Imagine if your HR partners told you they were going to use one metric to measure all of your employees, and that should be how you determine each one’s overall effectiveness.
You’d push back immediately.
You’d say that your software developers should have different measurements than your salespeople, who should have different measurements than your trainers and your customer service reps.
And you’d be right.
So, why do we think all companies should use the same customer measurement?
Spoiler alert: They shouldn’t.
Even within one business category, companies are creating very different experiences, and therefore should measure differently.
Consider your desired outcome
Publix and Aldi deliver very different experiences. At least prior to the pandemic, Publix looked to inspire shoppers with their food demonstrations and samples. With or without a pandemic, you won’t find those at Aldi. And while I don’t know exactly what emotions Aldi is looking to elicit, I doubt “inspired” is one of them.
Publix and Aldi – and Whole Foods and Kroger and Wal-Mart and Target – create different outcomes. So measuring the customer experience in the same way is ridiculous. (Not to mention an ineffective use of time and resources, especially when you consider what you could be measuring and learning instead.)
In addition, there are times when using NPS’s “likelihood to recommend” question makes no sense.
There are times when using NPS’s “likelihood to recommend” question makes no sense.
Here’s just one example: We’ve worked with direct competitors in the life insurance space, and both have used this question for all audiences. The NPS score was very low when we ran the survey for annuity customers, but when we asked a follow-up question – “Why?” – respondents told us, “I don’t normally talk about life insurance companies to my friends.” Given that, NPS makes no sense.
But when we look at the financial professional customer, the story changes.
For both of those insurance companies, revenue is completely dependent on their financial professionals recommending them. So using this NPS question makes intuitive sense. But you still need to test it.
Connect the dots
A number of years back, Capital One 360 looked at who referred customers to them. They have a referral program, and they linked the results with their two survey scores: “How likely are you to recommend us?” and “How simple was your experience?”
They expected the likely-to-recommend score to predict referrals. (Wouldn’t you?) Instead, they found that simplicity was a far better predictor of referrals than likelihood to recommend!
Use three steps when deciding what metrics to use to track your program:
- Identify the outcome you are trying to create in your customer experience. For an annuity customer, the top priority probably isn’t recommendations (although both companies would take that). The more likely ideal outcome is for customers to keep their account active (vs. closing it out) and to add more funds. That requires confidence. So customers indicating “this company has my best interests at heart” will likely be a better metric than likelihood to recommend.
- Test that ideal metric – along with a few others – to see which best predicts that outcome. Do customers who have stayed with you a long time and added funds to their accounts give higher scores on this metric?
- Socialize the results. Proactively help executives understand why this metric makes the most sense for your company (no matter what some speaker at a conference said).
Doing this will not only mean you’re measuring something that matters, it also makes it easier to show how your work is creating impact. And that’s the most critical outcome for any CX program.