Faster decisions can be made in every situation, if decision makers have access to the relevant data. And this is the challenge for organisations that are facing the very real possibility of being disrupted if they fail to keep up with the latest standards of digital engagement.
We spoke with Pete Williams, Former Head of Enterprise Analytics at Marks and Spencer, and the world’s 6th ranked most influential person in data-driven business about how data is changing, and will change society as we know it.
Read Pete’s Q&A below to see how data driven insights are transforming business, and what the public sector can learn from the changing nature of the customer in a digital age.
Can you give us a general overview of how data changed the business landscape over the last 5-10 years?
Fundamentally! I don’t think there’s an area of business that hasn’t been affected by one of the data related opportunities now widely available. Whether that’s the type of data we can now collect and the things we can now measure, the speed and volume at which it can be collected or the ability to process and deploy the results faster.
More informed, higher quality and faster decisions can be made when data is presented properly at every level of a business, ensuring focus on the areas of decision making that demand attention. In many ways the most significant impact of the big data fever has been visualisation, where skilled data storytellers can turn complexity into common sense to consider for the best decision possible.
These activities have been sped up by the increasing availability of data science and AI technologies, where patterns and inferences in the data you’re collecting can be automatically generated and applied.
Inside organisations, empowerment of business analysts with tools and capabilities is leading to a realignment of the relationship between business teams and information technology.
In the wider market the clever use of data and emergent platforms has allowed disruptive businesses to spot gaps in the market and rapidly pull together then scale a business model to challenge incumbents. This was never previously possible.
However, a series of even more interesting cultural and societal challenges are posed by this rate of progress in the next 5-10 years.
One of the most tangible benefits of data is that it enables better and faster decisions to be made. How can organisational strategies be developed with this in mind?
For a start, recognise that data isn’t the strategy. Whatever your business, your customer should be your strategic focus. That could be delighting existing customers or finding new ones. Data will be an enabler to achieve it at the appropriate time.
So, workout what you want to be in the horizon you set yourself to achieve it. We usually hear of 3 or 5 year plans. Data based insight can inform some of this visioning. Workout an honest assessment of where you are now. The gap between the two informs the tactical plans you need to put in place to achieve your strategy. I’d then expect data activities to be the enabler for some or all of these plans.
In this process I’d like to ban the use of the word data as it can be limiting (“we don’t have that data”, “data quality is bad”). Debate and decide what you need to know and do. Then decide how you’re going to do it. Then tackle those problems.
How has data fuelled the need for private organisations to put more care into the customer experience?
In areas of the market where disruption has been most celebrated and reported (Uber, AirBNB etc) these business have shown how readily available cloud platforms and open source or bespoke technologies empower an easy start and rapid scalability.
One of the key characteristics is that these business can start small and agile but driven by technology to a degree not previously possible. They use technology to be close to the customer requirements and experience. It can feel both incredibly personal and also remarkably easy to use their processes. And it’s a reciprocal arrangement as the constant measurement of customer experience and satisfaction is fed back into product and process development very quickly.
Larger, more traditional organisations can’t move at this speed. They have usually either not invested in being so close to the customer experience, or built layers of method and process so the insight is trapped and not impactful as it should be.
At the same time the rise of mobile has given customers the most immediate and powerful soapbox ever in social media. If things are good they’ll talk about it. If things are bad they’ll scream and shout into the social space so everyone knows. In an age where a brand is defined by the customer experience, not what you want it to be, an organisation cannot afford to be cut off from the data driven channels now available to customers.
Bringing all of this together in a customer centric culture, providing a digitally aware customer experience is vital to maintaining brand health and therefore commercial success.
What are some of the barriers to using big data?
It’s rarely the technology. While the cloud platforms and open source stack are different, many people and organisations can now take full advantage of them.
I believe the barriers are cultural and organisational and come down to the issue of large scale data literacy and leadership.
Understanding how data driven decisions could benefit the business processes is beyond many senior managers who haven’t been exposed to working with data. It therefore stands to reason that the sponsorship and drive that the journey to a data driven culture demands is underestimated.
