I recently had an interesting discussion with a friend of mine who happens to be the CIO of a major UK bank.

Working for a bank isn’t much fun at the moment, especially with the threat of eurozone collapse hanging over your head. This CIO, however, has a very specific set of problems which have little to do with Greece.

First off, his team is getting bogged down in a rapidly growing pile of data, secondly he has regulators demanding access to everything immediately, and thirdly he has what he calls the ‘complexity issue’ to deal with.

The problem stems from the fact that there is an ever-increasing number of new applications being released into the business while very few of the legacy applications are being retired.

This leads to added complexity, added costs and reduced IT flexibility. The way in which complex legacy systems have been dealt with has been simply to outsource them. However outsourcing in this case simply means handing the problem to someone else rather than solving it.

In fact outsourcers like complexity — there is money in it — and so they certainly aren’t minded to help you simplify matters.

The complexity issue is based on years of adding without ever taking away. It’s a bit like painting an old house.

The right thing to do is to strip away all the old paint, sand it down and repaint to give a fresh smooth coat but laziness means you just slap a new one on over the old. Eventually the cracks will start to show.

The problem is increased because at the sharp end of the business the demands are for greater flexibility, more agility and lower costs.

Driven by demand for a real-time response to customers, regulators and the market the bank needs to become faster and more flexible.

Although virtualisation and cloud technology can help to provide this, unless the legacy applications and IT complexity are dealt with first this is essentially putting lipstick on a pig.

This friend of mine isn’t stupid. He knows that he is storing up problems for the future but the task is so vast, so complicated, potentially expensive and frankly, so boring, that he can’t bring himself to tackle it and currently there is no-one for him to turn to for help.

Systems integrators struggle to get to grips with the size and complexity which results in snowballing costs and missed promises, while management consultants will focus on lengthy benchmarking and ‘as is’ studies which result in adding rather than taking away.

Whis is a great example of why IT departments should be run like a business. With internal P&L the savings made by retiring old and underutilised systems will pay for the investment in all the latest tools and applications that will help the bank become more aligned to customer needs — mobility, speed, 24/7 near-real-time access to data and so on.

The more you save, the healthier the bottom line… you could even reward bonuses based on savings realised.

Mushrooming complexity is a major pain point for companies in the financial sector because of internal and external governance pressures, but any organisation with a history will be living with a significant amount of redundancy within its IT architecture.

This means that the CIO needs to focus his architects on simplification and legacy removal as well as innovation and the ‘next big thing’.

It starts with assessing the level of complexity within the current architecture, applications, data and process flows. You then need to work out your framework for measuring success.

Thirdly, you need to think about incentivisation — not just within the IT department but also with the business users: understanding what they are using and how they use data is a key part of the process, so you need to engage with your internal customers.

So what do you do with the systems you identify as ripe for retirement? Just because a system isn’t being used doesn’t make it useless.

I am all for recycling especially because it means shelling out less on the new. There may even be a market for second-hand systems — a kind of Antiques Roadshow meets eBay meets second-hand car dealer for technology. When you consider the cost of this stuff that isn’t such a bad idea.

The fact is that in IT everybody looks forward. How many projects have been started and then stopped because the focus moved onto the next big thing?

How much old stuff is out there to be put out of its misery or dug up and revealed to be worth its weight in gold? Time for a bit of a clean-up?

Mike Altendorf is founder and former joint MD of web services company Conchango