Lessons from Retail Customer Analytics
By Ian Rutherford, Rachael Mcbrearty
The retail industry is perhaps one of the most advanced verticals when it comes to effectively using data analytics to engage customers at the right time, in the right context and with the right information to increase sales. Retailers excel at capturing data on customer behaviors, preferences and buying patterns and then analyzing it to create highly targeted, contextually relevant offers and services.
Fortunately, banks can emulate many of these best practices to attract more customers, keep their loyalty and increase revenue. With digital technologies and advanced data analytics, banks can accurately predict and respond to customer needs based on granular knowledge of the customer’s past, coupled with a real-time understanding of their present context.
Customer Traffic Patterns
The first step is to understand customers’ needs by gaining insight into what’s happening in the branch today. Video analytics and customer mobile data can help banks understand traffic patterns and interactions within the branch. By analyzing traffic patterns, banks can understand frequency of visits to the branch, what days or times customers visit and how they behave once they’re in the branch.
Walmart is a good example of a retailer that uses a context-aware mobile application to blend the physical and digital channels into a seamless experience for customers when they visit the store. The app is able to understand where the customer is standing in the store and what they’re looking at, then push relevant product information to their phone to support their decision-making process. With the proliferation of mobile banking apps, banks can employ similar strategies to create a seamless, personalized experience for customers when they enter the branch.
Traditionally banks have been reluctant to adopt these types of technologies due to concerns over client confidentiality, data security and regulatory compliance. However, banks can look to retail for best practices in these areas as well. Retailers like Walgreens have succeeded at engaging customers through their mobile applications while balancing privacy and regulatory compliance requirements related to prescription refills in the pharmacy.
As long as banks work diligently and transparently with their customers by proactively seeking their authorization and anonymizing data – and if the customer gives permission and connects to the branch WIFI – banks can understand who comes to the branch, what their motivations are and provide personalized responses in real time.
The second step is to get customers engaged in the mobile banking application and obtain information they’re willing to share and what type of engagement they wish to have with the bank. Begin by identifying what your customer profile needs to look like and how you can use your various customer touch points to capture the appropriate data on what to provide and when to provide it.
For example, did that customer previously start a mortgage application online only to abandon the process? Using digital technologies that recognize the customer’s mobile device, the bank can know this information as soon as the customer walks in the door and immediately put them in touch with a mortgage advisor to pick up the application process where they left off.
By better understanding customers’ past behavior and current context, data analytics can help banks understand how best to engage each customer in real time. These interactions can also be nuanced based on a client’s attitudinal profile. It would be helpful, for example, to know whether a customer is a financial novice who will respond positively to basic financial services advice or are already well-versed in finance and would be irritated by receiving such basic information.
Next, using advanced analytics, banks should mine the data for patterns and intelligence on customers’ past behavior in order to predict future behaviors. For example, what product might be of interest to the customer based on past purchases and life events? What does the customer’s decision-making process or purchasing path look like? The key is to understand what information is important to creating an effective response that reaches the customer at the right time, in the right channel.
Here are some tips for capturing customer data:
The more behavioral data the better. Capture data and insights from customer interactions through the website, mobile application, in-branch WIFI, rewards programs, employees and social media.
Request the right level of data based on the customer’s level of engagement. Capture data on the customer’s life circumstances, saving and spending habits or general financial acumen. Collection of this data should be designed into the process rather than collected through long survey forms.
Allow the customer to have visibility into and management over his or her data and communications preferences. Tell the customer how you will be using their data to their benefit.
Capturing the wealth of data from all available sources is actually the easy part. The challenge is in interpreting the data, identifying the most meaningful metrics and creating the appropriate responses that will entice the customer to take action.
Consider these tips:
Consolidate the different types of data, for example, marketing data, sales data, account information, data from different channels, social media and what’s gathered by employees.
Consider the most profitable customer segments and prioritize protecting valuable existing customers over attracting the desirable new ones, understanding that not all customers are created equal.
Find patterns in customer behavior, preferences, spending and savings habits that are triggers to purchase and retention. Build an aggregate knowledge by customer profile types and subtypes based on behavioral attributes, which will enable the bank to deliver interactions that have the feeling of a “one-to-one” understanding of each customer.
Iterate. Establish a baseline to evaluate the impact of the analytics. Data about the effectiveness of each banking transaction will be more valuable than the transaction itself in the early stages as it provides greater intelligence about what it takes to influence customers. Eventually organizations will be able to predict a customer need before they seek out the product or service.
Using the intelligence gained through real-time data banks can then optimize offers and services based on a customer’s profitability or financial level, likelihood to use a particular product or service, life stage or customer profile segmentation.
Once the data is interpreted, these steps will help in creating responses to those identified customer needs:
Define the “next-best-action.” Determine the most effective places to apply analytics in real-time within specific steps or sub-steps within the customer buying journey. Identify the customer needs and how you want to influence or assist them to determine the “next-best-action” as they engage with the bank.
Optimize each contact by setting business rules to manage frequency of customer communications across channels and by message types. Prioritize communications and offers based on customer preferences and specific marketing needs. Set the business rules to avoid over communication and send offers or messages to customers’ preferred channels, such as email, mobile application alert, SMS or direct mail.
Use machine learning to automate the matching and targeting of customers with high-propensity offers. Draw from the transactional behavior to gain intelligence about future behaviors and predict the best offers for each customer.
The final technology needed to bring this all together is data virtualization. The data analytics techniques described above require the aggregation of large volumes of disparate data sets. Traditional data warehousing methods are inefficient, expensive and inherently lose the immediacy of the data intelligence, instead becoming a historical analysis of what “might have been.” Because more customers conduct traditional transactional banking services through the website or mobile channels today, when a customer does walk into the branch it’s likely they’re seeking qualified, professional advice on higher value products such as mortgages, retirement savings and wealth management.
Data virtualization enables the real-time insights and immediate decision-making at the source that is needed for banks to deliver personalized service in the moment. There is a direct correlation to sales performance in the branch and immediacy of qualified advice. Nationwide Building Society in the UK saw mortgage sales performance increase by two-thirds as a result of using video collaboration and telepresence technologies to put branch customers in immediate contact with remote mortgage advisors.
As banks move increasingly from physical footprint to local presence with virtualized delivery of advice, they can also use the analytics of branch performance to make decisions about what services should be delivered in the future in that location, or if that branch is ultimately unlikely to deliver the financial results required.
Lastly, perhaps more important than the digital interactions are the personalized interactions that only staff members can deliver – particularly in the case of regulated products such as mortgages and wealth. The data intelligence provided through analytics and mobile technologies not only enables banks to deliver targeted, timely responses through digital channels, but also provides insights that enable staff to better engage and service clients.