Big Data: new or not ?
When I attended this year’s edition of a Big Data in Retail Financial Services conference, everyone — speakers and participants alike — seemed to agree on one thing: We collectively hated the term “Big Data”.
Why? It’s too technical. It’s often misused by vendors. It creates confusion (especially with senior managers). It inflates expectations. And is often reminiscent of other failed “Big” IT implementations (data-warehousing, business intelligence, CRM …).
Moreover, for banks, collecting, storing and using customer data is not a new thing: Credit risk scoring is common practice since the ‘90s; most banks have undertaken one or multiple customer segmentation exercises; and most of them have also been building up a portfolio or marketing response models through their in-house analytics units.
So, nothing new under the sun, right?
An opportunity to grab
Yet a lot has changed during the past five years or so:
- Customers are more than ever in the driver’s seat, visibly impacting loyalty and churn rates.
- Government regulation is impacting margins across a variety of sectors.
- Schumpeter doesn’t sleep lately as non-traditional players keep disrupting business models.
- And finally, the economic downturn weighs on consumer spending behaviour.
Also banks cannot expect anymore to gain (or keep) market share by just pushing a good product in the market. Rather, the battle for every customer is raging. More than ever before, this requires them to bring customer-centric thinking in the way they change, plan, and operate their business.
The idea of customer centricity is not new either. Ten years ago (and since then continously), strategy consulting firms were preaching this as the new normal. It took an economic turndown and some dramatic examples of digital disruption to create a burning platform for customer-led strategies at board level.
This is where customer analytics or business intelligence professionals have a unique opportunity to stand up. Without rushing to build a 360° customer database (another loaded term), they should aim at answering a list of key questions that can bring clarity in the link between customer data and business results.
Often the underlying raw data is available in-house. Sometimes it’s scattered across silos, sometimes its quality varies, sometimes it requires cross-silo thinking, but …. it’s there.
The toothbrush test
Over the past years, and in my previous role as analytics manager, I have been working on leveraging what we had (it’s always a good starting point, so it’s what I am advising now to customers as well in my current role). In the years before my stint at KBC, my predecessors had done a great job of building a customer data mart, along with a portfolio of models that predict various cross- and up-sell opportunities at customer level.
But the team I started to lead a couple of years ago, faced challenges in :
- The systematic adoption of these models by marketers.
- The integration of customer intelligence in sales support tools.
This is where Google inspired me with their so-called “toothbrush test”:
In one article describing Google’s product strategy, its head of M&A said: “We ask ourselves, ‘Is this something people use once or twice a day and does it solve a problem?’ That sounds like a toothbrush, at least for those of us who want to keep our teeth.”
This is what inspired me to go about the problem differently. Evolutions in the business intelligence landscape (see, for example, Gartner Magic Quadrant for BI & Analytics 2014) now provide cost-effective options to rapidly deploy dashboards and interactive views on considerable sizes of data.
Me and my previous team used one of these solutions to unlock the value of our data and ‘data products’ to a broader audience of users — all while keeping in mind that answering business questions requires regular interactions and several iterations. Hence, an agile approach. The visual dashboarding & discovery approach created a momentum of change for the team that I was looking for.
And that’s exactly what data professionals should be asking themselves: “How can data, big or not, be made relevant for employees and customers on a daily basis?”