Banking Process Management from A to Z

In our business lives as engineers, we all encounter with the famous adage “If you can’t measure it, you can’t manage it.” There are several claims about this adage being not accurate all the time. But it is definitely acceptable in some cases.

Let’s think that we are elected to design a bank’s processes from A to Z. It is a very challenging duty and it is hard to decide where to start. The steps listed below sequentially would allow us to complete this duty and more importantly; they would make us capable to manage our business with the power of measuring.

1. Process Definitions
As the first step, we should define our processes in detail. What are the processes and tasks we are dealing with? How we could complete them at the end of the day? What are the characteristics of these tasks?

To answer these questions, we should know and analyze our data in depth.
o Arrival analysis: The most imperative part of process definitions: How our tasks arrive? What is our job arrival pattern? Do our job arrivals fit with a statistical distribution?

o Processing times: Each process needs special requirements and differs in terms of processing times. To be able to plan our business, we should know processing times. Additionally, it would make our analyses more realistic to know the average, minimum, maximum and standard deviation of these times.

o Process statistics: Each process has its own characteristics in the banking system. For instance, a process could continue to the end state with probability p, or become cancelled with probability (1-p). Familiarizing this probabilities allow us to be more reasonable.
o Forecasting: We know our processes, processing times and statistics. It is all about our past data. However, in order to be prepared for future, we should be able to predict our future job arrivals. It is mainly important for planning and being ready for exceptional times.

2. Assignment
When we have our jobs at hand; as a next step, we have to assign them to the resources in order to get jobs done. There are two main methods for the assignment: Pull method and Push method. In the banking environment, Pull queues “simply distribute queued-up requests collectively to resources and rely on the individual agents to respond to the requests in order”. On the other hand, Push queues “distribute queued requests directly to each agent, so the agent has no choice but to respond to the next request that is served up personally to him or to her.” [1]. First, we think that system is run with the Pull method.

o Number of resources: We should know the number of resources that are able to complete the tasks.
o Competencies: Each task requires a specific competency to be completed and optionally, an additional competency too. For example, a specialist could complete a task which is in her core competencies in 3 minutes; but it would take 5 minutes to complete a task which is in her additional competencies.

3. Prioritization
The other assignment method is Push; and this method canalizes us to Queue Prioritization approach.
Queue Management: Each business, each department and even each process have a specific queue management method to be discovered. It is important to ken your business’ characteristics and analyze them with this point of view.

There are several queue management methods as FIFO (First in First Out) where first entry is processed first, or as EDF (Earliest Deadline First) where tasks line up according to their deadlines or as WPQ (Weighted Priority Queue) where each task has points and lines up with respect to their points. It is also possible to generate a specific algorithm for a business’ processes (or departments).

4. Resource and Norm Planning
Using all these analyses, next step is to determine number of resources optimally by taking into consideration resource presence ratios, locations and regulations to ensure reality of the system. This optimal number of resources will be able to assure specified KPIs.

5. KPI Specification

KPI specification is the other leg of the system optimization. Once we have our processes and assignment method defined and number of resources determined, we could optimize our KPIs like target SLAs or waiting times.

6. Branch Action Planning

When we are at this step, we know our processes and our resources’ competencies, our target KPIs and optimal number of resources. We know, for example, our Taksim Branch needs to have 3 resources at Front Office and 2 resources at Back Office to be able to complete all the tasks with target SLAs and waiting times. But we also know that our jobs do not arrive periodically.

For instance, our Front Office might be very busy on the Wednesday afternoons and our Back Office resources might be idle at that time. So, it will be wise to lead Back Office resources to the Front Office at the peak times and vice versa. Doing these plans periodically will certainly enable us to maintain required waiting times and target SLAs, together with managing seasonality of our job arrivals.

7. Simulation and Optimization
When we are at this step, it means that we take firm steps forward the top of the workforce and process management perfection curve. This step is for making strategic level decisions like: “What will be our SLAs if we decrease the number of resources by 3%, while our job arrivals remain the same?”, “Our business is about to grow 10%, how many resources should I hire to keep my SLAs the same?” and so on. Also, the interaction of our defined Prioritization method with the optimal number of resources might be subject to optimize.

Banking Process Management Diagram

When we complete all these steps, it means that we manage our workforce and processes perfectly and in detail. We are ready for exceptional times, and able to make our strategic decisions easily. In contrast to the opposing views about the relation of measuring and managing, we should never forget that “Scientia est potentia” (“Knowledge is power”) in the business world.


Product Marketing Manager