How do we fix the UK government’s IT woes?

The NHS seems to think that all us journalists ever want to do is knock it. That’s not true. When I was treated in Poole Hospital for an arm fracture the nurses asked me what I did for a living, so I told them and they immediately recoiled. I had to explain that I didn’t work for the News of the World and wasn’t interested in smut, jibes or pointless conjecture. They eventually gave me my morphine.

That said, there may be a more constructive route to critiquing the operations of public entities such as the NHS – and I am coming from an IT perspective of course.

One might argue that the large proportion of public sector projects these days are over-sized lumbering ‘elephant’ projects. It’s almost kind of "Let's fix the NHS" or "Let's do UK security" isn’t it? Sound familiar to you yet? You can imagine those words being delivered with a Blair or Cameron smile surely? Although I think a Brown grimace would be pushing it.

The problem with this approach, certainly from an IT systems and architectural viewpoint is that this leads to a bunch of misguided attempts to chip away at a problem that is far too large to steer and direct. If new solutions are developed, by the time they reach deployment they have little effect as the weight of the corporate ship is already on course for a new bearing.

Some argue that what is needed is an overall umbrella project ("Let's fix the NHS") and underneath this should be a set of small, interrelated projects, each of which builds on what has gone before.

I spoke to Clive Longbottom who is service director for business process analysis at analyst house Quocirca Ltd for his erudite opinion on this subject. “Take the NHS. If connectivity between the trusts had been looked at first, we could still have gone for disparate systems, which could exchange information via XML. There would be no need to try and come up with a one-size-fits-all ‘solution’ that suits nobody. We could then have looked at the XML templates that would be required to exchange information on bed availability, on consultant skills and availability, on waiting list times, on GPs moving records from themselves to consultants, hospitals and other specialists. Each of these could be addressed as separate issues, and as separate projects. No one project would be completely dependent on the overall progress of the ‘master plan’ - and so we'd be much further down towards where we need to be.”

That’s it then, it’s like the concept of how do you build a huge wall using Zen isn’t it? It’s a huge task, but first you take one brick, then you take one more brick – and so on. Or maybe I am mixing my mantras?

“There should be no single IT project that costs over £10m. As soon as it gets to that level, it should be broken down and carried out as multiple different projects. This maintains focus, concentrates the mind and makes project management per small project far easier. At the master plan level, more project management skills would then need to be in place to stop project creep and further stop those nice people from HP/EDS, Capita, Fujitsu, IBM, Accenture, BT, CSC, Logica and so on from trying to pull fast ones...,” added Quocirca’s Longbottom.

So just for the record, like the majority of you I’m sure – I am a big fan of the NHS. In fact, I am only up early to write this as I head off down to my local surgery for a blood pressure check. Let’s hope my doctor appreciates macro-level IT project management as much as some of us right?