I started my CEM Toolbox series back in January and haven’t kept up with it at all. There have just been so many great things to write about since then. I’ll strive to get back to it, and if you have any suggestions for the Toolbox, I’m all ears!
Last week, Howard Burns of Microsoft and I hosted a CXPA Roundtable about the Apostle Model. As we wrapped up the call, I knew this would be a great topic to add to the Toolbox.
I’ve written about the Apostle Model a few times, but it’s time to update the theory with some practical application.
As a refresher, the Apostle Model is a different way (as opposed to NPS, for example) to segment your customers for the purpose of understanding loyalty and driving customer-centricity in the organization. This model uses two questions (overall satisfaction and likelihood to repurchase) to create the segments. The image above gives you a quick look at the segments and how they are defined.
Toward the end of the Roundtable, Howard talked about his experience with rolling out the Apostle Model within Microsoft. Some things to consider if you plan to use this model in your organization:
1. Know and understand your customers. This is important for everything you do, but it’s important when you’re choosing a loyalty model/metric or deciding how to segment your customers. According to The Service Profit Chain, “Organizations that have not identified the customers they are targeting have a special handicap in achieving total customer satisfaction. They too often attempt to please everyone, creating too many ‘merely satisfied’ customers and too few ‘apostles’ in the very core of the customer base in which they should be investing. Further, customer satisfaction measures too often are averaged across segments and not related to other measures that could provide insights into profitable strategies.” Amen.
2. Understand the metric. What is the Apostle Model? Understand how it works and the fact that it provides a different view of the customer. Overall satisfaction is overall satisfaction, but adding a second dimension, i.e., likelihood to repurchase, like the Apostle Model does, creates a different perspective. Also, determine the best loyalty measure for your organization to use in the Model. Microsoft uses a composite/index as the value for the y-axis.
3. Sell the metric. Why are we using this one? Executives within your organization will not just accept a metric because you say it’s the one to use. Do your homework. Explain why this is the best one for your organization, for your customers. Prove it.
4. Educate. Once you’ve sold it to the executives, you need to socialize it with the rest of the stakeholders within the organization. As a matter of fact, you’ll need to educate everyone, not just stakeholders.
5. Validate. Clearly there will be several validation points along this journey, including before you even sell the metric to your executives. But after you roll it out, you will want to continue to validate to ensure this metric makes sense for you.
6. Now what? Prepare a prescriptive guide that outlines how employees will deal, respond, and interact with the different customer segments differently.Why? Again citing The Service Profit Chain: “Satisfaction scores provide useful early warning of problems, but… satisfied customers do not systematically buy more than… unsatisfied ones.” This is why they did the research that resulted in the Apostle Model. Once we accept the fact that satisfied customers are not loyal customers, we need to realize that we need to outline how we will interact differently with customers, depending on who they are.
Result: A personalized customer experience. Hmmm… what a novel idea.
There is a big difference between a satisfied customer and a loyal customer. -Shep Hyken
Customer satisfaction is worthless. Customer loyalty is priceless. –Jeffrey Gitomer
Thanks for sharing this model, Annette. I hadn't seen this one before, but I like the two-dimensional approach.
Your post did trigger a few questions that I hope you can answer:
1) How do people react to terms "Apostle" and "Terrorists"? (They seem like they may be potentially offensive or confusing to the wrong audience, or right on target to the right audience.)
2) The Apostle and Terrorist label seem to imply a word of mouth component, but the dimensions of the model don't address that. Does this model account for word of mouth in some way?
3) Can you offer any suggestions on when this particular model would be a good fit?
Hi Annette,
Thanks for sharing this model. Like Jeff, I can see its usefulness but am concerned with some of the terms in the model and their implications and meanings…..apostle, loyalists, mercenaries, defectors, terrorists, hostages etc etc. Personally, how we describe things and the words we use is fundamental to how we 'feel' about different things. Therefore, although I like the dimensions of the model I am not convinced about some of the terms used in the model to describe different customer groups.
Adrian
Interesting post Annette.
I have heard very similar terms applied to change management within an organisation. Depending on people's engagement with a change and their influence they can also be described as terrorists, hostages, loyalists and the living dead.
The last category is a particularly nice image. Adrian, maybe that is one to keep to myself
James
Maybe you should, James 🙂
LOL. Love that. I hadn't heard that use case before!
Hey guys.
Sorry for the delay in responding. Jeff, let me start with your questions, and I think this will address Adrian's concerns, too.
1. Keep in mind that the authors created this model in the early to mid 90s, so I suppose the labels were somewhat less offensive at the time? Nonetheless, in my usage of the models, I have changed the names to remove this particular concern. For example, Hostages are referred to as False Loyals. We discussed this very topic on the Roundtable call, with general consensus. We decided, for conversation purposes, to just refer to the original segment names.
2. This is a great question, too, and I think it's implied in their repurchase intent in that they will purchase again and again and by default also recommend. Not sure. Having said that, I have had clients whose primary objective was referrals and, hence, used "likelihood to recommend" as their loyalty question (y-axis) instead of "likelihood to purchase again."
3. We talked about this on the Roundtable call, as well, and we felt it applied to both B2C and B2B companies. In my experience, B2B companies have been more open to it. I've found that it works well at an account level, for example, plotting accounts on the grid.
Hope that answers your questions. Let me know if there's anything else!
Annette 🙂