Every quality management framework for human services in Australia has a standard / indicator / expected outcome relating to feedback. Organisations will generally realise this through a compliments and complaints system, and, of course, through surveys. Surveys can occur during any stage of the service delivery process; usually clients are surveyed when they exit, and at timed intervals during their service, e.g. annually.
But a common complaint about surveys is that nothing ever comes of them—management might give an overview of results, but there is often little action or change. In an environment where clients are continually (and increasingly) asked for feedback (both internally and for external audits too), being able to demonstrate that you really care about what they have to say is a vital part of a customer service culture.
Being able to get results from surveys that you can action goes back to how the survey is written. All too often, management will write or release a survey without knowing what they will, or can, do with the results. Before you start, first ask "what exactly do we want from this survey?" Here are some guidelines I use when designing a survey:
1. A targeted survey is more valuable than a generic, broad one. Look at your strategic and/or operational plan for inspiration on where to target your survey. Complaints or other feedback data can also be helpful.
2. Ask yourself what you really want to know—do you want to know if your service is achieving its purpose? If clients are happy with service times or how long they have to wait on the phone? Asking something like "do you feel that the service has helped you maintain your independence" (and following up with "if no, what could we do better") will get you a more useful response than "are you satisfied with your services".
3. Make sure you know your critical limits—or, what is a "good" response to you? If you're asking satisfaction questions, is a 70% satisfaction rate good? Or is 85% what you really want? What is the most important question, and what is the least? Defining this as you design your survey will help you respond to what is really important. Make sure you document this: "if the satisfaction rate for question x is below 85%, management will initiate a review of the x process".
4. Get a good sample—if you have 200 clients and 30 respond, you don't really have a good enough sample to be assured of the results. Remember that you have to factor in outliers (people who feel strongly either way and can skew averages) and incomplete or incorrect surveys. If you're not getting a good response, extend the survey and follow up personally.
Our clients and staff time is precious, and I do not believe in surveying for the sake of meeting a quality standard or any other reason besides a genuine desire to learn, engage, and change. Doing something with your survey results is a sign of respect. And (nerd alert) a beautifully designed survey that gets you results that you can actually action? Well, it's truly an exciting and invigorating thing.
I would love to hear about good or bad surveys and what you've learned; please drop me a line or comment.
Thanks for reading,
The Quality Nerd loves all things Quality Management and Internal Audit...too much is never enough!