Skip to main content
Live from EBRC - Business Rules Approach Applied

Next EBRC session is Business Rules Approach Applied with Alcedo Coenen, formerly of ING. He was talking about knowledge analysis and how it can contribute to process optimization.

Within banks, and other organizations, there is much to be gained from from process optimization particularly efficiency, decreasing the number of "defects", more transparency and increased agility. ING used knowledge analysis to do this and got some surprising effects on the organization. ING is a huge bank and insurance business with over 120,000 employees. The particular case was the mortgage application process across several ING brands in the Netherlands where ING has about 25% of the market and about 150,000 applications a year.

The basic process involved getting an application into the system and then checking to see if there is a reason to reject (decision) and, if not, an offer is immediately made. If there is a reason to reject, the next decision is to see if an adjustment was possible and resubmit it again if it was. If not, then the third decision was whether the request was inside the staff's mandate or not. If it was, someone else in the department would call and try and make an override and if not then it was sent to a special group who investigated more thoroughly and then tried to make an override. Initially only 16% got an offer automatically, 70% required an override and 14% ended up getting rejected! Clearly very inefficient as 70% were going through the manual process.

First step to improve the process was to apply 6-sigma/lean techniques to the process. For instance, only asking for evidence forms after someone accepted an offer. Overall, however, focusing on the process did not make a big difference and only handled efficiency and defects - it did not help with agility or transparency.

The second step was to use knowledge analysis - decision analysis if you will. Manual processing in the old system meant that the criteria were pretty simple (people could not handle complex rules) and the workflow was driven by authority (from those with the least authority to those with more). To analyze the decisions they took all the rejections and analyzed them for parameters that led to rejection. For each parameter that was changed or overridden, they tried to establish if they could formalize a rule to explain why someone was initially rejected and, if not, they built something to support a manual process. For instance, they found that people with insufficient income who were subsequently made an offer often were young and had a growth-oriented career. This would result in a rule to make an offer if the income was too low but close and the age and profession of the applicant were suitable. If they could not formalize a rule then they focused on guidance and support for the manual decision.

By turning the experience into new rules they were able to drive up straight through processing (STP) and decrease manual processing. They also found that people moved from manual processing to refining rules. This meant that operational work was gradually replaced by knowledge work and eventually the total work dropped resulting in lower staffing and so on. This in turn led to an organizational simplification where investigation was the only manual process and the people in the department were engaged in both handling real exceptions and defining new rules (true knowledge work). This knowledge analysis got great results and improved the agility and transparency. Process workers become knowledge workers as a decrease in operational or procedural work was replaced by increased knowledge work:

  • Manual Process -> Automation
  • Procedural thinking (order, sequence, authority level) -> Functional thinking (validity, control)
  • Simple rules -> Refined rules
  • Acceptance competence -> Analytical competence
  • Performance management -> Risk management

This was a great illustration of how applying decision management to a process increases STP rates, improves the process and allows staff to be deployed from manual processes to risk and business management. His slides are here.

Technorati Tags: , , , , , , , , , , , ,

related posts