With the help of Mobysoft it is now a successful reality.

Paul Lees, Chief Executive, Adactus Housing Group explains:

In 1980, I got my first job in housing as a housing officer (rent man) in Wigan. I was the only housing officer who didn’t have a car, so I cycled to my first house of the day and left my bike to do my round on foot.I then cycled back to the office with a couple of thousand pounds in my satchel to reconcile my cash for the day. I had usually finished by 11.30am. My car-bound colleagues cashed up at home and returned to the office at 4pm. I had a lot of ‘me time’ on my hands, so I decided to do all the office’s arrears.
I got the large, green-print, blue-bound arrears details that head office sent once a week and some standard letters run off the Banda hand-operated copying machine and set to work. I wrote every address out in full with the amount owed. I then put these letters in envelopes and put them in the post. Each week, I got through approximately 2,000 cases in the space of a few hours. The office only had one phone but it didn’t matter because hardly any of the tenants had a phone, so there was no one to ring. Tenants either ignored the letters or came to the office to make a payment.
Jump forward 30 years to 2010
I asked an arrears officer to explain why our arrears were not coming down. He ran through the now-computerised arrears system, jumped in and out of screens and pressed a variety of buttons that eventually came up with a standard letter that ran off the printer. It took as much time for our 21st century arrears process to produce one letter as it did for a bored 1980s novice to produce 30 of them. Something was wrong.
I reviewed the whole system and decided that it was time not just to revise our arrears process but to get rid of the whole housing management department – but that is another story.
We searched for a solution to why it was taking people so long to get through what is a very simple process. People sign a contract with us that clearly states that they will pay a rent in return for the use of a property. If they don’t pay, we need to highlight this breach of contract to them, explain what the consequence of not paying will be and offer advice if they are having difficulty paying. What could be more straightforward?
This was the question I put to the arrears team who worked in housing management at that time. The reasons why we couldn’t pursue arrears as simply as I envisaged were legion. Late housing benefit, true arrears, 48 as opposed to 52-week rent rolls, not being able to make telephone contact, visits to people who were not at home and so on.
 
True picture
After a long search, we eventually found Mobysoft, who had a product that used predictive algorithms to clearly identify true arrears. This meant that our arrears team could now see who owed what and when. Progress, one would think.
However, human intervention again put a spoke in our wheel. We might know how much people owed but then housing officers (ersatz social workers) had to decide why and how to escalate any action we might want to take to pursue the money owed.
The straightforward speedy process ground to a virtual halt as people ploughed through the data to make a decision and then decide which letter to send. The pen and intoxicating Banda ink of the 1980s were still winning.
We reviewed the process again and identified the time in the process when algorithms gave way to humans that slowed the process down. We asked a simple question – why involve a sentient creature in what is a cut-and-dried process? Why not get humans to do what they are good at and let computers do what they are good at – processing data? So we did.
Technology Solution
We got Mobysoft to link in with Housing Contact, which provided a messaging service that we were already using to undertake tenant satisfaction surveys.
We then linked our current arrears directly to an automated telephone message that told tenants that they owed us money and offered to put them through to our contact centre at no cost. If tenants didn’t respond the first time, the message repeated until they did. We consulted tenants to see if they liked the idea. The overwhelming majority of tenants consulted did, so we started the process. This was August 2013.
Data triumph
The automated, data-driven process we introduced in 2013 is still in use today. Tenants who owe us money either make contact to come to an arrangement to pay, or the arrears case is escalated through Notice of Seeking Possession to an application for court action.
At court stage, humans come back into play. Data showing how many times we have contacted the tenant, combined with the details of arrangements made and advice and assistance offered, chime extremely well with the courts and our success rate on positive action from judges has increased dramatically. Finally, the 21st century has beaten the 1980s into a cocked hat.
Since we introduced our cyber rentperson, we have:
  •  Increased weekly contacts by 700%.
  • Arrangements to pay are up by 46%.
  • The amounts that arrangements are made at have reduced by 38%.
  • We have received one formal complaint.
  • We have reduced the number of people chasing arrears by 50%.
  • We have increased the number of people providing financial advice by 50%.
Of course, the real proof of this algorithmic, data-driven pudding is what has happened to our arrears levels. Of course they have gone down – I wouldn’t be writing this article otherwise.
We had a vision of how arrears should be approached, but our technology partners provided the means to make this happen. Hats off to Mobysoft and Housing Contact, the real heroes of this story.

 

This article is taken from Inside Housing – 12th March 2015