Government by numbers: doing data differently

The challenges of government today are different to those of yesterday. Some things haven’t changed, though – attracting the brightest and best still requires a different approach from that of the private sector. Tight budgets are another – rare is the government with plenty of cash to spend.

However, some things have changed. The fierce media lens trained on governments has only been intensified by the ubiquitous presence of Twitter and Facebook. A change that’s less visible – but even more significant – lies in governments’ rich reservoirs of data. Thanks to new technologies, policymakers can now collect and analyse vast swathes of data about their citizens and the services they depend on. These data sets offer huge potential to a farsighted government. Using them effectively can deliver better-designed, bespoke public services and improved citizen outcomes.

But does this happen in practice? Well, yes and no.

Setting a clear vision

In its work with governments around the world, the Boston Consulting Group has seen some clear trends in the sources of value derived from big data and data analytics. The most critical element for government is to have a clear vision of the outcome they seek to achieve – without this vision, outcomes are often abandoned in favour of infrastructure.

A key mistake we’ve seen governments make time and again is acquiring technology for technology’s sake, building capability to no distinct end, and being unclear about the impact they want to achieve. Having a clear vision, and then understanding how to bring it all together – the governance, the appropriate risk profile, the right leadership culture, and an engine to drive the value – helps act as a sieve to ensure investments are focused on impact. Government can then create a platform or ecosystem across the public and private sectors to bring together the data that underpins its objective.

Now this is far from easy, but there are plenty of cases where it has been achieved. Take Transport for London, for example. At a talk given last year in Sydney, the UK capital’s transport authority discussed how they’ve tracked passenger volumes, analysed the data, and – by tweaking their services accordingly – have saved about £10 millions-worth of passengers’ time every year. It’s this way of using data to reduce the friction in people’s interactions with government that shows the value big data can provide.

Opportunity knocks

Traditionally, governments have been structured around controlling citizens’ access to administrative processes. The most exciting thing about digital is that it offers the opportunity to invert this relationship. Citizens’ access to the finite resource of public time need no longer be restricted. Government can define processes that reflect people’s preferences, letting them choose their own route through a series of user-friendly apps and artificial intelligence (AI) programs rather than making them follow a rigid departmental diktat.

In this context, data opens up two principal avenues of opportunity. One is around improving the usefulness and relevance of policy, and the other is related to service delivery and operational management.

When it comes to policy, we’ve seen definite advantages in being able to sense and understand what’s happening across government and to listen to what citizens are saying. This helps establish a more participatory democracy, so that people’s voices do get heard and we can understand where the true problems lie.

Data also highlights where attention and investment need to be focused. For example, New Zealand’s policymakers have used data analytics to identify the people most likely to be in long-term unemployment or on social benefits. Providing focused investment early on has improved the quality of life of its beneficiaries and reduced the government’s long-term liabilities.

We’ve also seen data help in the segmentation of service delivery – government can use its insights to interact with people through their preferred channels and deliver the appropriate levels of service. Averages are always suboptimal, and data gives us the opportunity to identify where we should be investing more or less than the average to achieve a better outcome. And in risk management, data analytics enables policymakers to identify where revenue leakages, such as tax fraud, may be occurring.

Making it real needs to start with multidisciplinary teams who possess not just policy, data and operations knowledge but also the users who are likely to be affected by the services that are being designed. Human-centred design requires input from those who are best placed to judge the services’ effectiveness – the consumers themselves. And reality being what it is, we rarely get things right first time round. The only way to allow for this is by defining a clear feedback loop in which people can provide human-level input about how well the process is working.

Towards an AI-powered future?

Beyond this, it would be remiss not to spotlight the potential for AI. This is because the same disintermediation and disruption faced by the private sector also apply in the public sector.

My sense is that no government will be able to evade the impending AI revolution – its potential for improving services and outcomes is just too great. And remember that in their interactions with government, individuals see it as whole rather than as a collection of individual agencies. Reducing this friction is an ongoing challenge, but – thanks to big data and AI – it can and will be done. It’s only a matter of time.


The Centre for Public Impact is investigating the way in which
artificial intelligence can improve outcomes for citizens.

Are you working in government and interested in how AI applies to your practice? Or are you are an AI practitioner who thinks your tools can have an application in government? If so, please get in touch.