Net Promoter Score (NPS) has become a KPI at many organizations, and for good reason. It is simple to measure and has the distinctive ability to indicate how the business will perform on several other metrics, including customer retention and revenue.
The challenge with current NPS data approaches
While NPS is easy to measure, it is difficult to improve. Organizations understand that detractors have low NPS because they have had an unsatisfactory experience at some point in their customer journey. However, identifying where or when the unsatisfactory experience occurred is a challenge. This is especially true for financial services organizations, where customers have unique and complex journeys through various different products and services. NPS suffers from these challenges for a few reasons:
- It doesn’t scale well. Determining where every negative experience occurred throughout each customer’s journey would require countless surveys or conversations. This would be a frustrating and time-consuming process for the customer.
- It forces organizations to be reactive in improving their customer journey. They only discover which areas are causing friction after customers have experienced them, meaning that they are constantly playing catch-up to improve these experiences.
- Analytics solutions require integrated customer data. Analyzing NPS scores in relation to specific experiences means you need to be able to tie NPS results to specific customers and their broader data sets. A holistic analysis requires a 360-degree customer view.
The challenge with using customer behavioral data to explain NPS
Financial services organizations are collecting data through many different touchpoints in the customer journey, all of which could be used to understand NPS – in theory. In practice, explaining NPS with this data is difficult due to the sheer depth and breadth of the data. Organizations need to understand the context around every individual experience in order to explain NPS, and because financial institutions track hundreds of customer interactions over the course of a customer journey, this data becomes too broad and confusing, overwhelming traditional statistical approaches.
Using a Sequence-Modeling approach to explain and predict your NPS
A related problem is scalability of analytical approaches. A typical data set for one individual customer could include thousands of touchpoints, log events, and variables. Building an understanding of which variables and experiences actually cause measurable and significant changes to a person’s NPS score is extremely difficult. Oftentimes, reframing (i.e., normalizing) the data leads to so much loss of information that conclusions aren’t actionable or clear.
This is where sequence modeling comes in. Customer experiences happen sequentially; for example, you visit a branch after receiving a direct mail solicitation; or purchase a credit card at or after the branch visit. New analytics tools look at sequence data more holistically, understanding what elements in the sequence impact the actual end result — that is, the NPS survey response. Such tools use AI, including probabilistic graphical models (PGMs), as well as deep learning approaches like Long Short Term Memory Networks.
Using AI-based sequence analysis tools like those above helps conserve the data and draw meaningful conclusions. Additionally, AI-based approaches are often predictive in nature, meaning they also can predict what an NPS score will be before a person is actually asked.
Business benefits of predicting NPS
The ability to predict NPS scores gives businesses the ability to proactively make improvements that will raise NPS and create more promoters. Using sequence-modelling, you have the power to predict where customers will have a negative experience. You can then intercept the customer journey, preventing customers from ever having this negative experience at all.
In a market where customers have more options and higher expectations than ever before, financial institutions must focus on developing higher NPS scores and more promoters. Using customer behavior data, AI, and predictive analytics, you can create a positive customer experience at every touchpoint.
Find out how you can use Canopy Labs to understand and improve your NPS. Get in touch with us today.