By Rowan Jackson
Not everything that can be counted counts.
Not everything that counts can be counted.
William Bruce Cameron
19th century shopkeepers knew that their business would thrive if they met their customers’ expectations. They knew that knowing their customer was essential, as was the shopping experience.
In the 1990s, Fred Reichheld, in his famous book The Loyalty Effect, quantified what shopkeepers knew 100 years previously. Increasing customer trust, advocacy and loyalty by 5% year on year increases profits by between 20-125% depending on the industry.
The result; an explosion of measurement. In the 1940s, organisations started using satisfaction as a measure. In 1995, Professors Tom Jones and Earl Sasser at MIT revealed what many knew. Satisfaction was not a reliable predictor of loyalty. Yet, despite this, many still use satisfaction surveys today to measure their customer experience. This seems to be yet another version of the famous Einstein quote: “It is a peculiar sort of insanity that thinks by repeating the same method you will get a different result.”
More recently, we and others have researched CEOs and senior executives what they wanted and did not want from their measurement processes. There are no surprises on their list. “Please give me:
- Insight that identifies and prioritises precisely where my customer wants my business to improve. Not many insights, just the vital few.
- A method that, like a doctor with a patient, diagnoses what needs fixing.
- A method that reliably predicts loyalty.
- A method that tells me how I stand with respect to my competition.
- Insights presented in a way that is easy for non-experts to understand and to communicate to their people.
- High quality analysis based on proven methods.
- Insight driven by the customer, that is strategic.”
Their list of what they didn’t want, but were often given, was equally interesting. “Please do not give me:
- Vast amounts of unsorted data.
- Internally driven, not customer focused data.
- Data for compliance requirements only. It is usually only the minimum acceptable level, and we want to go better than that.
- Tactical low level stuff!
- Satisfaction data: It does not predict loyalty.
- A long list of un-prioritised actions.
- Anything that does not diagnose what I need to do, like NPS.”
So here are a few guidelines:
1. If you wish to measure trust, advocacy and loyalty, then it makes sense to use the one method that drives these outcomes; the degree to which your organisation meets your customers’ expectations. Expectations matter and should be measured and not satisfaction.
2. If your measurement system is diagnostic, then it will tell you what to do in a prioritised order.
3. If your measurement system includes measurement of competitors, then it will tell you how they perform as well. Most of your customers have experience of your competitors too, so they can tell you.