We need to build those storytelling skills, to have tools that can help with making data visually exciting, because humans like drama! We all like a good narrative – Abhi Toraskar
The banking industry is blessed as one of the most data-rich industries in the world, with decades’ worth of customer transaction and behavioural insights built up and at their disposal. Yet, cultural inertia, widespread data illiteracy, and a myopic focus on product over people is preventing FSIs and their staff from making use of this invaluable asset.
Here, as part of FST’s digital banking panel series, data and analytics leaders from Australia’s top financial institutions discussed strategies for nurturing a positive, business-wide data culture, one that challenges ingrained corporate biases, methods to increase data literacy and connectedness across the organisation, and why an element of fun and indeed, humanity, ensures all employees can actively engage with the business’s data assets.
- Dr Catherine Lopes, Head of Data Strategy & Analytics, ME Bank
- Abhi Toraskar, Head of BFS Data & Analytics Platforms, Macquarie Group
- Michelle Pinheiro, Global Head of Data Governance, ANZ
- Meggy Chung, General Manager – Data Platforms, Westpac
Moderated by Cath Gullo, Regional Vice President, Financial Services, Tableau
Cath Gullo (Moderator): Research shows that extreme customer-centricity is paramount to retaining loyal customers and acquiring new ones. As data leaders, what do each of you feel are critical enablers to tapping into data for greater outcomes, for both the customer and your organisations?
Meggy Chung (Westpac): To me, it’s a problem that typically exists within most corporates, and that’s a very deep-rooted product and service-based structure. This tends to drive a very inward-looking, product-based focus.
People use data, for instance, in a very product-centric way, thinking, ‘How can I sell more products?’ or ‘How do I improve the services that I’m already offering to customers?’. That’s because the organisation is structured that way, it’s rewarded that way, and it thinks that way.
I consider this [product-centric thinking] to be the biggest obstacle for organisations in turning their culture into one that can reimagine data.
To offer a concrete example, at a previous employer, we came up with a model that worked out the next best offer to give a customer at any point in time, which we could send to any channel that a customer is using. To get that model live, however, we faced obstacles with the heads of our cards business who challenged us, asking why we’re not selling credit cards all the time or selling a mortgage then and there. That’s because the organisation thinks in a very product-centric and service-centric way – a very inward-looking one.
To truly be customer-centric, you need to have the muscle to start looking from a customer perspective completely, using the data regardless of what you have internally.
Having put all that infrastructure and all those capabilities in place at Westpac, this is exactly the space we’re moving into: How do we get the organisation’s culture shifting and able to reimagine data differently?
Dr Catherine Lopes (ME Bank): When we consider culture, and it doesn’t matter if it’s a data culture, it simply means that enough people think something’s in fashion.
It doesn’t matter whether you’re in IT, you’re in the technology or technical areas of the business or in the customer engagement or digital space, we need to ask: do you consider using data to be fun or useful? That’s the biggest obstacle to transformation I’ve faced over my many years.
You can have a dashboard to show you what data can do for you; we, as the data team, are very, very happy to share this. We want to use data, to help us, to guide us, to make decisions and to design our products. So that’s becoming a culture.
Michelle Pinheiro (ANZ): There are, for me, a few things. Firstly, you need to set up your data highways properly. And you need to have adequate data assets to work from. Obviously, the quality of that asset and how it’s being structured is really important. But you also need to have really reliable data pipelines, or data highways, so that if you’re going to hang operational processes or draw immediate analytics to help with customer decisioning, you need to have reliability in those processes. On top of that, having automation built into that is crucially important.
What I’m looking at now, in terms of building out a data culture, is that we see really good data strategies coming out from business units that want to use data to their advantage. And those data strategies map what their use cases are to what data they need.
Quite often I’ve seen data strategies that are more technology strategies than true data strategies.
From there, we need a strong lineage of metrics; we need to have throughput and to understand how infrastructure, in serving up data, helps with the metrics and the measures connected with the business strategy. That’s really important.
Cath Gullo (Moderator): Quite a number of you have tapped into my next question, which is really around creating an inquisitive culture. Research shows that curiosity at work fuels business innovation. And, in a data culture, people ask the hard questions, challenge ideas, which requires the changing of mindsets, attitudes and habits.
Abhi, starting with you, what are the biggest challenges in creating a data culture and a data-driven organisation?
Abhi Toraskar (Macquarie Group): Before I answer that, can I just say how wonderful it is to have so many women on this panel!
I’m an avid quizzer and inquisitor. It is a way of connecting disparate things within your world and making sense of it. We need to enable that within the enterprise as well. You will have different datasets that are seemingly disconnected: How do we make it easy for someone to link them together in a meaningful manner and derive insights out of that?
