The federal government says the public service must be “smaller, more digital, and more flexible”. Data analytics is one important solution.

In its 2013 submission to the National Commission of Audit, Accenture Australia argued the current approach to public service delivery via state-owned bureaucracies was unsustainable:

“Government [must] re-evaluate their choices and re-ask the question about what really matters to citizens and businesses.”

Government agencies will also, Accenture argued, have to exploit data analytics to become much more citizen-centric:

“The insights enabled by data analytics give clarity into the differentiated needs of citizens and communities, and a starting point for developing much more finely segmented service platforms to meet a much broader range of personal needs at a much more sustainable cost.”

Marek Rucinski, managing director of Accenture Digital — Analytics, says there is now a pressing need for government agencies to adopt data analytics to increase both relevance and efficiency by enabling them to much more precisely tune their services to the needs of citizens.

He suggests government agencies can learn from the private sector, with data analytics the bridge between, creating synergies that could be exploited for mutual benefit and to Australia’s advantage. Commercial organisations in competitive environments have been facing the pressures of moving to a more customer-centric approach for a number of years; Rucinski says government agencies will need to undergo similar transformations to remain relevant.

“The power function has moved from brand-holders to citizens and government needs to very much reflect upon that in regards to how agencies adjust,” he said. “Governments are the policymakers but the citizens are much more vocal and much better informed. They will push back much more than they did in the past.”

The digitisation of all industries and all value chains is highly disruptive and is creating new business models in the private sector. Rucinski says the government sector is ripe for similar opportunities where agencies can start working with each other and combining datasets to allow them to be much more efficient.

He acknowledges that some of these initiatives will need to be well thought out and executed to ensure privacy and security concerns are addressed, but argues they could even be extended to the private sector.

“Industry could start leveraging the data to create value, to create partnerships and new ways of interacting with the government,” he said. “That creates the potential for Australia as a country to become much more productive. We have very good leading edge thinking and we are very good at adopting technology in Australia.

“If we can explore and expand on these opportunities we could be creating a virtuous cycle to leverage digitisation and create new value for the country and increase our competitiveness at a global level, and potentially export some of those new services because they would be quite portable.”

Challenges posed by data analytics

The transition to being an organisation that is data driven, to one that makes effective use of data analytics, is not without its challenges. Again, Rucinski says government can learn from the private sector.

“A lot of our private sector clients are struggling with the problems of having new sources of data that are breaking down previous paradigms of how they operate, how they report, how they act, and that are cutting through supply chain capabilities, forcing them to re-invent their processes,” he explained.

He says also that new data sources and data analytics can greatly accelerate internal processes, putting further stresses on an organisation. “When we look at marketing capabilities, the planning around marketing campaigns used to take six to 12 months. Now it takes days or weeks. That change in the length of time for a process creates a huge ripple effect through the organisation. Government agencies are facing a similar challenge,” he said.

Rucinski says the transition to becoming a data-driven organisation is well established in the private sector and provides a clear example for government bodies to follow. “Once that realisation  [about the importance of data] reaches the C-suite it triggers the recognition that a governance layer needs to be put in place to manage that data properly, but with a clear proviso that the governance must be very agile because you do not want to create a strait jacket around data.”

C-suite leadership might be a prerequisite for a successful transition to a data-driven organisation, but is no guarantee of success, Rucinski says. Too much attention devoted to data can be a recipe for disaster.

“Don’t start with the data, start with the strategic objectives of the agency and the change agenda the agency is facing today and use those as the starting point to reflect how digital capabilities can be applied to help with achieving those strategic objectives,” he said. “It is classic strategic prioritisation problem that is well proven, but using the value creation potential of data analytics as the lens through which to view it.

“And because it is a well-understood and communicated objective it will not be foreign. You can point to it and say ‘this is an objective that is not new, it comes from our leadership, but the way we are going to solve it is new; the way we are going to create value is new and the ripple effect it will have through the agency as a result of introducing these data analytics capabilities is new as well’.”

Leadership that recognises the importance of data analytics, sound data governance policies and some strategic objectives are perquisites for an agency to successfully exploit data analytics. But, Rucinski says, projects can still come off the rails if the right execution model is not adopted.

“There are very clear patterns that we see in both our private enterprise and public sector clients. You see people doing analytics everywhere in a network fashion. That’s good but it can create islands of excellence with little connection, and people can go rogue. It can be very difficult to scale and to grow talent because the talent is stuck in a node with no career path,” he said.

The hub-and-spoke model

“The pre-eminent model we see emerging is the hub and spoke model. You have the centre that is connected to the data at the governance layer and connected to the business at a strategic level to understand the priorities and the connecting points between those priorities.”

This hub has the ability to understand the techniques and skills needed to handle the use cases and to provide direction as to the data needed. The spokes, on the other hand, are close to the end users of the insights provided by the analytics.

“You have talent mobility between the hub and the spokes so there is a career path in place. And you avoid the ivory tower syndrome,” Rucinski explained.

“Breakthrough thinking comes from bringing disparate individuals together to look at a problem. Magic happens then.”

He sums the end-to-end data analytics ecosystem with five “I”s: investment, innovation, involvement, integration and industrialisation. “You must have the integration of processes, otherwise value will not be realised. Then you have to industrialise the process. After you have done it once you must learn to crank the handle at a much higher pace,” he said.

But he stresses strategic objectives must guide the whole data analytics effort.

“The most important thing is not to start with the data, start with the strategic objectives for the agency and the change agenda the agency is facing today and use those to reflect how data analytics techniques can be applied to help the agency reach those objectives,” he said.