While at J.D. Power and Associates, we were visited by two gentlemen: Thomas Jones and Earl Sasser; they met with us about some research we were doing for Xerox. At the time, I didn’t realize the significance of their presence and those meetings, but a few years later, when I read their findings and groundbreaking thinking on “Why Satisfied Customers Defect,” (Harvard Business Review (c) November 1995. Sasser, Jones), I had an “a-ha moment.” I remembered our meetings with them quite vividly. Subsequently, we used this model with clients when I joined CustomerSat.
I bring this up now because the research created a loyalty model that gets over-shadowed by NPS, but I believe it’s a model that deserves a second look. (NOTE that I don’t advocate any one model; I think each one has its place and is right for the appropriate industries, uses, and reasons. Having said that, I’m versed in several loyalty models and typically discuss the pros and cons of each when asked by my clients.)
While NPS uses just one question (likelihood to recommend) to segment customers, the Apostle Model uses two: overall satisfaction and likelihood to repurchase. The graphic below shows the naming convention for the different customer segments, based on where they fall in each quadrant. When you look at the definitions below, you’ll understand why it’s important to look at customers using this two-dimensional approach: using just one or the other doesn’t give you the full picture. Just because you are satisfied doesn’t mean you are loyal, and vice versa.
Here’s a quick rundown of each segment:
Loyalists are satisfied customers who will buy again, and they are an extension of your sales force because they will spread the word about your brand. There are two subsets:
- Apostles are extremely satisfied and extremely loyal; they are repeat purchasers and happily spread the good word to friends and family. These customers have an emotional connection to the brand.
- Near Apostles are satisfied and loyal, but not as strong as Apostles.
Mercenaries are satisfied but not loyal. These customers often switch depending on which way the wind blows! They are price sensitive and will switch whenever the opportunity arises. I always use the example of a Tide detergent customer who buys All because they got a $3.00 coupon for All in the mail. (I previously mentioned Mercenaries in a blog about Black Friday shoppers.)
Hostages are not satisfied but they will purchase again. These customers are typically locked into a contract or have a corporate directive telling them they must purchase the particular brand. The products may have little or no competition, or there may just be a huge barrier/cost to switching.
Defectors are both dissatisfied and not loyal; they will not purchase your product again and bad-mouth you on the way out the door. There is one subset of this segment:
- Terrorists are extremely dissatisfied and extremely disloyal, i.e., they will never purchase again. These customers are your worst nightmare; they’ll share their dissatisfaction with their 10,000 friends!
The beauty of this model is that it provides a robust way to segment and understand customers, and it can help you take appropriate action by company, customer, industry, etc. Let’s look at how to prioritize addressing these segments. Keep in mind that it’s cheaper to retain customers than to acquire new ones.
Priority #1: Address the Defectors, especially Terrorists. This may mean figuring out how to “fire them” the best way possible (so they leave but don’t spread more ill will), figuring out the ROI of fixing their issues (may not be worth the time or resources), or designing a win-back initiative to turn them around. Be sure to understand why they’ve landed in this bucket – changing needs could be addressed in a cost-effective manner that will allow you to turn these customers around. They complain more frequently and can and will drain your resources, so move quickly to figure out an action plan. Also, if they do leave, they will share their reasons and experiences – and with social media, we know that the megaphone is on. They represent an extreme potential for negative value for your company.
Priority #2: Focus on your Apostles and their needs. For obvious reasons, you want to keep and grow this segment. They want to help your business succeed; that emotional tie they have achieved means they are now encouraging others to use your products and services. Facilitate community building for these folks. And don’t forget to regularly thank them for their business.
Priority #3: It is worthwhile to do whatever it takes to bump your Near Apostles up to Apostles. The slope of the curve where they fall is steeper, meaning that a small increase in satisfaction can yield a big increase in loyalty. Of course, you want to grow your Apostle segment, and this segment is the easiest to address to do that.
Priority #4: As little control as you may have over Mercenaries, the best way to ensure that they stay is to communicate the value they receive for the price they are paying for your product. Remind them of the benefits; after long periods of use, customers may forget and only think price. Also ensure that they are using the right product or service for their specific needs.
Priority #5: Hostages, who are exhibiting a false loyalty, should also be addressed because someday they may be scooped away by competition. Understand their needs, their reasons for dissatisfaction, and work through it with them. Communication is a key retention strategy for this segment. Again, cheaper to keep the customers you have.
You can see from the priorities that, similar to NPS, you want to reduce the percentage of Defectors and increase the percentage of Apostl
es. Also like NPS, this model works for employees, too.
This model is by no means perfect, and it has its proponents and opponents. One of the arguments against this model is that, because the segments were never clearly defined (e.g., Loyalists are Osat = 4, 5; LTP = 4, 5), it (a) is misused and (b) combines scores that should not be combined. The authors used a 5-point scale, but in the graphic in both the article and in The Service Profit Chain, where they also discuss this model, they never labeled the exact points that fall into the buckets for each segment. The issue it raises is that, as they state, a customer giving Xerox a 5 on overall satisfaction is six times more likely to repurchase than a customer who rates satisfaction with Xerox either a 3 or a 4. So these customers cannot be lumped together. At CustomerSat, we used a 10-point scale and created our own definitions around that scale (based on our findings) to mimic the spirit the authors were outlining in their graphic. Keep in mind, though, and I have not mentioned it yet, that the authors also noted a “Zone of Indifference,” which is basically a super-segment that excludes Terrorists, Apostles, and Near Apostles (loosely – again, values are not clearly defined). Basically they are saying that those three segments are your main focus … again, like NPS: reduce Terrorists, increase Apostles.
While the original model uses likelihood to repurchase as the loyalty question, the authors wrote about an updated model in 2003 (The Value Profit Chain. Heskett, Sasser, Jones) that incorporated likelihood to recommend, or “referral value,” as they called it.
My personal opinion… every model has its pros and cons. The key to using these models is that it gets the organization to focus on the customer, specifically, different kinds of customers: what makes customers happy, what doesn’t. Most important is to look at the experience from the customer’s perspective, figure out what it will take to fix whatever processes or operations are broken (creating Defectors), and make it right while at the same time continuing to do the things that delight customers (creating Apostles). My favorite thing to say is that it takes the right people, the right tools, the right data, the right touchpoints, the right processes, and the right culture to get the organization focused on the customer experience. Consider using a loyalty model such as the Apostle Model as one of the tools in your tool belt.