How to Conduct a Survey – Analyzing the Data
If the earlier steps were done well, analysis is where the payoff begins. Your responses are now numbers, and their distribution can be plotted to show where most respondents land on the scale. The question is what to do with that data — and how not to be misled by it.
Compared to What?
Seventy percent of your customers liking your customer service sounds like a win — until you learn that 92% of your competitor’s customers like theirs. Without a benchmark, you cannot judge whether a score is good or bad. NBRI’s ClearPath Benchmarking© exists precisely to give you that comparison point against your industry.
Are You Sure That’s the Issue?
Knowing customers dislike your service doesn’t tell you why. Root cause analysis is how you find the small number of issues that, when fixed, move the most other results. But correlation is not causation: ice cream sales and swimming deaths both rise in summer, yet neither causes the other — summer heat drives both. The same trap can lurk in your own survey data, so take the time to think through whether an apparent driver is really the cause, or just a symptom of something else.
Every business and every survey is different, so no single article can catalog every pitfall. Take your time, involve other perspectives — department managers and coworkers often see things you don’t — and re-survey to clarify a question if you need to.
NBRI Research Team
National Business Research Institute
























By submitting this form you agree to our