Why do people use Google? People like to ask questions, and they know they can get a really good answer, often just one click away. We need to have a similar ecosystem where it’s easy to ask questions and you’re not going to be bombarded with reams and reams of data that is very technical and that doesn’t make sense. You need to have data that has business definitions attached to it, and it is easier to make sense of it.
One of the key challenges around data-driven decisioning is that we need to accept that, as humans, we’re very much intuitive beings.
We do stuff by instinct, that’s how we’re lived for tens of thousands of years. It’s only now that we are in a data-rich world where data has just exploded. It’s a mindset that we need to get into; but for that, as I mentioned, we need to have all data aggregated in a meaningful manner so that it becomes an augmentation to your intuitiveness.
Cath Gullo (Moderator): Do you think there’s a need for an internal community, where people are excited about data and its potential impact on the organisation, or is there a better way of achieving that data culture?
Meggy Chung (Westpac): I want to borrow Catherine’s words of making it ‘fun’. In fact, to get that cultural shift, facilitating our imaginations is so important. I’ve used that word ‘fun’ before in creating real-time dashboards. You can talk about getting “real-time data” until the cows come home and people will go “Yeah, yeah, yeah”.
But if you put together dashboards and start showing people what’s going on at a certain point in time, you can begin to paint a picture of the art of the possible.
And then the imagination starts coming and people will start thinking, “Oh, that’s what I can now do!”. That, for me, is very important.
One example I’ll share from a past institution I worked in, we were looking at real-time data specifically from affluent customers. We realised there are a bunch of people, individuals who’d always been using digital, suddenly visiting a branch on the first Monday of every month. We did not realise that was happening. After investigating, we found that there was a particular set of parameter combinations that meant our digital app was locking a bunch of our customers out at the end of every month. We only came about this because people were being inquisitive about the data.
Back to this idea of community. For me, communities are very important. All throughout my career, I’ve driven a lot of ‘women in technology’ and ‘women in data’ activities. But here at Westpac, I have ‘Data Girls’.
Data Girls’ objective is to create a tribe of confident women at the forefront of data.
And in running this younger community, rather than simply running some panel discussions, I’ve said to the girls: “Try to do what you think is good to achieve the objective we want to achieve”. They came up with some really innovative ways of nurturing that community. For example, they said, “Let’s teach women how to debate!”, followed by “Let’s have a debate on big data!”. Next month we’re running a hackathon on humane technology, asking, ‘How do you protect banks from fake news?’ It’s really about allowing younger generations to come forward and consider data differently; that’s a very important, interesting, and fun way of getting people interested in data.
Dr Catherine Lopes (ME Bank): I saw the post of the debate, it was a wonderful concept, and a very interesting format, similar to a community we had at ANZ when I was there. ANZ’s first data science community, which started with just 10 members, within a year, grew to 200, and then just grew and grew.
Michelle Pinheiro (ANZ): Building that data culture and that excitement of using data to discover new human behaviours, it’s quite satisfying knowing, “Wow, people are doing that!”.
The biggest killer for the poor individuals doing that data work is that 80 per cent goes into the preparation of that data for analysis. That kills that data culture.
To streamline that and have that as something that data scientists, the data workers, don’t have to do, builds that excitement, satisfaction, and delivers much faster insights and change to the business. That, I feel, is the biggest contributor to a data culture.
Cath Gullo (Moderator): We talked about an inquisitive culture. Research suggests that data literacy is more than just purely technical skills – it’s being able to ask the right questions of data, to assess the relevance and the validity of that data, and to communicate findings to a wider audience.
Staying on the topic of culture and looking at data literacy, does your experience support those findings that data literacy is more than just technical? And what are you finding is the most challenging in terms of the skills to find?
Meggy Chung (Westpac): From my perspective, I own the technology as well as the analytics.
To me, the technology is the easy part. It’s building that data literacy that’s more challenging.
It isn’t building it in my data platforms team, it’s building it across the organisation.
At Westpac, we have a chapter model where, as leads of chapter, we drive literacy programs across all relevant lines of the business. We have sessions where we talk to the lines of business about the latest and greatest in data, how to talk data, what’s important about data, what is lineage, how do we approach lineage, and then all the way from how we do data management, how should we be looking after our data, to how do we do data democratisation (getting data out to people in a different way, to analytics, and to value creation)?
The good thing about this chapter model is that you get communities together, you start sharing ideas, they start exchanging ideas, and they start leveraging off, for instance, data management or visualisation.
Dr Catherine Lopes (ME Bank): Being that my background was in academia (though I’d call myself a recovering academic!), I wouldn’t mind sharing some thoughts on this. Like a master’s degree or an MBA within an organisation, I’ve built data analytics ‘academies’ within a few places I’ve worked. At ME Bank, I’m building another academy. Similar to the university model, the goal is to ensure that everyone is literate, to help everyone build a data-driven mindset. But, when we build these academies, we need to ask, do we really need management or an executive learning SQL and other such technical things? We, therefore, need to design different curriculums, and leverage whatever their baseline is – we want to lift them up, but to ensure it’s very well linked and designed across the entire organisation.
