In my experience, most surveys I’ve taken and seen (includes both those I’ve been asked to provide input on as well as those I couldn’t bear to actually complete), don’t really get actionable information as a result.  I get the sense that most people think surveys are easy to do and really, how hard can it be to come up with a bunch of questions?  In a way, they are right.  It’s not usually the questions that trip you up (although there are those and we’ll get to that in a bit).  The hard part, the tripping point, is really purposeful information.

1. If you had the answer to this survey question, what action would you take with it? 
2. Pretend you’re the opposite of a lawyer and don’t ask any questions when you already know the answer. 
3. Would you take this survey if you got it? Really?
4. Avoid conjunctions.  I don’t know why this comes up so often.  Conjunctions don’t conveniently shorten your survey by grouping concepts, they cloud your results.  (And, but, or, yet, for, nor, so).  “Please rate your satisfaction with the cleanliness and quality of our bathroom.”  Really, is cleanliness the same as quality?  Perhaps it was clean but the toilet paper was of poor quality… but I digress to another potential blog here.
5. Don’t be so wordy.  Overdone corporate speak, really long questions and acronyms are the usual culprits here.  Keep it short and simple to get the best results.
6. Have at least one open text/feedback question.  Instead of trying to cover every possible base, leave the customer the chance to share what they want to share, the way they want to share it.

And my last thoughts on the manner – keep the whole survey short and focused on your point.  Tell me how long the survey is or will take in the invite.  There’s always more but these are the things that always seem to come up in one form or another.  Now you tell me – how many times have you seen these simple rules violated?