Skip to main content
grow: mark healy

Two weeks ago I wrote about approaches and techniques to deal with typical weak spots in customer surveys. There were many comments about the poor quality of customer surveys received on behalf of all manner of organization.

Customer surveys can be powerful tools for small and medium-sized businesses trying to better understand any number of issues, from customer experience and brand association to loyalty, satisfaction and quality. However, they must be intelligently designed and well executed.

We all have experience with the poor surveys Globe comment writers discussed. It's not that putting together a good survey is hard, it's just that – like anything else – there is a process to it, and attention to detail really matters. Here are some ground rules for building a solid survey and driving above-average participation.

The process: Construction

The process underlying a solid survey build is straight-forward, but rarely followed.

1. Identify the target audience.

2. Select the medium of distribution.

3. Determine, specifically, the goal of the survey.

4. Sort out the three-to-five metrics – things you want to measure, or graph, or report on later.

5. Write the questions.

This is where people usually go wrong: they start by writing questions. That is like flooding your backyard before levelling it or putting up boards. You'll get a rink but it won't be any good. Here is an example of a good consumer survey construction process.

• A restaurant chain wants to better understand how to appeal to youth. It would first define "youth" – age 6 to 18, 14 to 21?

• Next up it would pick the medium that makes the most sense to get at that segment. If the 14- to 21-year-old set is selected, an online survey probably makes more sense than an in-restaurant deployment.

• The first meaty decision is to really understand the goal of polling customers – in this case it might be to understand what marketing tactics (not what menu items) would draw in this crowd.

• Metrics are next – determining what really needs to be measured. In this case, the chain might select TV advertising effectiveness, contest effectiveness, social media effectiveness.

• Finally, questions would be crafted to address each of the metrics. For example, for contest effectiveness, two potential questions could be: 1. If we offered free "dinners-for-two" as prizes in random draws, would you be more likely to eat here more often? Yes or no. 2. On a scale of 1 to 4, how much did our X-Box contest influence your decision to eat here? (We prefer four-point scales with a N/A option over five-point scales since a large count of "3's" really doesn't mean anything.)

The rules

There are 10 simple rules to building and launching a good customer survey:

1. The first rule of fight club is: do not talk about fight club. Never use the actual word "survey" or "questionnaire" in any of the communication about the survey or in the questionnaire itself – potential respondents will turn off immediately. Spam filters also object to these questions. ("Please take five minutes to provide us with your input on…" or something similar works better.)

2. Never start a survey with demographic questions – it is off-putting to some – ask these questions last.

3. Start with the most relevant questions – if customers drop out after four or five questions, at least you have data for the questions you really care about.

4. Always start a survey with a question that will likely elicit a positive response – a yes or a good news answer – this puts people in a positive frame of mind and you are less likely to have people drop out of the survey.

5. Keep questions short and punchy, ensure they are unambiguous, and that each question goes after only one thought or idea.

6. Answer options are as important or more important than the questions themselves. If topic-driven (price, quality), ensure some research is done ahead of time so that options are well informed, and also include an "other" category. If it's a numerical scale, ensure it fits the question (0-10 for satisfaction, 1-4 for degree of influence). Open ended (no answer option) questions can be hard to code or interpret later but often elicit the best answers.

7. Put a note at the top of the survey or in the communication that goes out with it from the top dog (president, CEO, head) appealing to the reader ("help shape the future of your YYY…") – this increases response rates by as much as 50 per cent.

8. Online customer survey respondents become bored and frustrated after as early as the 10-minute mark. Keep surveys short, and advertising up front that it "will take three-to-five minutes" helps.

9. Categorizing questions by topic is a convenient way to organize questions during the writing phase. However, arranging these questions to form a story helps respondents see the purpose of each question in relation to the whole survey.

10. The response rate is always higher on the weekend for online surveys – so launch on a Thursday or Friday to achieve the most rapid initial response.

Note: Points one to five assume you have the screener questions out of the way (questions that qualify the audience you want for the survey – for example, only primary shoppers aged 16 to 21).

The process: Deployment

Again, there is a process for "launching" good surveys. Typical response rates for consumer surveys are in the 2-per-cent to 5-per-cent range. I've seen response rates driven as high as 40 per cent though smart planning and execution:

1. Get a second set of eyes on the survey before distributing anywhere. Any mistake – big or small – will lower participation and could skew the data.

2. Test the survey. Carve out a few test participants from your survey audience and send it to them first – then see how questions are answered and ask test participants for feedback on length, format and wording.

3. In parallel with No. 2, get in touch with the participant audience and let them know a survey is coming and why it is important their voice is heard, how it will benefit them later. This is the crucial step in driving participation rates above the average.

4. Launch the survey.

5. Send a follow-up communication to the participant audience, imploring participation, to get one more bump or spike in completes.

What you do with the data once you've got it is a whole other story. But putting together a tight survey is the first step toward getting good data. Done well, surveys can unlock deep insights and inform important decisions.

Special to the Globe and Mail

Mark Healy, P.Eng, MBA, founder of Torque Customer Strategy, is now a partner at Satov Consultants – a management consultancy with practice areas in corporate strategy, customer strategy and operations strategy. Mark's focus areas inside the Customer Strategy practice include consumer insights, customer experience, innovation and go-to-market strategy. He is a regular speaker and media contributor on topics ranging from marketing to strategy, in telecom, retail and other sectors. Mark is known as much for his penchant for loud socks and a healthy NFL football obsession as he is for his commitment to Ivey and recent Ivey grads. He currently serves as chair of the Ivey Alumni Association board of directors. Mark lives with his wife Charlotte and their bulldog McDuff in Toronto.

Report an error

Editorial code of conduct