We have some fun sessions for young graduates and technical sessions for data linage, the tools we’re going to use. For the academic and engineer, we might have a session on understanding how we lift data science analytics from the dashboard to decision-driven automation in the cloud. I’m quite passionate about this academy, and it’s certainly the way we’re going to transform our organisations into a data culture.
Abhi Toraskar (Macquarie Group): Catherine, you mentioned SQL.
The main foundation of literacy is language. And the language of data for all purposes is SQL.
You need key people who are good with that, but not everybody can be good with SQL. Business folks cannot be. A lot of them use Excel. And tech people sometimes look down on Excel as their tool for data analysis, whereas, in fact, we should be encouraging the best use of Excel. We need to look at training around SQL or Excel, and then go into advanced tools like Tableau or other self-service tools. For me, it’s quite important to ensure that we have training in all variety of tools.
The second aspect of literacy is storytelling. If you just look at data, it’s often very dry and voluminous.
We need to build those storytelling skills, to have tools that can help with making data visually exciting, because humans like drama! We all like a good narrative.
We need to have those soft skills training as well so that we can make data-based insights and outputs more acceptable within the organisation.
Dr Catherine Lopes (ME Bank): It so very true! I could not agree more about Excel. I’ll give you a real-world case. Our finance department was widely using Excel, until another division decided Excel is no longer allowed. That’s not how you’re going to encourage people to transform into a data culture; they simply don’t have the experience. We need to start from somewhere they’re comfortable with to bring them on that journey. We need to help them use pivot tables, and then perhaps progress to using dimensional tables. This opens opportunities to now learn SQL, and from there, perhaps using dashboards and now automation. This journey took us a year to go through. But to just shut down one department’s use of Excel, that’s not going to work.
Cath Gullo (Moderator): With a significant digital transformation sweeping across the industry and the emergence of fintechs, how can traditional institutions best leverage your biggest asset, data, to compete and really excel, particularly from a customer engagement point of view?
Michelle Pinheiro (ANZ): A while ago the attitude was, “The fintechs are coming and there’ll be disruption!”. I think it’s fairly clear there are some pretty big market barriers to becoming a bank, connected largely to having enough capital.
But rather than seeing it from a competing perspective, it’s really about chasing the best experience for the customer and being able to provide them with personalised insights. Also, from an altruistic perspective, it’s about trying to really improve the financial position of customers – that’s what we’re really chasing here. It’s not about the best thing for the bank; it’s about the best thing for the customer.
The perspective, therefore, going into this decade is that it’s less about competing and more about partnership. How can banks partner with fintechs to provide that best experience?
Large organisations, such as the major banks, are really working on their speech delivery and enhancing their digital strategy so that they can leverage the power of their data for the benefit of the customer to provide them with insights to have greater financial literacy and be better off.
In terms of access to information, and indeed appropriate access to information, using information in the right way so that it is within the reasonable expectations of customers is vitally important. But then also providing them with those insights that we were talking about earlier, behaviours that were coming out of the data, and providing that back to the customer and informing them around their spending or savings behaviour and tweaks they need to make to be able to be better off. It’s really quite exciting. There are lots of fintechs that have some great ideas around how you can do that, and they can plug into those customer ecosystems to enhance that experience. That’s really exciting.
Meggy Chung (Westpac): Exactly as Michelle said, it’s less about competing and more about spotting those opportunities that would help you deliver more for your customers and more value for your customers.
If I can use a couple of examples from within Westpac. With digital banking being at the centre of our strategy, our partnership with 10X to accelerate the build on our banking-as-a-service platform provided us with an opportunity to significantly leapfrog our existing capabilities. Another good example is our partnership with Afterpay.
Later this year, Afterpay is going to be launching on this digital platform a set of cash management, savings, and budgeting tools that will be available to our customers, but anchored around our data.
So, we’re leveraging their products and capability and creating it around our data, and then offering a better set of services to our customers. I truly believe that it’s in this space, in spotting the right opportunities and ensuring our customer service and products create more value, is absolutely the way forward.
Abhi Toraskar (Macquarie Group): As a digital-only bank, Macquarie Bank is, in a way, between the big four and the fintechs. With banks, we have to understand that we have a much deeper relationship with the customer. We have customer relationships going back a number of years, even decades and we have customers that have different products with us. It’s important for us to leverage that information, which means building customer-360 and then providing greater personalisation to the customer off the back of that data. This means having much more robust sales and marketing, backed by tools like Salesforce in analytics. We get the best use of tools like that, underpinned by all this rich information we have. That’s how we can differentiate from fintechs.
This is an edited extract from FST’s Banking Digital Discussion – Unlocking the Business Value of Data digital panel, hosted on 1 June.