At the recent Earnix Excelerate Summit, Wei Ke, PhD, addressed the topic of “Mastering Personalization in Financial Services”, inviting the audience to holistically look at customer data, and to stretch personalization to its limits.
There has been significant discussion about personalization in banking, including on this blog, and its role in improving the effectiveness of risk-based pricing in the ultra-competitive fight for new customers. Banking personalization has also been shown to be highly effective throughout the customer lifecycle, contributing to the maximization of customer lifetime value (LTV).
The rewards can be significant: Wei reports that “we typically see 20 to 30% improvement in sales as a result, with a significant increase in lifetime value as well.”
As with all Summit sessions, this one was wide-ranging and informative. Here are just some highlights of the discussion.
Personalization – Something Old, Something New
Wei opened his session with the observation that “[personalization] is an interesting topic, because it’s sort of both an old topic and a new topic.
“The underlying models that you have to think about in personalizing – for example, pricing and promotional offers – have been around for quite some time. But the pandemic certainly has drastically changed the landscape, because very few of our customers are going to branches, especially during the lockdown periods over the last 18 months or so.”
Realizing the Most in Personalization – A Structural Explanation?
“If you look back at the last, say, 15 to 20 years, [banks] have shown a desire to change. But many are organized as product organizations or into product silos, with groups of products that can come together for retail banking services.
The associated marketing activities are created in silos, without much consideration of the broader client relationship. There’s not a whole lot of coordination between these messages, and at the end of the day banking organizations think about marketing on a calendar schedule.
“A lot of banks, for example, schedule campaigns during summertime, because usually that’s when people move banking relationships – they’re moving for a different job, moving to a different location for school, and so forth.
“[Instead of calendar-based marketing], if we can come up with a set of triggers that’s identified through advanced analytical models, like machine learning (ML) and so forth, using a comprehensive set of data to understand customer behaviors, then we can come up with a personalized communication strategy. Those communications can consist of both sales messages as well as non-sales messages, with an overarching objective, and you can then sequence those messages. That’s the end vision of what we’re hoping to achieve here.”
Playing the “Long Game” – Marketing Throughout the Customer Lifecycle
“If you look at a vision of a more orchestrated and personalized messaging strategy or marketing strategy, what we want to think about is to actually lay out the entire client journey from the moment that this particular client became a client, at the time of onboarding for their very first product, then to retirement and eventually wealth transfer. This is a very long-term kind of a horizon for that particular client.
“If you have an overarching objective like maximizing customer lifetime value (LTV), of course, that’s a financial metric that benefits the bank. But at the same time, it may be also improving the overall financial health and wellbeing of that particular client – there are many things you could do to help improve both objectives.”
Four Pillars of Marketing Orchestration
As a way of reaching this more evolved method of marketing, Wei laid out four pillars to guide the transformation:
Pillar One – Define the Objective – Very Clearly
“Pillar One is to define the overarching objective, whether it’s lifetime value or financial well-being, or some other objective – you need to have that very clearly defined.
“Banks tend to use fairly basic indicators that don’t really meet the needs of the client. And they tend to use these metrics in a way to construct what looks like a ‘carpet-bombing’ exercise, [based on] very myopic business objectives. Those outbound campaigns tend to focus us on immediate revenue maximization, or if you think about deposit campaigns, they tend to focus on new money inflow, without really much consideration of the bigger objective, which is lifetime value of the customer.”
Pillar Two – Collect the Right Data
“And the second pillar is to collect the right, relevant data. And here we’re looking at both big data sets and small data sets. Things that might be structured data as well as unstructured data, of course in compliance with the various different jurisdictions and regulations on privacy.
“The first data we could use is what I call ‘flow of funds’ information. Basically, looking at how a typical customer uses their various accounts, for example, checking accounts and savings accounts. As well as the other side of the balance sheet, which is their lending activities and payments activities.
“If we are able to actually map out the entire financial network of that particular client with information within our bank, but also if a checking account is involved and there’s actually an external banking relationship, a lot of our clients do have two to three outside banking relationships. We should at least be able to guess at some outside activities that these clients have. This gives you a very rich set of data on the [financial] behaviors and the needs of our clients, when you start with flow of funds.
“Then you augment with some of the other stuff, for example, social media information. This data gives you a sense of the different interests of our clients. And that could be useful for prospect targeting purposes, in identifying new customers. The revealed life stage information that these clients have will dictate product needs [and augment] the overall relationship data that we have collected within our bank.”
Pillar Three – Advanced Modeling
“Of course, we will need to develop the required models to measure client behavior as well as to predict their reactions.
“I would just caution that the more classic definition of ‘lifetime’ tends to be based on a very stylized understanding of customer behavior or customer contribution to the overall profitability of that relationship.
“Building a model that’s not necessarily based on steady-state metrics, but rather trends – metrics that look at the starting point of that relationship during the first months to the first two years – and then leading to the eventual steady-state data is something that works much better.”
Pillar Four – Omnichannel Communications Coordination
“Finally, let’s make sure that there is an omnichannel coordination strategy, a client-facing digital experience. A typical organization would have to do this in stages.
“The first stage, the simplest approach, is to create trigger-based communication, to identify triggers that will lead to the messages that you will send.
“And then, once you have built that capability, you move on to what is called the chain communication capability – here you can sequence those messages together.
“And then finally you go into predictive communications. You’re now starting to not just be reactive to the customer behavior, but rather you’re proactively forecasting customer behaviors.”
As mentioned earlier, the conversation with Wei was wide-ranging and informative, and we’ve only scratched the surface in this post. For the full video, please be sure to visit the Earnix Excelerate section of the Earnix website.
The post Personalization: Building A Holistic View of the Banking Customer appeared first on Earnix Blog.