When you need numbers.
WHAT SURVEYS ARE GOOD FOR
Measuring how common something is, once you know what to ask.
WHAT THEY ARE BAD FOR
Discovering what matters.
WHY
You can only ask about what you already thought of.
WHAT TO DO FIRST
Conversations, then a survey to measure what they revealed.
WHAT MAKES A SURVEY USEFUL
Short Clear questions Answer options covering the real range Sent to the right people
HOW SHORT
Few enough to complete in a couple of minutes.
WHY
Completion falls sharply with length.
WHAT MAKES A QUESTION BAD
Two questions in one Leading wording Options that overlap Options that miss the real answer Assuming knowledge the respondent lacks
WHAT TO INCLUDE
An option for none of these, and a free-text box.
WHY THE FREE TEXT
It is frequently the most useful part.
WHAT TO BE CAREFUL WITH
Who actually responds.
WHY
People who respond differ from those who do not, sometimes substantially.
WHAT THAT MEANS
Results describe respondents, not the population.
WHAT TO DO ABOUT SMALL NUMBERS
Treat them as indicative, not conclusive.
WHAT TO AVOID
Reporting percentages from very few responses.
WHAT TO TEST BEFORE SENDING
The survey, on a few people, watching where they hesitate.
