Good Surveys start with Good Goals
Posted by Jeffrey Henning on Thu, Apr 09, 2009
Don't let your survey project go astray from the start: be certain to focus on a specific goal. Be precise about what information you need to gather and what you plan on doing with it.
If your organization hasn't done a survey in a while, the tendency is for every department to chime in with questions they want you to ask. The result: questionnaire design by committee, taking up more of your and your coworkers' time to prepare, then producing a survey that is too long and tedious for respondents to complete quickly.
By setting a narrow goal you will be able to relentlessly simplify the survey and keep it on target. To do this, talk to the stakeholders who will use the survey data. Develop an understanding of their wants and needs. What are the specific decisions they need to make? What information do they currently have? Where are they uncertain about what the target audience believes?
Once you have come up with a list of goals, ask yourself whether a survey is really the best way to gather this information.
- Can you meet the goals without doing a survey at all?
- Are your potential respondents the only source of this information?
- If you work for a large organization, is a coworker in another department doing a survey on this topic or researching this issue?
- Do your CRM, web analytics or other systems hold data that would help you reach your goal? You can ask web-site visitors their favorite sections of your web site, but using web analytics to study the traffic volume for the different sections will provide you better data.
To help you, here are some examples of bad goals and better goals.
| Bad Survey Goal | Better Approach | Improvement |
|---|
| Maybe customers would like it if we provided support after 6 p.m. | Prioritize ideas for expanding customer service. | Don't ask about one product or service improvement in isolation. You need to compare it against other possible ideas to determine the best initiative for your organization. |
| Find out how satisfied customers are with customer service. | Find out how satisfied customers are with customer service and intervene with unsatisfied customers to address their issues. | Don't approach customer satisfaction as simply a strategic concern, a matter of measurement, but use it tactically as well to identify individuals who are unsatisfied in an attempt to fix their problems and satisfy them. |
| What are our customers thinking right now? | Instead of doing a survey, conduct one-on-one unstructured interviews with customers to determine their current attitudes. Use the results of those interviews to determine other research needs. | Surveys are great for structured information gathering that you can extrapolate to your target audience (quantitative insights-see sidebar). When you are looking for qualitative insights instead, just talk to customers. |
| Let's create a committee and see what they want to find out from a customer survey. | Instead survey the employees about the customer information they are currently lacking and use that to determine the priorities for your customer research. | Committees can be driven by individuals with pet-projects; a survey of employees will help you develop an assessment of how many people need to gather specific types of information. |
| Determine how often callers into help desk check the customer-service portal before calling. | Record customer-service portal log-ins and knowledge-base searches into the CRM system, so that representatives can see the searches as soon as they identify the caller. Run reports on the percent of knowledge-base lookups that lead to calls and track that over time. | Actual usage is more reliable data than self-reported assessments of usage through a survey, and customer do not need to be interrupted to take a survey. |
I'd love your examples of bad goals you've seen for survey projects over the years.