8 lessons from digital megaprojects


Project: Oyster card
Lesson: Roll out slowly and keep testing

oyster cardThe Oyster card is arguably the most successful political initiative in a generation. It revolutionised travel in London for the better. It meant commuters could board buses and Tube trains with a touch of a card. Queuing for tickets became redundant. And it gave Transport for London a detailed record of passenger routes, allowing scheduling to be refined. The TranSys consortium behind the project was made up of Fujitsu, Cubic, EDS and WS Atkins. The strategy switched early on from a big bang implementation to a slow rollout to give time to iron out glitches. A testing facility ran London Underground through mock-up stations. “The system was given a very thorough thrashing before it was let loose on the public,” recalls Pat Morey, project manager for TranSys. It worked brilliantly. The approach deserves to be studied and copied by anyone attempting infrastructure projects on a similar scale.

Project: Halving the admin budget at BIS
Lesson: Big goals trigger high performance

halving the admin budgetIn the drive for austerity after the 2010 general election, the Department for Business, Innovation and Skills (BIS) was whacked harder than any other. The administration budget was to be halved over the course of the parliament. Yet BIS managed it, with no noticeable reduction in function. How? The big goal had project managers thinking creatively. Agencies merged, such as the Office of Fair Trading and the Competition Commission, taking the number from 70 to fewer than 50. A new Enterprise Performance Management IT system gave managers visibility on efficiency. Finance functions were automated. Not for nothing did BIS win the National Audit Office’s Excellence in Reporting award in October. Director general of finance Howard Orme kept staff onside with honest messages about the changes. “My approach is to be upfront,” he says. “Get people involved in the change, get them to plan it – even those areas closing down.”

Project: Crown Commercial Service launch
Lesson: Train staff

crown commercial serviceProject managers in the public sector had a common moan prior to 2011 – the contracts they were executing were often drafted appallingly. They would lack basic key performance indicators, the procurement deals were ruinous and some projects would be 20 years long, with no exit plan if third parties failed to deliver. So in 2011 chief procurement officer Bill Crothers with Francis Maude at the Cabinet Office founded the Crown Commercial Service (CCS). Procurement officers would be trained in the new Commissioning Academy. The new training stressed that contract duration should be cut to five to seven years. Projects should be “open book” so managers could see profit levels and other metrics. In 2014 the CCS vetted £15.1 billion of IT spend, saving the DVLA £57 million on a package of contracts. Overall the initiative claims to have saved £5.9 billion in 2014-15.

Project: 2012 Olympic Games streaming
Lesson: Sometimes you’ve got to shine

olympic games streamingThe Olympics is more than a festival of sport. It’s a chance for the host nation to showcase everything it is good at. And when London hosted the 2012 Games, the BBC wanted to prove it was still the world’s best broadcaster. It vowed viewers would never miss a moment and that the coverage would prove the claimed obsolescence of the BBC is nonsense. The BBC would stream every event simultaneously online, with up to 24 high-definition channels providing 2,500 hours of sport. Access was via mobile, tablet, PC, games consoles and TVs with a red button. There was social media integration, live alerts, an app, and guides to every athlete and event updated in real time. Selected highlights were in 3D. The result? The nation unified behind Team GB. The BBC got its halo effect. And the nation got the Games it yearned for.


Project: Universal Credit
Lesson: Confess errors and listen to critics

universal creditThe idea of Universal Credit is to fuse 30 welfare programmes into a single system. This will make it possible for the Department for Work & Pensions (DWP) to see the financial position of each person, ensuring that in every case the benefit of working outweighs the rewards of staying on government handouts. From that promising starting point, Universal Credit has been a disaster. Delays, grotesque overspend, technical failures and a lack of enrolment means it’s quite possibly the worst administered programme in modern memory. The National Audit Office blamed “a ‘fortress’ mentality within the programme team and a ‘good news’ reporting culture”. Further reviews said the team was isolated and defensive and gave misleading interviews to the press. The Office for Budget Responsibility condemned “the recent history of optimism bias in Universal Credit plans and other projects of this sort”.

Project: FiReControl Centres
Lesson: Is your project necessary?

firecontrol centresEven the stony-faced analysts at the National Audit Office (NAO) were aghast at the sheer blithering incompetence of the FiReControl project. The goal was to replace 46 fire service control centres with nine purpose-built facilities. A £469-million budget went up in smoke over seven years as the centres were built. None opened. Five years after the whole scheme was axed in 2010, seven of the unused centres stood empty, bleeding taxpayers’ cash. Amyas Morse, head of the NAO, says: “This is yet another example of a government IT project taking on a life of its own, absorbing ever-increasing resources without reaching its objectives. The rationale and benefits of a regional approach were unclear and badly communicated to locally accountable fire and rescue services, who remained unconvinced.” In other words, there was zero point to the project. But no one dared say so. Tragic.

Project: Rural Payments Agency
Lesson: Gets your specifications right

rural payments agencyThe Rural Payments Agency IT system is a start-to-finish horror show. It pays farmers their subsidies. The cost was four times higher than the original forecast of £75 million. An extra £680 million is needed to administer the system. And it didn’t work – not even after umpteen relaunches. At one point farmers were told to revert to paper forms. A key problem was the lack of research into how farmers would use the system. The mapping software included in the Rural Payments System takes a lot of bandwidth. Farmers are often stuck in broadband black spots, so couldn’t use it. Worse, if they did, the numbers of applicants would overwhelm the servers. It was nowhere near being able to cope with 120,000 farmers and 1,200 land agents. Re-engineering the system caused outages. A Public Accounts Committee would ask why the chief executive of the agency was still in his job following the traumas.

Project: Surrey Police IT system
Lesson: Does the product already exist?

surrey police systemAll Surrey Police needed was a standard reporting package to log crimes and track cases. Other police forces had workable systems. But Surrey Police decided to re-invent the wheel and build a bespoke system. £15 million went down the drain before the whole project was canned. At which point Surrey Police adopted an existing package which worked just fine. Grant Thornton looked at the wreckage and identified a long list of shortcomings, including a failure to track costs, technology errors and a lack of agile methodology to adapt when things went wrong. The mission was an “ambitious project that was beyond the in-house capabilities and experience”. It is a grisly reminder of the tendency for project managers to dream up elaborate solutions when a simple one will do.