Companies in the financial sector have made enormous strides in getting a grip on their data, analysing and acting on it – whether it’s reducing risk, meeting regulatory objectives or monetising customer insights.
This trend is set to continue. According to industry analysts IDC, the financial sector’s spend on big data and business analytics will reach US$22.1 billion in 2019 – second only to discrete manufacturing.
However, it’s also fair to say that for many financial services organisations, the creation of data strategies and structures is still fledgling. They are not alone in this predicament as today’s analytics climate still feels highly unsettled. Models for governance, prioritisation, data and information architectures, talent development and approaches to problem-solving, are still ill-formed or even absent.
Just who is responsible? Are you the CIO? Chief Analytics Officer? Chief Data Officer? Chief Data Scientist? How about the CMO or CFO? You’re one – or likely more than one – of them at the same time. Either way, you are the person the CEO is pointing his finger at to figure out the way forward.
Doing so involves a number of balancing acts, which you should be aware of as you embark creating the strategy and framework that will determine your organisation’s future data collection and use of analytics.
Prioritising decisions over data
When we conduct analytics workshops with banks and financial services companies, we usually find that their main challenges fall into two categories: concerns about data – sources, formats, information governance – and concerns about decisions – where to apply analytics, picking the correct metrics, creating actionable insights.
Your own background – whether it’s IT, marketing or finance – may determine where you put the emphasis. However, on the whole, the industry’s focus is much more on data-related questions, at the expense of decision-making. This is like putting the cart before the horse and has to be avoided at all cost.
To avoid such an imbalance, companies are now starting to apply the principles of product design to their data analytics strategy. Design thinking puts the limelight on to users’ needs, asking questions such as: who are we designing for and what problems are they experiencing? Based on the answers, designers then nail down what the purpose of the design is.
In forming your data strategy, you should apply the same approach and look at users and purpose first of all: who will use the data and how are the resulting insights going to be applied?
Balancing creation with consumption
In addition to designing your data strategy with a purpose, building an analytics framework on this basis will also stem data overload.
Production of data is already outstripping our ability to use it, yet organisations are continuing to generate exponential amounts of “data for data’s sake.” There is a misguided view that more data, more dashboards and more reports mean more insights.
This is fallacy. Throwing more resources at something doesn’t always come up trumps, and more data doesn’t necessarily equate to more insights.
Consequently, planning your data strategy must start with your company’s business decision-makers and the measurable business outcomes they are aiming to achieve. Any data created has to link back to these, otherwise it will just serve to clog up your datacentre.
This is not to say that you should only focus on the business bigwigs and ignore the wider organisation. There are thousands of decision supply chains in your organisation, and it’s your job to ensure transparency across them, so that the data you generate benefits all parts of the company.
Asking great questions comes first
Your team will be under fire all the time to come up with answers – and quickly. No-one likes to appear ignorant or on the back foot, and after all, providing answers is what the analytics team is there for.
Rather than following human nature and blurting out answers frantically, a better approach is to hang fire and make sure that the questions being asked are the right ones. As the saying goes, silence is golden.
The reason is that data does not provide answers in its own right: it needs to be analysed and interpreted, and as such can be moulded to support virtually any position. Asking better questions can lead to a significantly more appropriate route to analysing data and resolving a given problem.
You many need to take a higher-level view of the problem, challenge basic assumptions or get a better understanding of the problem first before you tackle the data.
Are you having a pricing issue with a particular financial product, or is this just masking a much bigger underlying problem with how the product is sold? Do your in-branch salespeople understand it well enough? Is your website structure conducive to a smooth sales process?
You need to give your teams time – and training on what good questions look like.
Solving problems vs. problem solving
With the daily onslaught of questions you’re likely to face, it’s easy to get lost in firefighting. This means your team has not got time to look at the wider – often undiscovered – problems data could help with resolving. And they are missing an opportunity as a result.
Businesses face a myriad of small, often interconnected problems. Mapping these problems and their interconnections represents a fertile ground for innovation. Unlike the ‘big issues’ your company or the financial sector as a whole face, small problems often only require small solutions. So they can be resolved more cost-effectively and profitably. Plus, yours may be the only organisation in the sector to have come upon any particular constellation of problems.
Creating order from chaos
In creating your data strategy, you need to establish a framework for encoding problems in a consistent manner, and for balancing different forms of analytics – descriptive, inquisitive, predictive and prescriptive.
Recent research we undertook showed that companies mainly use descriptive and inquisitive methods – describing what happened in the business and why it happened – compared to predictive and prescriptive methods, which look at what’s likely to happen in the future and how the company should act in anticipation.
It’s important to get out of the descriptive/inquisitive comfort zone and ensure that a mix of analytics is used to unlock their full potential. This is as much a question of mind-set as it is one of time and resources.
In terms of overall governance structures, you need to make a tough decision between centralised analytics serving the whole business – but less responsive to changing needs across the organisations; decentralised analytics, which are more agile but likely to lead to information silos; or a federated model that combines the two. The latter is the most difficult to realise but it has the greatest potential payback and stands to foster the most collaboration across traditionally siloed departments.
Getting the culture right
The final balancing act en route to a sustainable data strategy is that between technology and culture.
Much of the public discourse around Big Data and analytics revolves around technology and the lack of qualified users of said technology – data scientists.
Little attention is paid to how to create an effective culture that supports the effective use of analytics throughout the organisation. Analytics should serve to render the organisation more flexible and able to adapt to an ever changing business environment.
For this, not just the analytics function but the whole company needs to adopt a growth mind-set which puts learning over knowing. Instead of relying on ‘tried and tested’ methods, continuous experimentation becomes key, and failure not only acceptable but desirable. Staff should be encouraged to provide authentic feedback regularly, so the company can act on and adapt to it.
Another major factor in this evolution is encouraging an interdisciplinary perspective both within and outside of the company. After all, it was technology businesses which revolutionised the music, hospitality and ground transport industries – not record, hotel or taxi companies.
With a solid data strategy, the right mind-set and an effective culture, your company should be in a good place to anticipate the next ‘disruptor’ rather than just reacting to it.
By Tom Pohlmann, Head of Values and Strategy, Mu Sigma.