This white paper discusses two-headed IT departments, their problems and offers a solution by eliminating one head.
By Bradley W. Lucas, February 2000
IT departments at large financial firms are two-headed monsters. They are bloated bureaucracies, often contributing more problems than solutions. Large numbers of financial professionals can’t ask for and receive anything of substance from an internal development group. They are at the mercy of development managers who complain of being short-staffed, yet build environments that are hostile for working software professionals. Dibert-isms are rampant in the financial world, and something must be done about it. Why? Just wait until customers of large financial companies realize how much money is wasted running 500+ person IT departments, with a >35% attrition rate and fewer than 15% of all initiated projects reaching completion.
But, how many heads are really needed on this IT dragon? Well, in reality only one.
The first head is responsible for the company’s systems: networking maintenance, systems’ monitoring and computer installations. Some companies have tried to outsource this function, but over time it has become apparent that on some level organizations require a department focused on the daily nuts and bolts of its own internal support requirements.
For the second head, the news is not so good. The second head exists to develop internal, custom software applications — theoretically to leverage an investment firm’s proprietary knowledge to customers in a manner that differentiates them from their competitors. Integration is a key component, as well as proprietary development in “the name of” competitive advantage.
The second head is completely failing. Leaders of large financial companies believe this head is a necessary evil, because they are understandably very protective of their technological advantages. Their aim is to control in-house development in order to retain proprietary knowledge and keep competitors at bay.
The trouble is IT departments rarely complete a project they start. The frequently break every best practice known in the software development field, including: Preventing access to end-users in the requirements gathering phase. Eliminating input from the scheduling process. Selecting technologies and platforms before designing the system. Adding developers in increasing numbers as the project deadlines slip, falsely hoping that sheer numbers will solve the problems. Repeatedly, creating environments conducive to “Death Marches.” Visibly dividing “teams,” distinguishing employees versus contractors thereby creating cultural chasms. Ignoring user satisfaction as a basis for project success. Selecting immature technologies more for resume building than ultimate project success. Building inadequate environments for developer satisfaction. The disaster is further compounded when financial firms unabashedly market this supposed ability to customers as something that differentiates them from their competitors. Typically, large IT departments are run by mangers with experience working in large corporate environments. Unfortunately, their past experience causes them to see all the problems as a normal cost of doing business. To a certain extent, it is this mindset that causes very little change. As contracting body shops continue to spring up in increasing numbers, it is clear the attempt at solving these problems stems from an old-world management viewpoint: buy more people and try to muscle your way from A to B. At some point, business leaders will question this process. They will look around the Internet and see great things being developed in short periods of time with very few people — while their 500+ IT departments are doing what? One avenue is to reorganize these organizations. This is possible but impractical. Large organizations are slow to change, much like oil tankers that take miles to turn around. Real change will come from outside the organization. Business users in search of a new custom application will look outside the organization to places that can demonstrate the ability to elegantly build quality applications in months rather than years — applications that do what they want and work reliably. “Looking outward” can be as fraught with danger as not looking at all. One must make sure an outside vendor knows the financial industry. Today’s “hot” business development idea is to start a Web integration firm. Read up on Web technologies and go into companies pitching the latest technologies to webify business processes. Typically, these businesses are growing as fast a possible with the goal of going public and doing so by hiring a lot of people — recent college grads with very little experience. This is fine for building web pages, but disastrous for financial firms with sensitive data needs, high performance requirements and security concerns. “Looking outward” also requires doing research. How can you be certain a firm can do what you need? There is no rating system or government accreditation. Many who look outward suffer from Titanic syndrome: they mistakenly say to themselves that size matters. They find the biggest consulting company and assume that is the best path. They quickly find themselves impressed with the founders’ and senior partners’ experience, but end up actually working with a room full of recent college grads billing out at an exorbitant rate. Such situations are a recipe for disaster and a precursor to giving consulting companies a bad name. What financial services companies need are firms that, first and foremost, focus on the financial services industry. Secondly, they need firms that are staffed with experienced people, people with more than 10 years’ development experience. They need companies focused on building solutions and relationships, not IPOs. Growth should come only as a measure of success based on true client satisfaction and properly engineered software solutions. Such firms do exist. You just have to find them. The searching is worth the effort since these firms can build reliable software in reasonable timeframes, and can do so at a level of development that is sensitive to ever-changing needs of the financial world. For information on such a company, visit http://www.BeaconHillSoftware.com. Bradley W. Lucas is President of Beacon Hill Software Boston-based financial software services company dedicated to slaying the IT dragon. Copyright © 1996-2001 Beacon Hill Software, Inc. All rights reserved.