Secret team sinks app dev project

Why politics and software don't mix

I was born in Cameroon, and if there’s one thing I learned during my 16 years of doing IT in Africa, it’s that politics and technical work are mutually incompatible. I had the “honor” of designing and deploying a number of large banking systems, and with one exception, all those projects were successful. But what an exception that one turned out to be!

I was chief MIS officer for a bank that was upgrading from a very old banking system to a new one. In Cameroon, financial posting and processing rules are based on the French banking system, which is very different from the American model -- so I had already done lots of customization on the original software, which had been designed for the North American market and needed to be adapted for central Africa. The new project was fully funded, with a green light from management.

We hired a team from a Canadian company with a good banking application to help manage deployment. I was overseeing the project for the bank, writing a lot of Visual Basic, as well as specialized financial code for the banking system.

Unfortunately, in the middle of the process, a large German firm purchased the Canadian software company. As the Canadian team lost focus on our project, it slowed to a crawl.

We badly needed the bank’s general manager to meet with the new German owners to convey the urgency of the situation, but despite our pleas, nothing ever seemed to happen. During meetings, the general manager would even promise to set up a conference with the Germans, and once he went so far as to direct me to work up a list of topics for that imaginary meeting. Sadly, that was as far as things ever went.

After several informal chats with key figures close to the general manager, I began to grow suspicious that something dubious was going on. I decided to morph myself from an engineer into a private eye. That’s when I discovered that a secret, underground team at the bank was working with a local developer to build another banking system from scratch and sell it to the bank. Worse yet, I learned that the general manager was a key member of this underground team, with major shares in the project. The horror, the horror!

I knew the project was doomed. We were not technicians any more; we were politicians -- and that’s a skill set I never mastered very well. So I went out and found another job.

Meanwhile, as I had predicted, the project failed. And, despite all the efforts of the underground team, the secret project never made it to the winner’s circle either. The bank continued using the original software, but after the general manager and his pals were discharged, new software was finally installed. As I had already left the bank, I came in as a consultant to manage the migration from the original package to the new one.

The lesson I took from all this was that for your project to be successful, you need to make sure that management is at least 90 percent behind you -- and don’t rely on official reports or meetings, either. I’m guessing that this rule applies to the rest of the world as well, not only to Africa.