In December, I omitted to post my Times column on government IT and digital policy:
The travel chaos last Friday was a reminder of just how much life depends on Big Software doing its job. The air-traffic control centre at Swanwick was six years late and hundreds of millions over budget when it opened in 2002 in shiny new offices, but with software still based on an upgraded, old system. Unnoticed and unsung, however, this government may actually have found a way to bring the horrid history of big, public IT projects to an end.
Remember the chaotic launch of Obamacare in August, with only one in five users able to access the healthcare.gov website. Or the NHS patient record system, abandoned last year after costing the taxpayer nearly £10 billion. Likewise, the BBC wrote off £100 million last year after five years of failing to make its “digital media initiative” work. It’s equally bad in many big companies: McKinsey found in 2012 that 17 per cent of IT projects budgeted at more than $15 million fail so badly they threaten the company’s very existence.
There is something uniquely troublesome about big IT projects. So it may surprise you to hear that I think a genuine success story of this government is that it has finally learnt how to prevent such problems and design new systems so that they work. In essence, it has begun to adopt the principles of evolution, rather than creationism.
Largely unheralded, the government digital service is one of the current administration’s success stories. Tirelessly championed by Francis Maude, as minister for the Cabinet Office, egged on by Baroness Lane-Fox, and run by Mike Bracken, brought in from outside, the GDS is not just trying to make government services online as easy as shopping at Amazon or booking an airline ticket. It is also reshaping the way the public sector does big IT projects to make sure cost and time overruns are history.
What Mr Bracken calls the “waterfall” approach to such big projects in the past consisted of “writing most when you know least”. The people in charge wrote enormous documents to try to specify the comprehensive requirements of the end users, did not change them as technology changed and issued vast, long and lucrative contracts to big companies.
Instead, Mr Maude and Mr Bracken are teaching the civil service to start small, fail fast, get feedback from users early and evolve the thing as you go along. So those designing an online service begin with a discovery phase, lasting six to 12 weeks, then build an “alpha” prototype of a working software service in less than three months, followed by commissioning a private “beta”, to be used by a private audience of specialist users. Only once rigorously tested is this opened up to the public, sometimes in a controlled way. And only later is the old service turned off.
This is exactly the sort of recipe for success championed by the economist Tim Harford in his book Adapt. Harford pointed out that whether pacifying Iraq, designing an aircraft or writing a Broadway musical, those who succeed allow for plenty of low-cost trial and error and incremental change. It’s the mechanism Charles Darwin discovered that Mother Nature uses. Rather than a grand “creationist” plan or a big leap, natural selection incrementally discovers success through trial and failure. From the English language to an airliner, everything successful has emerged by small steps.
The successful IT systems we all use, from Facebook to BBC News, were all built this way. Yet government kept trying to do things by grand plan. The history of information technology explains how we went wrong. In the beginning all things related to the web, in public and private sectors, became the property of the high priests in the IT department, as the only people who understood the technology. They thought mainly of the needs of producers of content, rather than users. Much of the private sector wrestled digital content out of the hands of the IT department long ago, but in much of Whitehall that’s where it still lay until recently.
Mr Bracken and his lieutenants have turned Whitehall upside down, collapsing the profession of chief information officer (head of IT) altogether. These CIOs had often been recruited from big consultancy firms, and they were in the habit of outsourcing big projects to those firms. The plan was to outsource the risk, but it never happened. Indeed, in many instances the big IT consultancies royally ripped off the taxpayer by designing big systems in the waterfall fashion so that they overran their budgets and timetables.
Mr Maude began by centralising controls so that he had to sign off any IT contract of more than £1 million (now raised to £5 million), then built up an in-house capability to offer cheaper and better design, and opened procurement to smaller companies. Government contracts with outside IT suppliers are now shorter and smaller. Some of the savings on offer were so vast that civil servants refused to believe them. In one case, 98.5 per cent of the cost of an existing contract was saved by letting a contract to a small British business rather than an incumbent multinational IT firm, and it worked better. With much of the system still to tackle, Mr Maude reckons he will have saved £4 billion a year by 2019-20.
The big question being asked in Whitehall right now is whether Universal Credit will prove to have been the last of the old or the first of the new. It began as an old-fashioned waterfall project, but is being reformulated in parallel in the incremental and digital fashion.
It is not just me who is starstruck by what Mr Maude and Mr Bracken are doing. Other governments are noticing. Last week the government hosted an international summit called the D5, an exclusive club with Estonia, Israel, South Korea and New Zealand in it. Along with Britain, these countries see themselves as by far the most advanced in digitising government services in a way that makes them easy for users to use. The United States government is deliberately emulating Britain’s approach.
The rapid roll-out of the new government web platform, gov.uk, now used by 300 government organisations, won the prestigious Design of the Year award last year, beating the Shard. Mr Bracken relentlessly demands simplicity from government websites, reversing a 15-year build-up of useless information, banishing 40 per cent of the content to the National Archives, purifying forms, closing 1,800 separate government websites and saving 60 per cent of costs. One-click government at last.