And neither should you.
In CX, we love to talk about our survey results. We report the latest Net Promoter Scores, or show how we did in satisfaction last quarter. Yet, as I talk with CX practitioners, I hear about how it’s often hard to get the business to care.
And, really, why should they? Satisfaction and likelihood to recommend are just our research-created constructs. We know that satisfied customers defect, and that those who say they will recommend you often don’t.
The best relationship surveys take this a step further by showing a driver analysis, showing which factors most improve the dependent variable of satisfaction or likelihood to recommend. For those not familiar with driver analysis, it’s too much to cover here, but it essentially shows which of your many survey questions most impact the top score. It’s unfortunately neglected by many, but it’s critical. However, it still suffers from trying to predict the NPS or Satisfaction scores, rather than true business outcomes.
As an example, one client asked us to test six different metrics, from satisfaction to likelihood to recommend to other more emotional outcomes. What we discovered shouldn’t be too surprising – the top drivers (and thus our recommendations) varied substantially depending on which question we used as the top-line score. What drove satisfaction was very different from what drove likelihood to recommend, which again was very different from more emotional scores. What was the business to do?
There needs to be a better way. And there is.
If you’re struggling to engage the business to create customer-focused change, perhaps the change needs to start with you, and what you’re reporting.
Forget about starting with satisfaction or likelihood to recommend. Instead, show the gain or loss of customers or other critical business metrics as your top variable. Then show which of your other questions most drive this.
By starting with a business metric familiar, you can better engage your business partners. Then go ahead and tie in satisfaction or NPS if you must. But a better idea is to focus less on these, and instead show a driver analysis on the business metric itself.
Imagine in your next reporting session if you were to begin with share of wallet. Show the share of wallet from your survey respondents. Then show what most drives this. Skip satisfaction and NPS altogether.
Or start with revenue changes from the last reporting period, and show that X, Y and Z are the top drivers of increased revenue. Or perhaps number of orders per month, profit per order, or something else that your business cares about. That will grab their attention, and lead them to want to know more.
This isn’t easy. Business metrics are messy, and researchers often aren’t comfortable using messy data from external sources. Who said your job was supposed to be easy?
If you’re struggling to engage your business, it may be because they don’t really care about your metrics. They care about the metrics their bosses ask them about, such as revenue, gained or lost customers, or share of wallet.
Try an experiment. Start your next report with the business metrics that your internal customers use to report to their bosses.
I guarantee they’ll take notice.
