Cookies disclaimer

I agree Our site saves small pieces of text information (cookies) on your device in order to deliver better content and for statistical purposes. You can disable the usage of cookies by changing the settings of your browser. By browsing our website without changing the browser settings you grant us permission to store that information on your device.


Last week, you will have read our blog looking at the highlights from the first half of 2020. The half year, and specifically second quarter of 2020 has been unlike any quarter we have previously seen. The impact on judgment levels seen in this period was greater than that which occured post financial crash in 2008.

I knew, that this was going to make for a fascinating set of quarterly statistics.

As an essential part of the Data Analyst role here at Registry Trust, I have become a dab hand with some of the methods we use to analyse data. So I thought I would take you behind the scenes and show you how the project comes together from start (extracting) to finish (publishing).

First of all, I have to introduce my software. Key to facilitating my work is Tableau to produce all our statistics in table and visual form. Tableau’s mission is to ‘change the way people use data.’ This is something that really resonates with me. If data is accessible and transparent it is more likely to be received and understood by those outside the stats bubble – which for us at Registry Trust is where the majority of our audience lie.

So how are they produced?

To get started, I need to create an extract which includes all the data we need to do the analysis, which remains constant or fixed. This is important because our data is naturally dynamic. Every day we receive judgments that are not necessarily all from the day before. They can be from months before, and sometimes even years before. In order to make sure data is replicable and reliable we extract the data and transform it into a static dataset.

For quarterly reporting, I actually extract the data for the 15 months prior to the last day of the relevant period. This ensures data comparability. For this set, the final day was 30 June 2020 and so we’d usually use a start date of 15 months ago. This time was a little different though, as we were also looking at the half year. This meant I needed to include the entirety of HY1 in 2019 and so the extract began on 1st January 2019. That’s a lot of data!! So I get on with other work whilst Tableau produces the extract, occasionally trying to second guess how many rows of data will be revealed.

BTS 1.png


I have preformatted workbooks that I use for quarterly reporting. This ensures consistency across all our reports but also means that when the new extract is made, the tables automatically adjust and the new tables are produced with ease.

I then export the jurisdiction-specific tables, which you will see on the end of one of our quarterly press releases, and send them off to our communications team so that the press releases can be produced.

While that goes on, I get to do what I think of as ‘the fun stuff’.

After taking the top-lines of the quarter as whole, I look to each jurisdiction to try and identify any patterns or trends in each area. We introduced commentary slides in our statistics book (the coloured page in area jurisdiction section) so that we could offer those interested a bit more depth into the judgment behaviour for the quarter.

For example, this half year alongside the falls in the number of records there were also increases in the values for judgments. For consumers in Scotland, the median had risen 12% year on year. To investigate what had been happening, I broke down the 1,419 we had received in Q2 2020 by amount (see graph below). In doing so, the visual highlighted immediately that there has been a large decrease in the number of judgements that were between £100 to £500 (5 percent points, year on year), but an increase in the number of judgments above £1,500 (7 percent points). This evidence allowed me to conclude, that during the economic unrest experienced in the quarter, attention was given to reclaiming greater amounts of debt owed and lower value decree were less of a priority.

BTS2.png

Being able to draw further analysis out of our statistics means that we are able to provide a more holistic view of judgment data and how judgments are experienced in real terms. If we were to talk about judgment movement in isolation, there will be little appreciation given to the fact that judgments are real things that have large impacts on people’s lives. To me, these slides are some of the most interesting to create.

Once the statistics book is fully visualised and written, it is time to check through it. This can be a laborious task, but we take pride in ensuring what we publish is correct. The press releases will have come back at this point so I will check, check and check them again too.

As explained earlier, the statistics report are a snapshot of our dynamic data. In order to bring some dynamism back, I create a dashboard using Tableau to allow those who like to interact with data actively to be able to explore the quarterly data too. Using another Tableau workbook, I visualise the top-line statistics, such as the amount breakdown, proportion of satisfactions by jurisdiction and mapping of judgment density.

BTS 3.png


To design the appearance and placement of graphics in a dashboard, Tableau uses containers (the grey shaded are on the image above). These can be fiddly but makes sure that graphics all fit inside the pre-set dashboard dimensions. It can just be a good idea to have an idea of where things are going because it can be very annoying when you have to start again because you can’t place the container where you want it to go. Just like restarting in banangrams (if you know, you know).

Once I am happy with the dashboard and the information it is providing, I will upload it onto Tableau Public so I can then embed the link onto the specific page on our corporate website.

I think dashboards are great places for those of you who like to investigate data visually. Tableau neatly presents the data in graphs and maps and then can offer users greater information when they are interested. But if they are not wanting the specifics, the general trends are seen just through looking.

BTS 4.png

(If you like interactive dashboards, you should check out Registry Trust’s Financial Stress Tracker. It is fully interactive and will filter graphs when you selects measure you a specifically interested in).

At this point, we are ready for publication. We upload the statistics book, the dashboard and the press releases onto the website.

To finish off, I will write a blog the following week to summarise our most interesting findings as the final method in which we communicate our stats. This format is my favourite as blogs come with a chattier tone (if you hadn’t noticed).

I hope you have enjoyed this behind the scenes look in to how we produce our quarterly stats. See you next time.

If you want to keep up to date with our latest blog posts, you can click here to subscribe for updates.