Main content begins

Customer Analytics: Moving Beyond KPIs

April 11, 2017

An interview with Jonathan Adler, Director of Insights and Analytics at Lenati.

Many organizations compile customer data and establish key performance indicators (KPIs), but there is immense value in going beyond the high-level numbers. As Director of Insights and Analytics, Jonathan Adler helps global brands understand what is required to become data-driven organizations. He leads the Insights and Analytics Practice at Lenati, a marketing strategy firm committed to helping the world’s leading companies build customer loyalty and generate business growth. We asked him to share his perspective on customer data and analytics, and their importance in loyalty strategy.

How do you help companies derive value from their customer analytics and data sets?

At Lenati, we help companies design, maintain and manage their loyalty programs, and data is essential to this process. Data serves as evidence and the business-case validation for deciding which program tactics will influence customer behavior.  

The easiest place to get that evidence is through the existing customer data. We leverage existing data – or conduct primary research if needed – to understand the potential value of a loyalty program. We run ad hoc analyses and data simulations using machine learning for understanding if the program changes are influencing the right customer behaviors.

Also, a great deal of our work involves measurement – whether that’s setting up dashboards to track overall program performance or doing periodic deep dives to see how specific customer segments are performing. This is where choosing the right metrics to track customer behavior, program performance and customer lifetime value (CLV) becomes very important. Often, marketers want to go straight from a program design to picking key performance indicators (KPIs) to monitor a program. This sometimes works, but without careful thought, KPIs can be problematic and there are other options.

What are some common challenges with key performance indicators (KPIs)?

KPIs are very good at informing you when something changes. But what KPIs don’t do is provide deeper analysis of what happened and why it happened. For example, suppose retention was down 3% last month. Is 3% a big drop, or is it the kind of month-to-month fluctuation that should be expected? Was it from customers across all segments or just one type of customer? Was it at one location or across the company? KPIs typically won’t answer these deeper questions.

Marketers sometimes think having more KPIs will provide more insight. But, having too many can make it hard to understand the real story. It can be overwhelming to see some KPIs showing positive results while others show negative ones – which should be trusted? KPIs function as a snapshot, giving you a good idea of what’s going on at a specific point in time, but they don’t provide context.

What are additional methods beyond KPIs, and how can companies become more mature with data?

Companies can take two approaches beyond KPIs. The first is to do exploratory analytics, meaning you take the data and view it in different ways to get more context of the larger story. This involves taking the time to delve through the data to understand the problem. The proper tools for this are either business intelligence tools like Tableau and Excel, or data science programming languages like R or Python. Exploratory analytics allows you to slice and dice, pinpoint and interpret the issue, making it much more actionable.  

The second approach is predictive analytics: using advanced statistics and machine learning to find the relationships in the data to predict behavior. For example, suppose there’s a retention problem. You can use predictive analytics to build a statistical model to understand the drivers of customer churn, and predict which customers are likely to leave next. This informs the strategy on how to minimize churn. Predictive analytics allows marketers to be proactive and agile in their business versus reactive, when it might be too late to course correct.

What recommendations would you give to a company trying to better understand their customers through data?

First, establish a consistent reporting structure: a digital dashboard, a monthly report, an Excel sheet – just a consistent method to view customers and monitor program performance on a continual basis. A sure sign of an analytics problem is if people within the organization are constantly asking, “How are we doing?” and everyone has a different answer. It’s critical to have consistent data reporting that gets everyone speaking the same language.

The next step is to have the capability for exploratory analysis. If your reports indicate something is off, you can quickly dive in and assess the issue. Furthermore, establish a strong relationship with your insights and analytics resources. Ideally, you have someone who can translate a business question to a data science question. Then go find the right data, and translate that data science answer back to an answer that can be widely understood by the business. Asking your team to explore a data set or respond to a specific question is very different than requesting them to “make this graph.”

With those pieces in place, start thinking about the predictive models, which involve more advanced data science around customer lifetime value, churn modeling and segmentation. When should you start thinking about predictive analytics? Only when you’re already able to identify issues and find answers quickly, and feel like you have the pulse of your customers. Predictive modeling for loyalty can provide tremendous value, but the framework of data reporting and resources must be in place.

And finally, most companies are maturing in terms of collecting customer data. However, you still need to be conscientious about the customer ask: request only what’s needed and what will be helpful. There’s a tendency to try to obtain as much data as possible, but there can be a very real cost to gathering it. Customers get frustrated by extensive or continual data requests, which worsens the customer experience. You must always look at downstream impacts of gathering customer data – and consider the customer experience first.

Specific to loyalty, insights and analytics – done well – can be a tremendous asset. There’s potential for much greater ROI with simple adjustments. The process does require time and investment, but there’s great potential for long-term gains when you have the evidence to support your decisions.