Case Study: Investment Banking

Reducing Complexity of Investment Banking Operations; A Case Study

by Roberto De Micheli, Rule Financial

1. Background

The economical and socio-political environment of recent years has put a downward pressure on banks’ IT investment while still coping with rapidly changing regulatory and de-risking requirements. The lack of strategic investment and the increase of tactical solutions have resulted in complex IT landscapes which are costly to maintain and operate. Our clients, the business and IT heads of global market operations of a Tier-1 investment bank (“the bank”), need to achieve a transformational change that enables the business to operate on a significantly reduced and sustainably cost-effective base.

The strategic goals of the bank can be summarized as follows:

  • reduce costs, including operational and personnel
  • increase efficiency
  • reduce the number of IT systems while streamline and simplify those that are left
  • introduce globalised operational and IT models

2. The Challenge

Financial and regulatory pressures have left banks, particularly the larger ones, with an almost frozen and potentially dated systems landscape. This situation has not been helped by the fact that many banks have been expanding through acquisitions leaving them with a large and complex portfolio of systems applications with significant amounts of functional overlapping, data quality issues and cumbersome integration. This is compounded by the historic tendency to buy or build best of breed solutions for each banking business line and then further stretched by the perceived need to deploy system instances across multiple regions and/or countries. 
Case in point the “as-is” application portfolio of the client: in the image below every box represents an IT application. The applications are grouped by business function and colour-coded by fate (e.g. red = to be decommissioned)

The impacts of this complexity spill into the Operations world which must navigate a maze of applications with different UIs and different ways of representing and handling the same data.

3. The Engagement

This assignment required the delivery of a new target operating model, an implementation roadmap, and a set of business cases to enable the bank to achieve a transformational vision for its global markets operations function. The scope of the assignment included Global Operations for Securities, Derivatives, FX and MM. It was a Change Preparation engagement; at the time of writing, the bank is deciding about budget allocation for 2014. The analysis was performed by a team of six consultants, two of which were part-time. The team included architects, business analysts and business SMEs (Subject Matter Experts). The overall assignment duration was 4.5 months during H1 2013.

4. Our Approach

We initially formalized the drivers, objectives and goals of the client. This helped the bank articulate why they needed to engage in this initiative, how they were going to achieve the intended practical results and how they were going to measure success. The success measure was particularly important because many of these initiatives do not have any well-defined success points and thus the business never is able to determine what, if any, return it received from its investment.

To define success, we used an approach based on the same principles of Roger Sessions’ Snowman Practice.The philosophy of The Snowman Practice is that the ultimate goal of any transformation is the simplification of the IT landscape. The “Snowman” description relates to the end design of an architecture guided by this practice, a collection of business functions implemented by a group of services supported by a clump of data with strong vertical partitioning separating individual snowmen.
The Snowman Practice includes metrics for measuring complexity, a methodology for minimizing complexity, and tools for guiding the overall process. Unfortunately, these tools were not available at the start of our engagement so we created our own Excel based tools. We are in the process of evaluating integrating our tools with the Snowman Toolset.

Having agreed upon the overall goals (simplification) and a metric for judging success (an important one being the Snowman complexity metric) we then studied the bank’s functional model and classified every function in terms of current and target level of automation. This gave us the information we needed to define a global Business Technical Architecture that provides the technical services needed to support the target functional model. We studied the bank’s current operational model and mapped all the pain points to the functional model. 

Using The Snowman metrics, we could calculate the as-is and to-be complexity. The Snowman metric calculates complexity in Standard Complexity Units (SCUs).

We considered five possible approaches to replacing the existing systems. After a workshop with senior stakeholders, a clear favourite emerged, the use of a particular vendor providing the bulk of the required services. This vendor provided increased automation, simpler exception handling, and better optimization for operating in a global environment.

But we still needed to answer the critical question: how would this impact the overall IT complexity? The Snowman metric gave us the answer. The existing landscape had 5.9M SCUs of complexity. The to-be approach would reduce this to 260K SCUs, an overall complexity reduction of better than 95%. The Snowman Practice predicts that this will translate to a reduction of 95% of the cost of running the IT operation not including the cost of acquiring the vendor systems.

We pulled all the data together into our Roadmap tool. This tool allowed us to schedule all the projects over 3+ years based on dependencies and resources availability and to plot the costs and benefits. Thanks to the tool we were able to model different scenarios by playing with project scheduling and resourcing and by changing the benefit assumptions. We settled on grouping the projects into 10 groups organized in 3 phases and using the Snowman metric calculated the overall complexity reduction of each.

The following table shows the SCU complexity scores for the application architecture at the recognized stages, which proves how the proposed roadmap reduces the complexity of the IT infrastructure fairly dramatically.

Note: due to the nature of the engagement, we did not have the bandwidth to do a complete application-to-service mapping. The predicted complexity has been calculated assuming that each application provides all the services associated to all components the application is mapped to (discounting component-to-component interaction complexity).

5. The Outcome

The project has been considered a success by our client. At the end of the engagement we left the client with a new target IT architecture taking the bank from 264 processing systems to 17; an assessment of operational effectiveness across each of the business product lines in scope; an illustrative implementation roadmap of projects to achieve the target architecture; and a general approach for measuring the overall effectiveness of any strategy with respect to its ability to reduce operating complexity.

Smaller, simpler, and more effective. A good goal for any transformation.

For more information on Rule Financial, see our web site www.rulefinancial.com or contact the author at roberto.demicheli@rulefinancial.com

For more information on The Snowman Practice, see Roger Sessions’s web site www.objectwatch.com.