The best L&D, HR and OD pros understand the strategic value of information gleaned from employee surveys and know how to put it to great use when planning and executing their programmes.

However, achieving good response rates and getting the right kind of information requires careful planning and great communication.

My ROI (return on investment) work includes assisting organisations to develop and roll out employee surveys.  I’ve seen first-hand what works well and what can cause problems so I wanted to share some of these insights with you.

Be clear about what you want to achieve by surveying

It sounds like a no-brainer, but it’s amazing how often surveys are run with no real thought given to how the resulting information will assist the organisation to achieve its strategic goals.  Knowing this will help to ensure questions (and responses) are strategically useful. For example, is the ultimate goal to improve employee engagement and therefore productivity, or is it to ascertain how well communication is flowing throughout the organisation?  Each is a perfectly valid reason to survey, but there will be differences in the questions asked in order to ascertain this information.

What do you really need to know?

Surveys often fail because they’re too long. When I’m working with a client to develop a survey, we hone in on exactly what we need to know and build our questions around that. The question ‘Is this essential or just nice to know?’ gets asked a lot at this point.

Once questions have been drafted then we think about how to make it easy for people to respond. Most people will likely skim-read the survey, so we boost navigability with clear sections, headings and instructions.

Design questions to yield useful data

Avoid non-committal or vague answers by eliminating halfway house responses – the most obvious being the mid-point ‘sometimes’ or ‘neutral’ option in a Likert scale. Although many believe it is best practice to use a 5-point scale, my preference is to have an even number of options to prevent fence-sitting.

Avoid ambiguity

Confusing questions result in people dropping out of surveys. Keep questions short (fewer than 15 words per question) and test them on colleagues to make sure they can’t be misunderstood. If I’m concerned that people won’t give the level of detail needed, I sometimes give a sample answer as a model.

‘Double-banger’ questions can also result in ambiguous answers because people may have different responses to each part of the question.  Here’s an example of a double-banger question:

“On a scale of 1-5, how fairly and equally do you feel employees are treated?”

How could someone provide a ranking if they felt people were treated equally, but all were treated unfairly?

Carefully check to make sure that each question contains only one question.  Seeing the word ‘and’ in a question should warn you that it could be a double-banger.

Sense-check your response options

Ensure there is a logical alignment between the question asked, and response options given. It is surprising how often mis-matched questions and response options find their way into surveys.

Here is an example of a confusing set of question / response options:

Example Statement: “I often access online modules.”

Choose one option from the following: * Very often * Fairly often * Sometimes * Never

A more logical question would be:  

Which of the following options best describes how often I access online modules?

Choose one option from the following: * Very often * Fairly often * Sometimes * Never

Pilot testing a survey on several people before it is rolled out will help to catch those confusing questions.

Keep it confidential

One of the many benefits of online survey platforms is the reassuring sense of anonymity they offer. And if you’re doing it the old fashioned way by gathering information on paper, it is extra-important to assure people that you will keep their responses confidential. I’ve found that giving people a postage-paid return envelope boosts response rates in this situation.

Make it worth people’s time

It is common to offer rewards for completing surveys – but you may not need one if you have made the survey easy for people to complete. If you do decide to offer a reward, then consider splitting your budget between several smaller prizes to increase individuals’ chances of winning. If the odds of winning are high, we let people know – most people enjoy a flutter!

Say thank you

The words ‘thank you’ mean a lot when you’re asking for people’s time. We always include at least one thank you within a survey and thank people again via the methods we used to communicate the survey.

Communication matters

I always remind clients that it’s no good having a great survey if nobody knows about it.

It is important to sell the survey’s benefits, just as would be the case for any other initiative. For example, make respondents feel special by explaining why their comments are valued and let them know how their ideas will make a difference. If possible, use local managers to spread the word and encourage responses. Depending on the audience, consider including a note from a senior manager in survey communications.

Then, once the survey is underway, schedule reminders to prompt extra responses.

Find the best combination

I love online survey tools such as SurveyMonkey and SurveyGizmo.  Even the free offerings often provide enough flexibility to gather detailed information and easily collate the findings.

I hope you have found these tips useful!

Book Recommendation

Finally, I recommend the Survey Basics book co-authored by Drs Jack & Patti Phillips of the ROI Institute® and Bruce Aaron that can be purchased here.