When I heard that the UK Government had lost its court battle over cancellation of the e-borders project and was ordered to pay the contractor, Raytheon £224 million, I was reminded of the observations of Carl von Clausewitz.
In von Clausewitz’s view, ‘Countless minor incidents – the kind you can never really foresee – combine to lower the general level of performance, so that one always falls short of the intended goal’. This observation helped to introduce the concept of ‘friction’ to describe the myriad factors and events that prevented plans being executed as they had been envisaged. Such a shame the Home Office didn’t have von Clausewitz available to advice on the e-borders project. He could have saved them a fortune.
I expect there were ‘countless minor incidents’, and maybe also some quite major ones, that conspired to drive the e-borders project off the rails, and von Clauewitz’s analysis give us an insight into the timeless nature of this problem.
With government projects, great store is always placed on ‘getting the specification right’, as if merely stating what is required in great detail should be sufficient to deliver it. Of course this never happens. And one might have expected those in the Government to know this – after all this is only the latest in a long series of failed projects.
It’s important that public sector and private businesses face up to the realities that accompany major projects spanning many years, and improve their management of them. To get them started, they might take a close look at how the size and cost of these projects is estimated.
At present, most Government contracts are awarded after a competitive tendering process. The suppliers, keen to win, will generally quote the lowest price they feel they can afford. The customer, in this case the Home Office, will generally feel duty bound to accept the lowest bid that looks like it can deliver the required outcome.
Where’s the flaw? The fact that neither party wants to take friction into account. The supplier wants to assume there won’t be any or, if there is, he can charge extra for it. The customer wants to assume there won’t be any, or else he can blame the supplier for it. And that’s exactly what happened on the e-borders project and why the Government ended up in court.
Since it is the customer’s responsibility to select the right solution, he cannot afford to be wearing ‘rose-tinted glasses’ when evaluating the various suppliers’ bids. He should make certain that friction is accounted for before letting the contract.
There are plenty of recent examples of failed Government, and business projects, that could be used to judge the effects of friction and to support the creation of accurate estimates, so why not use these experiences to inform future estimates? Of course, this may mean increasing the estimated project cost and duration by a factor of 2, 3 or even 4. But if this is a realistic estimate, isn’t that the one they should be using?
Assuming that everyone’s expectations as to timescale and cost are set too low at the outset, is it any surprise that the arguments start when friction inevitably causes the project to fall behind its unrealistic schedule? This may not be critical for a project that is scheduled to last 3 months and cost, say £200k.
But for the e-borders project, which had already been running for 3 years when it was cancelled, the amount of money wasted is high. According to James Brokenshire (no, really), the Minister for Immigration and Security, by its cancellation in 2010 “The contract, signed in 2007, had already cost the taxpayer £259.3 million and yet wasn’t delivering.”
Major projects need to be broken up into manageable parts and tackled using an iterative approach in which all stakeholders collaborate closely from start to finish. This means representatives that are responsible for achieving the outcome working in the same room as the analysts, developers and testers that are creating the code. All of the parties engaged in the project need to be together, work together and accept collective responsibility for delivering the required result.
The private sector has increasingly been moving towards this ‘Agile’ approach to project delivery, in which large projects are broken up into bite-sized chunks that can be delivered by teams of the customer’s and supplier’s people working closely together.
This collaborative approach has significantly reduced the risk associated with major work programmes and is fast becoming the standard methodology for private sector IT projects. Recently, the Government Digital Service has shown that the Agile approach to project delivery can work very well in creating new online services for taxpayers and rolling these out in weeks rather than years.
Such an approach may be difficult for the public sector at large to embrace, since it requires an acceptance that the situation will change and that priorities will have to shift in response to the ‘countless minor incidents’ that will inevitably occur and that cannot be foreseen. This lack of certainty at the outset will naturally worry procurement teams. But, as we have seen from the e-borders fiasco, there will always be a lack of certainty. Pretending that it doesn’t exist can never be a smart way of embarking on a large-scale, multi-year project.
It’s time to abandon an approach to the management of IT projects which is based on a frictionless delusion and replace it with a practical method that acknowledges the fact that all sorts of things will go wrong and no one can know in advance what they are or what effect they will have. As Carl von Clausewitz might have said, ‘Friction is going to happen anyway. Understanding and accepting that fact is the first step in overcoming it’.