MY FIRST PROPER JOB

My first ‘proper’ job was as an e-government programme manager for a local authority in 2004. The job felt an ideal fit for me as I was a certified practitioner in PRINCE2, a well-established project-management methodology, with a degree in electronic engineering and a penchant for technology. I also have a strong belief that technology can be hugely beneficial.

My role was to ensure that the authority achieved the outputs set out by the e-Government Unit (eGU). The eGU was part of Tony Blair’s Cabinet Office charged with securing the savings outlined in the 2004 Gershon Efficiency Review, and one of its chief targets was making 100% of council services accessible online.

It wasn’t long before concerns started to gnaw away at me. PRINCE2 is very clear that projects should be initiated on the back of a project mandate giving the project manager clear direction on what needs to be achieved and why. Very few of the projects had a mandate, and managers set to ‘benefit’ from a project were by no means certain if it was needed or why.

The real mandate for the project appeared to be meeting e-government conditions, along with the plausible but nebulous benefits of reducing operational costs and improving the service experienced by citizens.

As time went on, it became evident that many of the projects would not deliver the expected benefits. This was a common concern shared by e-government programme managers in other councils. I have yet to see a summary or reporting of the actual savings achieved nationally.

Enthusiasm from Whitehall

So why did national support for the programme continue? What did the architects of the programme believe would happen?

The eGU believed online services to be both more convenient for citizens and a cheaper transaction channel. The logical implication was to put all services online and ‘encourage’ citizens to use them.

Although I was not involved in the formative meetings of the e-government programme, I have observed discussions about digital transformation in financial-services companies, which are usually conducted on the basis of opinion and founded on assumptions that a) the current service offering works well and b) digital provision will enhance it.

In the e-government case, that led swiftly to the conviction that services should be ‘digital by default’.

The present government is equally enthusiastic about ‘digital by default’, its rhetoric is similar to that of Gershon, only now with more urgency and greater emphasis on financial pressures.

Thus in November 2015, the Autumn Statement and Spending Review declared that ‘a modern and reformed state is built on the understanding that higher spending does not automatically mean better services, and that by harnessing today’s technological advances, government can modernise public services, saving money and improving citizens’ interaction with the state’. To achieve this it announced a £1.8 billion ‘investment in digital transformation’ of government services.

Health Secretary Jeremy Hunt followed up a few months later by promising a £4.2 billion drive to bring the NHS into the digital age. A ‘paperless’ NHS, he said, would make services faster and more convenient, while the investment would also help the NHS save £22 billion by reducing waste and increasing productivity.

Dangerous enthusiasms

This is a dangerous enthusiasm. Over recent decades a spate of technology-based improvement initiatives has proved problematic for one or more of the following reasons:

• Delivering unnecessary functionality

• Not delivering necessary functionality

• Generating need for follow-on projects

• Exceeding budget

• Missing delivery dates

• Being cancelled before completion

These failings have not gone unnoticed by the digital and tech community. Much effort and innovation has been focused on how to become better at producing code, applications and devices. This enthusiasm for leveraging technology to deliver better outcomes for users is laudable. But it can only generate useful results when IT specialists work with those delivering services to first understand the problem from the perspective of the service user. After all, the role of code developers is not to develop code, but to overcome problems through the use of code; if they are not engaged on the right problem they will never produce the right code.

The human consequences

The most prominent current example of digital transformation is the Department of Work and Pensions (DWP) Universal Credit. If ‘digital by default’ is not an explicit strapline, it is not far from the surface. As The Guardian reported:

The work and pensions secretary, Iain Duncan Smith, is refusing to set up a freephone number for the estimated eight million people who are set to claim the new universal credit over the next four years.

When challenged on this a junior welfare minister, Justin Tomlinson, stated in a written parliamentary answer that they expect claims to be made online, although the government’s universal credit website did also advertise the chargeable number for the helpline.

Regardless of whether calls should be chargeable or not, the need for a telephone help with online access makes an absurdity of digital by default.

The experience of a Universal Credit claimant predictably looks like this:

Universal Credit – the user experience

1.  As an applicant for Universal Credit, you apply online (not easy for some) and receive on-screen confirmation that your application has been submitted. Weeks go by. After enquiries and complaints, your MP establishes that there is an intermittent error in receipt confirmation. Your claim has not reached the back office (and will not be back-dated to the time of the initial submission). ‘Please re-apply by telephone’.

