So my former employer is now a year behind schedule on their shared services project, and a cool $250 million overbudget. Obviously they were right, they didn't need my help. Not that it would have made any difference, but I did try to explain to them that managing projects - at the highest level - was a matter of managing interests, not schedules or resources. Of course it turned out that it was in the interests of just about everyone who had executive level control over the project to make it go on for as long as, and cost as much as, possible. My prediction is that they'll give up trying to centralise all Government processing within the next twelve months, hand out the software (in whatever state it happens to be at the time) to the Agencies who have watched another 'fond hope' rise and fall, and call it a rip-roaring success. After a year or so the Agencies will ditch the software when they find the cost of having it modified is 'too high' compared to other 'off the shelf ' products and we will have gone full circle. Was there, is there another way? Of course, but nobody is paying me for opinions these days so they can go - and I think this is the expression - 'whistle Dixie'. Am I being just a little bitter? Heck yes, but I feel better now. Well enough in any case to make this observation:
The answer now lies in language, not 'product', and certainly not performance. So where the Government proposed to 'roll in' agencies into this central processing unit, they should now start talking about 'rolling out' the technology (and the 'capability') to the agencies. A slight change in nuance, and if you look at the post in this blog about 'anticipating organisational death', you'll see it's really just a matter of 'bringing the schedule forward'. It's not too late to 'take control' of the train wreck, by steering it 'home' (back to the Agencies where Agency IT staff at least have a chance of knocking it into shape) rather than allowing it to run out of steam somewhere in the wilderness. In fact there's a certain poetic beauty in the concept of giving the unwanted child back to the Agencies and saying, 'you look after it now'. And of course if it fails it's then due to Agency neglect, not failure of 'concept'.
The beauty of having an 'Organisational Death Plan' (in this case to centralise the systems and the 'best practice', then refine it, then hand it back to the Agencies as a fully automated system that line and production managers could use directly (with fewer HR and Finance people needed to 'translate' and act as intermediaries)) is that it would have made it possible to make the current situation (and my 'rolling-out' concept) 'look' like it was anticipated, and that we were in fact 'on track' (indeed ahead of schedule!) - and that we had some notion of what that track was and where it was heading. And of course it then becomes more than 'just' language, but in fact a 'workable plan', because the consumers of the language are not just the anxious public and politicians, but the people who have to implement and use the systems.
A little additional nicety is that the 'rolling out' will of course be accompanied by the return of some funding to the Agencies (that was taken from them in order to fund the Central Agency). But of course the money 'handed back' will be less than the money taken away, because the system returned to the Agency will now incorporate 'improvements and streamlining' that didn't exist in their 'old' separate Agency systems. A saving! For the first time in the history of the project. And of course the Agencies will be charged with making 'further refinements' to the system, but no additional resource will be provided for that. And if the Government wants to be particularly cunning, it would make each Agency responsible for some particular aspect of the refinement, so that all of the other Agencies (waiting for that particular refinement) would provide the pressure on the sponsoring Agency 'to get the job done'. Heck, real motivation - and that might again be a 'first' again for the project.
Ah, but no mention of roll-back, that would be too unfortunate.
Comments