In July 2015, the Department of the Prime Minister and Cabinet released its Public Sector Data Management report. The project aims to create a roadmap that will unlock the potential of public sector data to drive innovation, efficiency, productivity and economic growth.
The report found Australia’s capacity to remain competitive in the digital economy to be contingent upon its ability to harness the value of data. And it found Australia to be lagging on several counts.
Australia lags the United States and the United Kingdom, it argued, in releasing public data for business, and lags New Zealand in the application of data to policy design:
“Some countries started with a clear mandate to release data, while others focused on sharing and analysing it — there is no optimal sequence.”
Access to data is a prerequisite to any kind of analysis. According to Marek Rucinski, managing director of Accenture Digital – Analytics, it is only one of four preconditions that need to be met for any organisation to usefully extract value from data. He calls them the “four As”: availability, accessibility, affordability and ability.
Unless an organisation needs external data, he says that, for most organisations today, the first three preconditions are likely to have been met, because data is digital and the costs of storage and processing technologies have plummeted in recent years. Where organisations struggle is with ability. Not only do they need the analytic talent, but the organisation must be receptive to the outcomes of the analysis. “That’s a key pressure point within most of our client organisations,” Rucinski said.
“Even after you have addressed the data, addressed the tools and you have the insights, the end processes of the organisation are often not attuned to receive these and the workflows are not ready to act upon them to generate and convert the theory into value.”
He says a number of Accenture’s clients are at the stage of recognising, at the highest level, the need for an analytics capability. “They are feeling the competitive heat. There is a sense of urgency. So it’s a question of how they move from being a novice [in data analytics] to becoming a fully-fledged practitioner, and there is a clear pattern to how you do that,” he said.
Being passionate about analytics
Accenture defines an organisation that has fully embraced and exploited data analytics as one where “leaders behave analytically and show passion for analytical competition” and where “Analytics is integral to the company’s distinctive capability and strategy”.
Rucinski says it’s possible for a data analytics initiative to start from very small beginnings and generate its own momentum towards such a goal, using data visualisation tools like Tableau.
“There is a lot of value to be derived from basic organisation and visualisation of data,” he said. “If people are presented with data that is visually pleasing and easily consumed and easily comprehended, then insight generation can be self-driven.
“Once you get over that initial stage you start saying ‘I want to predict, I want to prioritise, I want to rank’. Those types of questions imply more complex analytics models and then you need to move up the food chain. Everyone wants to know more: they start asking new questions and those are invariably more complex and they lead to more advanced methods. Ultimately you get to a point where people want insights based on real-time data and presented in real time.”
This, he says, can lead to the development of centres of analytic excellence within an organisation. “That situation can persist until the C-suite or the leadership team becomes aware of the potential exploiting those insights can have across the organisation,” Rucinski said.
At that point the game changes. “Centres of excellence do not cut it any more,” he said. “There is a mandate from the top and you get the whole analytics construct coming to fruition. Organisations start seeing some of the more complex use cases they can leverage, not just marketing and customer data, but maybe supply chain and manufacturing data.”
But achieving this level of maturity in analytics also brings potential problems. “When you start to connect the data sources to create those insights you will need stronger data governance to allow you to say who owns what data and how you manage the connections and so forth,” he said. “You need a stronger governance layer to decide what the organisation should go after. What assets you need, what tools you need and how you organise the talent.”
Balancing governance with freedom
Loss of talent,Rucinski says, can be one of the biggest dangers as organisations’ analytical capability matures.
“If you have pockets of excellence you don’t want to destroy them,” he said. “What we usually see is that a new strong central governance entity develops that acts as a co-ordinator not a strait jacket, but it is a very fine line. If central governance becomes too strong it will kill the innovation at the edges.”
He recommends implementing data analytics through a hub and spoke model where a central team of data analysts looks at data holistically and works strategically, collaborating with analysts in the spokes who work more tactically on day-to-day issues and, most importantly, are close to the users of their insights.
“Proximity to users is directly connected to action on the insights,” Rucinski said. “The more remote you are the less are the chances of success. That is a very clear pattern we have seen over several decades.”
Another advantage of this model, he says, is that it creates a career path for analytical talent: if they feel they have exhausted possibilities for advancement in one “spoke” they have the option of taking on a different role in another, or at the centre.