2. You apply by telephone. The agent takes you through the questions you answered in the online application. Three-quarters of the way through, the portal ‘falls over’. The agent assures you that ‘It does this a lot; the data won’t be lost, but I can’t see it or complete the application right now. I’ll phone you back – possibly tomorrow’.

3. The agent calls back to complete the application. You make an appointment for a face-to-face interview to review your evidence and (with luck) get an estimate of the credit you will receive. You receive a letter confirming the interview time and the four items of supporting evidence that are required.

4. You attend the interview, at which you are informed that the letters are wrong and a fifth piece of information will be needed before your claim can be progressed. This is inconvenient and expensive (so many people give up), but ‘Can I bring it in this afternoon?’ ‘No, you have to make an appointment. No, we can’t make appointments, you have to phone the contact centre’ – which is busy taking calls from people making applications.

5. You go back to 3. Having supplied the necessary information, you experience the standard six-week wait, after which you chase up progress, to be assured that things are taking their course.

6. A letter arrives to inform you that the application is missing some information. On calling the contact centre you are told, ‘Ah, the letters are wrong, it’s all OK’.

7. You finally receive your Universal Credit, although the amount is not the same as what you were told and it is only backdated to a week after the date your telephone application was completed.

This is what the experience of a single adult with simple circumstances looks like – i.e. circumstances that the system should cope best with. It is unlikely to be the experience that the (then) work and pensions secretary envisioned and hard to reconcile with the idea of ‘digital by default’.

Human by default

The starting point for the effective deployment of technology should not be to decide what technology to use and then how to use it. Before any technology is even considered, it is necessary to analyse what happens when human beings put their hand up for a service. In other words, the approach should be not ‘digital’ but ‘human by default’. When human beings study demand, they learn what other human beings are asking for and why. They learn about the predictability and variety of demand.

They also learn that there is massive variation between people’s experiences of the service, even when they have similar needs. Some people find huge difficulties in getting a service, while others experience none. Some people get the service they need and want, others never get what they need at all.

All this may seem logical and obvious, but in my experience, it is rarely understood.

The traditional approach to understanding how the user interacts with a service is to gather a number of interested parties in a meeting room to map out on a process chart what should happen, rather than what actually does happen.

‘Human by default’ in social services

A typical Special Educational Needs Transport service, under pressure to reduce costs while continuing to fulfil statutory responsibilities, decided to take a radically different approach. Rather than jumping to a procurement exercise or buying off-the-shelf software costing hundreds of thousands of pounds, it studied its service.

They focused on the needs of the individual children and young people, providing them with more appropriate transport solutions and developing, where appropriate, their ability to travel independently. In the old system, no one helped the young people to travel independently, leaving many young adults institutionalised.  In the redesigned service, 60% of young people at secondary school age were identified as having the potential to travel independently.  Others reported an increase in self-esteem and feeling more involved in their community. In the first three years of the new design, the annual expenditure of £3m fell by half. In the years since, many more young people have been helped to learn to travel independently and as a result, the annual budget for this service has remained at £1.5m – all this without the benefit of new technology. Instead, the service changed the logic and assumptions that underpinned the design and management of the work.

As part of redesigning the new service, they discovered that employees needed access to data about service users, none of which was captured by existing software and applications.  So they commissioned bespoke software to enable them to call up the information whenever necessary. That is the rule: redesign comes first, IT development second, and then only if it is needed.

Technology is not to blame

Technology-based attempts to improve service and reduce costs often seem plausible and fit with the rhetoric of modernisation. Unfortunately, because it is so often used to solve the wrong problems in the wrong way, inevitable subsequent failure is typically blamed on technology. This is both inaccurate and unhelpful, resulting in years of multiple repetitions of the same error.

The straightforward antidote is a change in thinking from ‘digital by default’ to ‘human by default’. There is a huge amount of energy and enthusiasm in the digital community to leverage technology to provide useful outcomes for users. Starting with a position of ‘human by default’ ensures that this energy is used wisely.

Looking back on my days as an e-government programme manager, I realise my first proper job wasn’t as ‘proper’ as I first thought. It would have been a far more rewarding experience had our purpose been to ‘design services that work’ rather than ‘make services available online.’ I still have a penchant for technology and believe it has huge potential for good – but my starting point now is always to understand and improve the way the service works. Modernisation through technology will not in itself improve lives and reduce cost. Only human beings can do that.

So beware of naive and dangerous enthusiasms.

---

Read similar articles in Edition Two of The Vanguard Periodical: The Vanguard Method in Financial Services. Download it here