May 16, 2017
In my working life in IT, I have encountered a number of organisations that wanted to move from a working IT system to a different one, in the belief that it would be better and cheaper. I’ll be describing a recent example in this piece. Although there is not on the surface any obvious connection with the Brexit campaign, there are some interesting parallels in the behaviour of the people involved.
I’ve chosen the example because I had the opportunity to study it in detail. The organisation concerned is large and has a particular IT system handing around 50 applications, of which some are very critical; they must work. A complicating factor is that a number of requirement changes have to be included each year. I won’t say what the organisation does nor where it is located, except to note that it’s not in the UK.
At the time of my involvement, the applications were running in a very stable platform, which provided the high level of security required by the nature of their work. (For those not familiar with IT, I use the term ‘platform’ to mean the computer hardware and all the software, apart from the applications.) The supporting personnel looking after the platform and the applications were very capable, and implemented the regular changes on time. Everything worked well.
A number of senior people in the organisation were convinced that it would be better, financially and otherwise, if they moved the applications to a different platform. To further justify the move, they falsely asserted that the current platform was due to be discontinued by its supplier. I was asked to work with a colleague to investigate the economics of the proposed move. This involved three activities: calculating the annual total cost of ownership (TCO) of the present system; the TCO of the new system; and the cost of moving the applications from one platform to another.
The first activity was fairly easy. Because the system existed, all the data required to calculate the TCO – primarily the annual cost of hardware, software and people – were readily available.
The second activity was obviously more difficult as of course the system did not yet exist. We approached the problem by deriving a configuration capable of meeting all the requirements, based on general industry data. From this, we could then calculate the likely TCO of the new system, once it had started stable operation.
That left us with the cost of the move, which we calculated using recognised information for projects of this size and in the same business sector. The project was very large; the number of lines of application code ran into the millions. We derived optimistic and pessimistic estimates of the effort required and the likely cost.
The results were interesting, and rather as we expected. We found that there would be a potential TCO saving of about 10%, but the cost of the project to accomplish the move would result in a return on investment of the order of 20 years, a lifetime in IT. We also warned that the risk involved in IT projects increases with size. As this would be a very large project, the chances of missing out on time, cost and performance were substantial. We presented our findings, along with warnings about risk, to the organisation’s management.
What has this got to do with a venture such as Brexit? Both represent a significant change from a known to an unknown state. In each case, and other similar examples of change, the driving force comes from a number of people who obsess about the subject. Wishful thinking dominates their viewpoint as they construct a vision of a future of consummate excellence. Any evidence to the contrary is either ignored or dismissed as wrong. The costs of making the change are grossly underestimated. Half-truths and lies are used to help bolster the case for going ahead.
Brexit of course is still a work in progress. The UK only triggered Article 50 at the end of March, and is, as I write, in the throes of a general election campaign. The belief in the current and probable future UK governments is that a deal with the EU that is ‘best for Britain’ can be worked out, and trade agreements with other countries will be easily arranged. In the words of Boris Johnson, ‘we can have our cake and eat it’, although all the evidence from early discussions suggest that this is an unlikely outcome. We will see what happens.
What happened to the IT project I described earlier? The organisation’s management ignored our advice and went ahead with the proposed move of the applications. After spending about five times our most optimistic cost estimate and three times our more pessimistic figure, the project was abandoned after about two years, leaving nothing to show for it. The applications now happily run in a new level of the original platform.
Notes and sources
 More information is available about public sector projects. The private sector manages to conceal problems more effectively. One freely-available source is: Overspend? Late? Failure? What the Data Say About IT Project Risk in the Public-Sector. Budzier, A and Flyvbjerg, B, 2012, University of Oxford, Saïd Business School: http://eureka.sbs.ox.ac.uk/4952/1/1304.4525OL.pdf