How DBS adopted the 'Kaizen' approach and reinvented itself

DBS' Chief Data and Transformation Officer Paul Cobban shares with HRM Asia how the Japanese philosophy helped modernise its processes.

The art of Kaizen

I still remember my first day at DBS. When I told the Singapore taxi ‘uncle’ where I was going, he said, “Ah DBS – Damn, Bloody, Slow.”

There’s no doubt that in 2009 DBS had a well-earned reputation for being a bureaucratic, unimaginative and unresponsive bank. 

We also had a new CEO, Piyush Gupta, determined to turn that perception around. We knew we had to do things differently but where do we to start?

Our inspiration came by looking outside the banking sector to the automotive industry and in particular the concept of Kaizen.

Developed by Japanese management guru Misaki Imai, Kaizen is a process of continuous improvement, initially across car production lines. It is the foundation of the lean workflow methodologies you read about today.

Having had the privilege of meeting Misaki Imai personally during a tour of Japanese factories, I could see the power of this process in modern day manufacturing. But could it work in a bank? We gave it a go anyway.

We started with a project we called the RED programme and created a type of workshop which became known as Process Improvement Events, or PIEs.

We had learned from Imai that to improve processes we needed to take out waste, defined as anything that doesn’t add value to the customer. One Monday morning, my colleague and I walked into a room armed with nothing but butchers paper and post-it notes and led a small, slightly bewildered cross-functional team through a five-day process that dramatically reduced the time and effort it took to handle one specific problem - retail account opening by mail.

Making the change

Day 1: Walkthrough and map the current state. First we needed to know what we were up against. We forced the team to physically walk the entire DBS process step by step, taking notes, interviewing staff and recording times. We then created a large "current state map" on the wall indicating each step with timings and issues. We marked the steps that added value with green dots and those that didn’t with red dots. We calculated effort and time for the entire process. It wasn’t pretty.

Day 2: Map the future state. We asked the teams to create a new version of the process for opening this account. This time, we tried to eliminate as much waste as possible. We estimated the resultant levels of effort and end to end times. The results were much better, but so far this was only a theoretical exercise on paper. 

Day 3: Decision-making session. Here, we invited the seniors responsible for the process along with risk executives to review the current and future states. Then, we asked them to go through the list of changes required and give a decision on each as to whether the team could proceed.

Day 4: Refine the solution. Based on the direction given, the solution was refined and an implementation plan developed. We wanted to execute as many of the changes as possible, and we wanted to do it immediately.

Day 5: Outbrief. The team presented the solution to the senior team, and we went ahead.

Moving away from the 'Stone Age'

By refining that one process, we reduced the turnaround time for opening a new account by post from 21 days to five.

We initially ran a pilot of five PIEs across various parts of the business to test and refine the approach. We soon stepped that up to 50 PIEs and then over 200.

After we had run the RED programme for two years, we estimated we reduced customer waiting time by 250 million customer hours and that’s a lot of time. If you go back in time 250 million hours, you end up in the Stone Age!

Add new comment

Plain text

  • No HTML tags allowed.
  • Web page addresses and e-mail addresses turn into links automatically.
  • Lines and paragraphs break automatically.
CAPTCHA
This question is to ascertain that you are a legitimate user and prevent automated spam submissions.
5 + 9 =
Solve this simple math problem and enter the result. E.g. for 1+3, enter 4.