It is that time of year again, when organizations are putting the final touches on their project funding plans for tomorrow. Over the past several months, a strange tale has played out across the globe, a farcical tale that if you tried to tell anyone outside IT about they would scarcely believe you. It plays out like this:
Someone in the business gets something resembling an idea - not something fully formed, but at least an inkling that suggests that there might be some benefit achieved in developing something. A few sentences are written about it and here is where the silliness begins. All these little fragments of ideas are gathered together and then IT is asked to estimate the cost and schedule - based on little more than the few sentences of description. If they are lucky they might get to ask some questions.
Everyone knows this is ludicrous, they will say, but they say they have no choice. They put all manner of disclaimers on the estimates but these are forgotten over time. What is recorded and remembered is a time and dollar cost estimate for each project - based on, remember, a few sentences of description. This sounds so absurd that I know you would be laughing if you did not see it every year in your own organization. But wait, it gets worse.
The worse part is that real funding decisions are made based on comparisons of costs and benefits for the projects. The costs we have just heard about. The benefits are often even more tenuously connected to reality; in some cases they are simply fabricated, and even in the best case they tend to be wildly unrealistic. One client with whom we worked observed that when all the claimed benefits for proposed IT project were summed they exceeded the annual revenues of the company - an obvious impossibility.
There is an unfairness about all this that I come to next: while the business is not held accountable for wildly optimistic estimates of benefits, it is normal for the IT project team to be held accountable for performance against an estimate that they were uninvolved in producing, and which was simply an educated (and sometimes not so educated) guess. The project budget, scope and schedule are often set by this almost literally insane planning process. No wonder it is often claimed that IT projects fail most of the time: they are doomed to fail from the start!
There are some additional underlying flaws in this sort of planning process: investment opportunities do not come along once a year, and they are not uniformly distributed across years. Some years may require greater investment and others may require less, depending on other strategic goals.
Software is a long-term investment that needs to be tied to the strategic planning of the business. No business today can run without software, and in order to make good business decisions the cost and timing of software projects needs to be considered at the time business decisions are being made, not as part of a disconnected annual process in which an arbitrary pool of IT funds are allocated. More on an alternative approach to funding in my next post.