I call this environment the Big Data Analytics Ecosystem, where the technology is a subset of a full scale business digital transformation to ensure the benefits can be realised. Business leaders need to be trained in asking the right questions and having confidence in the generated answers. Managers need to be capable of making a fast, data backed decision. And the whole organisation needs to speed up – you can’t be thinking in terms of weekly or monthly data anymore, even daily could be too slow.
Data is so fundamental to better decision making that it deserves a board level presence and an experienced data leader (not a technology leader, but someone who understands how to enable data for commercial benefit) to show how it can benefit most situations and take accountability for it happening. I back the current need for the Chief Data Officer role as a CEO appointed advisor until boards are generally more data literate. But the buck should always stop with the CEO as a visible owner and champion for data driven transformation.
What are the biggest benefits?
To be blunt, you’ll stand a better chance of staying in business. Statistics show that companies using data effectively outperform their peers that aren’t. If you’re in an industry being disrupted by the new digital business models then it’s your only chance. If it hasn’t happened to you yet then it’s likely you’ll be able to outpace your competitors by using data to enable and drive your strategy.
With the right steps taken you’ll have a more agile organisation, more centered on your customer, with a more motivated workforce. It’s likely you’ll experience higher growth and profits, and be better prepared for market eventualities.
In your opinion, do you believe the complexity of government changes the nature of how they use data and its insights?
I’m not sure it changes the nature of how they use data and insights as I believe these will always be the way to the best decision. But the position brings a requirement for greater responsibility due to the potential impacts on individuals and society.
This alone makes it even more important that the right talent is attracted and robust processes are followed. I actually think the complexity and volume of data in government mean it’s essential and highly beneficial for government to become expert users of AI and predictive analytics.
A disconnect may come, viewed from an external perspective, between the right answer and a publicly acceptable answer. After all, there are “lies, damned lies and statistics” and data science is at its heart, statistics.
What skills sets should the public sector look for when building big data capabilities?
The same technical skills for individuals as for the private sector remain valid. A curious mindset and a self-starting nature is vital with the willingness to ask questions.
The ability to handle large volumes of data, cleansing, sorting and munging.
Mathematical capability. Data storytelling and visualisation flair. Given these are hard to find I do advocate ‘building the unicorn’ – that is ensuring all the skills are present in small, effective teams. Seeking individuals with all the skills is unlikely to bear fruit unless you are paying at the top of the market and offering a deeply satisfying challenge. Even then, they will be bombarded by other offers and may not last long with you.
However, these individuals will need strong guidance and defence as they execute and deliver their analysis. Senior leadership with a high level of data literacy is vital. That includes the ability to surface important issues to be resolved and the strength to tackle setbacks and failures which will happen.
What are some of the key lessons the public sector should keep in mind when developing their data strategies?
A data strategy has components of understanding data that you have stored; where it came from, how it’s transformed, how it ends up where it is, where it’s used and who sees or uses it. The public sector needs to pay particular attention to the elements around this data governance section. The public needs to know that the most sensitive of data is absolutely safe to be confident to engage.
The public sector also needs to understand the value it’s going to derive from the data. In a commercial organisation a significant output of the data strategy would be the proposed generation of business value. What would the public sector equivalent be where expenditure must be rigorously accounted for?
Another consideration might be silo behaviour. Corporations can be compromised by a sense of ownership in local teams or departments over data they maintain – in the public sector this could create significant blind spots to understanding and insight.
Do you see any potential for machine learning and predictive analytics within the public sector any time soon?
If this isn’t being used already I’d be disappointed. The public sector is exactly the place where these techniques could have a massive impact. I say this based on the volumes of data and the need for deeper understanding.
There’s valuable work that could be done on populations, towns, infrastructure requirements. How much could models be updated with new techniques or possible data sources? Could they run more efficiently or frequently to provide better foresight? What sort of forecasting could be done on public finances? Or scenario planning on external and economic factors? How could social data be used to keep a better ‘finger on the pulse’ of the public?
Pete Williams is delivering a keynote presentation at the GovInnovate 2017 Summit in Canberra. His keynote presentation is taking place at 9:45am on the 10th of October, and it will focus on how government organisations can begin to look at citizens more like customers. Register here to ensure you don’t miss the opportunity to learn new ways of delivering streamlined civic services.