Perfectly executed and statistically valid survey results won’t help you unless you actually do something based on the results. The trick is to design the survey and your business processes to enable action. Here are a few tips for doing so.

What questions will help us to determine results and make changes?
This is the most important of the three: considering what questions will best help you to improve the program, service or product about which you are gathering data.

If you offer a training seminar, is the satisfaction of the attendee during the event the most important aspect or are the improved outcomes they create on their job using this new knowledge? Probably the latter, although it is much more common to survey the former.

Who can best answer those questions for us?
Following what to ask is then who to ask it. Continuing the example above, if the key metric is improved job performance, then the right people to ask about that are probably the supervisors or managers of the people attending the event rather than those who were there.

It can often be extra work to identify and contact those who can actually answer your questions but it is critical if you want data that is valuable and enables you to take action.

Build in time and resources for analysis and action.
Even when you ask the right people the right questions, you won’t get anywhere unless you build in time and resources to really go through the results.

Who can make decisions about what to do? Who can best analyze the results and recommend alternatives and changes if warranted? Put time on their calendars before the surveys goes out so they will convene to review results and make decisions.

If you do the above you’ll be much more likely to improve the quality of whatever it is you are surveying about.