of the licence.


In cases when census is not feasible, we aim to collect a representative sample of actual performances as the basis for our distributions.

This method is effective when:

  • music usage is highly repetitive or there is a small breadth of total repertoire used. An example of this would be songs performed live in themed bars on Cruise Ships where the repertoire is often limited to popular 'standards' and repeated on cruises throughout the year/reporting periods.
  • we can collect this information cost effectively.

A sample may include a smaller group of venues to represent the wider group of venues of the same type. Alternatively, it may consist of a few days of broadcasting to represent a TV channel’s music use for the whole month. This is used when those few days fairly represent the month as a whole.

However, sampling is not always an appropriate distribution basis if data collection is expensive and there is a very high volume of music usage. For example, it would not be cost effective to collect a statistically representative sample of background music played in pubs because such a variety and large volume of music is played and spread across a huge number of venues.


An analogy-based payment is used for any source of revenue where pay-per-play and sample methods are not feasible. This is usually because sufficient data is unavailable or it would be too costly to use these other methods. This is the main distribution method currently in use for most uses of recorded music in public places i.e. in shops, hotels, restaurants etc.

Here is how we identify a similar of "analogous data" set to calculate our royalty payments:

  • We collect source data by sending music researchers to pubs, clubs and hotels all over the UK to record the music being performed in public
  • The collected data is then compared to the data we have received from different radio and TV broadcasters
  • We run these comparisons to find a ‘best fit’; a set of radio and TV broadcasters who music is most similar to the venues we have surveyed

We then distribute the public performance royalty for a particular revenue stream e.g. Pubs across the ‘best fit’ data Note – a separate calculation is made for different types of venue e.g. pubs, shops and restaurants, Jukeboxes, Karaoke

Once compiled, the set of stations and channels either remains the same for three years, or until distribution policy changes result in a recalculation of the analogy. Because of this, the set of stations and channels which make up a given analogy are temporary and subject to change.


In some instances, combinations of all three approaches may be used to calculate our royalty distributions. We are continually reviewing our distribution methods to keep pace with market developments and new technologies. Our goal is always to ensure distributions are as fair, accurate and cost-effective and transparent as possible.

Where does the money come from?

Browse through the various royalty sources available for members of PRS and MCPS.

Follow us on M Magazine Twitter Facebook Linked In YouTube
© PRS for Music