(623) 123-4567

Customer Loyalty & Engagement

Customer Loyalty & Engagement


Over the past month I’ve been deep in survey design thought, absorbing material from Coursera, Marketing Profs, and Research Rockstar. I’m playing with the idea of a Brand EnSat study model that provides a robust picture of the engagement and satisfaction of a company’s customers.

Part of this thinking was inspired by a webinar Marking Profs distributed for Web Surveys at Cvent. Mike Phillips, the Director of Feedback Strategy, presented Value-Based Segmentation: Leveraging Customer Engagement Using Segmentation. It’s worth the 31 minutes to watch the broadcast on demand or even just download the slides on Marketing Prof’s website at: Marketing Prof’s website.

Aha moment:

The net promoter score is not predictive of loyalty. People who are likely to recommend your company may not be loyal customers themselves.

What is the Net Promoter Score? The NPS is a 0-10 scale question measuring a person’s likelihood to recommend your company to a friend or colleague. Customers are then grouped by their scores into segments like Promoters, Passives, and Detractors. Read more at: Net Promoter.com

According to Cvent, there is a pyramid leading to customer engagement. Similar to Maslow’s hierarchy of needs, first a customer must be satisfied, then he’ll become loyal, and finally he’ll be engaged. The net promoter score has been used as a predictor of loyalty, but Cvent says a person likely to recommend a brand may not be loyal themselves. Think about that. Have you recommended a product or service that you didn’t necessarily feel loyal to yourself? I have.

So if likelihood to recommend isn’t a predictor of loyalty, what is? Cvent says engagement is the predictor. They tested this theory by surveying people in their database before their renewal periods came due. In these surveys they asked satisifaction, net promoter, and engagement questions.

Ater the renewal period expired, they compared the answers to the questions with the actual actions taken. Were those who said they were likely to recommend the company the ones who were most likely to renew? No. Statistical testing said it wasn’t likelihood to recommend. Instead, the engagement questions proved to be more predictive of renewals.

Cvent lists these nine statements as validated engagement statements for segmentation purposes.

  • I am a committed [company] customer
  • I feel a sense of loyalty toward [company]
  • [Company] is my favorite provider of [product/service]
  • [Company] is very responsive to my needs
  • [Company] will do anything to make me happy
  • I feel a sense of personal relationship with [company]
  • I know what to expect from [company]
  • I am satisfied with the [company] experience
  • I use [company] because I get a good value for my money

What can you do?

This exercise is something that you could replicate within your company. For instance, if you work with subscription models you could survey a slice of folks who will be ready to renew in the next 3 months. You will need to tag survey respondents to identify their subscriber record so that after the renewal period passes, you can analyze which survey metrics statistically predicted if they would renew.

What you find can then lead you to segment your customers in new ways, modify the renewal messages you send them, or alter the approach you take to engage them.

A couple experiments you may want to run include:

  • determining if changing the wording from “likely to recommend” to “likely to “renew” makes a difference in the predictive power of the net promoter score when it comes to subscription-based service providers.
  • testing other engagement ideas. Are the engagement agreement questions Cvent uses relevant to your industry? Are there other engagement questions that should be tested for your particular business?

Contact me if you’d like to talk more about how to replicate this study with your own customers and create a better engagement model for the future.