Analysis The thankless nature of IT service provision is often demonstrated by the fact users only tend to notice it when it goes wrong: if there's no screaming, then it's working fine.
In that respect, at least, HM Revenue and Customs' £10bn IT contract with Capgemini and Fujitsu has been a success. Particularly if one considers the users in this scenario are every tax-paying adult in the UK.
Compared with the litany of IT disasters across government over the last ten years it has been a relatively smooth project, a fact acknowledged by the National Audit Office last year. The NAO said the contract "has delivered certainty and continuity over the past decade."
However, the contract's enormous cost has been described by MPs as one of the biggest civil IT projects in Europe. HMRC has rightly come under fire for its failure to negotiate a better deal with the suppliers.
The total cost of the contract reached £7.9bn in March 2014, which HMRC admitted was expensive. From that, Capgemini and Fujitsu had secured a combined profit of £1.2bn – a profit margin of 15.8 per cent. Unsurprisingly, the deal soaks up 84 per cent of HMRC's IT budget.
Now that contract – first let in 2004 – is finally being broken up, with a tender out next month for the first suppliers of its "phased transition." HMRC has emphasised its eagerness to work with small business, and wants to spread the love across a 400-strong supplier base.
But as that highly risky transition begins, many will wonder if the £10bn price tag for a decade's worth of guaranteed £500bn annual tax revenue was an expensive price worth paying.
Several insiders have already been in touch with The Register to share their concerns about HMRC's plan.
Key to its success is HMRC's strategy to bring more skills in-house. As such, around 300 Capgemini and Fujitsu staff are being transferred to a privately-owned but HMRC-run limited company. That structure enables the government to pay more competitive salaries than if staff were directly employed as civil servants.
However, one insider told The Register a significant amount of talented staff working on Aspire have already opted out of the transfer agreement. Instead they will keep their jobs with Capgemini and Fujitsu where they will continue to receive bonuses – and in some cases better pension terms.
That raises a serious question as to whether HMRC will be left with enough capabilities to pull off the transition. Lack of skills has been named as one of the biggest blockers to the government realising its "digital transformation" ambitions.
Secondly, there is the risk of adding greater complexity by managing many more suppliers. One source noted: "After 10 years working with Capgemini and others, it's actually turned into a good relationship and IT is pretty stable."
Sensibly, HMRC appears to be mitigating some of the disaggregation risks by retaining the incumbent suppliers on a number of key systems. It is thought Fujitsu is actually keeping the majority of its Aspire contract, although the bigger supplier Capgemini will keep much less.
Difficult, legacy core systems will stick around for a while under various arrangements, added one source.
400 throats to choke
Smaller contracts being let include supporting IT, such as desktops. Although there is some concern that some of the core admin systems could be farmed out to the public cloud without proper security consideration. And there will certainly be time and cost implications of managing lots of small suppliers involved in such a key department – as opposed to the old “one throat to choke” model.
Another issue is the extent to how much appetite HMRC has for the new arrangement. "HMRC doesn't necessarily think it's a good idea. They're doing it because they were told to by the Cabinet Office," said a source.
As it stands, the Aspire contract conflicts with current government policy on how departments should buy technology. The Cabinet Office has said long-term contracts with a prime supplier do not deliver optimal levels of innovation, value for money or pace of change. As such, no single contract within the programme will be worth more than £100m.
The sentiment behind disaggregation is perfectly reasonable; after all, the government has paid through the nose for crap IT for too long.
But as the disaster of the Cabinet Office's involvement with the Rural Payments Agency's Common Agricultural IT system proves, having full buy-in for projects is essential for success. Hopefully there will be a much clearer line of responsibility and accountability in this project.
On the other hand, if the contract transition works, it will be a useful model for the Cabinet Office to showcase how IT can be done in the post "oligopoly" era.
Colossal disruption costs
Finally there is the question as to what extent the new form of the project will really save money. The business case for the transition is to shave £200m from the contract per year by 2020/21. HMRC's chief digital and information officer Mark Dearnley had, however, told MPs in March last year those cost savings would kick in by financial year 2019/20.
According to Dearnley the total transition costs will be £600m over the next five years, with £100m of transition cash earmarked for this year. He said the project had been flagged as "amber-red" by the Major Projects Authority – meaning it was high risk, a recognition of its importance and complexity.
One source said: "I'd be surprised if it ends up less expensive overall, if you take into account the colossal disruption costs."
History tells us that project costs always tend to be underestimated.
HMRC appears cogniscant of all the risk factors involved. Dearnley told MPs: “By being phased we are not talking about a big bang, we’re talking a lot of little bangs: that hopefully don’t go bang.”
No doubt if they do go "bang", the incumbent suppliers will be waiting in the wings to step in... for a fee. ®