How to identify gaps in the payment of non-domestic rates

By Richard Duffield, Senior Consultant at GeoPlace  Dicky

Maybe you’ve heard the phrase “I’m sorry for writing such a long letter, I didn’t have time to write a short one”? A quick Google search suggests Blaise Pascal is credited with saying this in 1657. I know how he felt. This is how I feel when analysing data for clients. Increasingly we all feel data rich and time poor. The people we work with are happy to look at data that helps them solve their problems, but they need it to be concise, understandable, and usable.

This is the challenge we faced when asked to help local authorities identify gaps in the payment of non-domestic rates.

First, let me give you a bit of background to the project…

Business rates is a hot topic. Until 2013 every councils collected business rates into a national “pot”. Central government distributed this back to local government as a grant, using a formula to decide how much each council would receive. Under this system councils may have received back more or less than they collected.

In 2013 this changed and since then councils have been entitled to retain 50% of the business rates revenue they collect giving them a clear driver to ensure that they are collecting 100% of the income which is due.  In fact, there is an ongoing discussion within government on the Government’s commitment in October 2015 to allow local government to retain 100% of business rates raised locally, making full collection of rates even more of an imperative.

Public sector senior managers increasingly recognise fraud and error as an area worthy of attention and investment giving us belief that this is a problem that they want to solve.

It was easy to see that an accurate property register is essential for local authorities who want to make sure they collect all business rates and avoid fraud and error.  We weren’t the only people to realise this. Our partners at Ordnance Survey (OS) and the Chartered Institute of Public Sector Finance Accountants (CIPFA) were already on the same page. They approached us to work together to find a way to help local authorities.

Often the challenge for data professionals is gaining support and funding from senior managers. The key is to find out what their top priorities are and show how data can be used to solve these problems. Knowing that senior managers were already thinking about these issues gave us a great opportunity to target our messaging.

So, we needed to come up with a solution. We had a clear view of what we needed to produce: big numbers. Lots of properties in the report would mean lots of uncollected business rates which would mean a large return on investment.

Salford City Council volunteered to work with us on a proof-of-concept study bringing together their property data, business rates and fraud and error teams.

Councils produce not only a business rates register but a definitive database of all properties known as the Local Land and Property Gazetteer (LLPG). It is this LLPG data which is compiled into the national address gazetteer and made available as AddressBase Products®. We compared their business rates register to their LLPG and the results looked promising. Just as we hoped, the numbers were big. One record stood out – a hotel. This would have a high rateable value and so would provide great return on investment.

We were excited to send the results to the team at Salford but before we did we wanted someone to check them, so we sent the data to our data analytics team. Unfortunately, the news was not as good as we were expecting.

There was good news. The report proved the concept and highlighted properties worth investigating. The benefits appeared to outweigh the costs. The bad news was that many of the records were false positives. The data analytics team found that by looking in detail at individual records they could find them on the business rates register. This wasn’t because the gap analysis was poor, or defective. Often it was simply due to differences in the way properties are recorded in the two datasets. For example, we were correct, the hotel was missing from the business rates list. However, next to the hotel was a public house. One company occupies the hotel and the pub and so the business rates list has just one record which holds the address for the pub.

We then spoke with Salford and asked them what they would like from our service. They said that most of all they wanted to avoid false-positives! Like many councils Salford manage limited resources and competing priorities. They were happy to review the report but were only able to commit a limited amount of time. Based on previous experience they also challenged us to produce a report which quickly built credibility. A report containing too many spurious records is likely to lose credibility and be ignored.

Our users were clear that individually reviewing large numbers of false positives was not an option. We needed to find a slicker way of removing false-positives while keeping the valuable ones. If not, we would have to accept that we could not help them which we didn’t want to do. So, we rolled our sleeves up and worked through the data to find patterns to help narrow down the search. By using our detailed knowledge of the data, we were able to build algorithms which selected the records most likely to yield a result and remove the rest.

This produced a shortlist – literally a short list of addresses – which proved to be much more cost-effective. Salford were really pleased to see such a short report and even more pleased when they found it to contain records which were missing from their business rates register.

Running data reports is not particularly difficult, the tools are now available to make this easy. The key is to create user-friendly and cost-effective reports which takes time, care and skill. This is what we should be striving for if we want our data to be used to bring about change. It takes longer but it is worth it when we see our work having an impact.  Our challenge for this year is to produce concise reports and not to find ourselves, like Pascal apologising for producing long ones.

Find out more about our service at


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s