Conduct Your Own Customer Satisfaction Survey


In these customer-focused times we often hear of how important it is to know more about our customers. One of the best ways to do this is to conduct a customer satisfaction survey, but it has to be done the right way to get the information you need.

The best way to begin is by identifying the elements in your business that create customer satisfaction. This includes such things as your products, your sales team, your warranty and even your business premises. These things will become the focus of the questions in your survey.

Next, think about what it is you want to do with the answers. What do you want to accomplish with your survey? Be as specific as possible. There’s no point in asking questions that won’t lead to actionable results. Have a targeted use for every bit of information gathered.

Review any information you have from previous surveys. Maintain continuity in your questions from one survey to another so that you can detect trends or attitude shifts.  If you decide to modify the wording of an earlier survey be absolutely certain you’re still getting at the same thing.

Questions should be framed objectively and not lead the respondents towards a particular answer. You’ll need to decide whether the format will be open-ended (‘What do you think about our product range?’) or simply require a ‘yes’ or ‘no’ response. Open-ended questions generally take more time to answer so this format restricts the number of questions you’ll be able to ask but can provide a lot of valuable opinion.

Keep questions as short as possible and avoid asking compound questions – ‘Where do you shop and what time of day do you shop?’  Each bit of information you want should be the subject of a separate question. Don’t use industry jargon or slang; it will only confuse respondents and can lead to inaccurate answers.

Now decide on the actual information you’ll need to collect. If you have more than one place of business you’ll need to find out which one a particular customer patronises to know if their answer to ‘Are you satisfied with our service?’ is related to factors that apply to just one outlet. Depending on your products or services it might be useful to know if there are different views held by male and female customers, or by different age groups. This is why surveys usually start with a few questions to establish the demographics of the respondent.

When you’ve completed writing your survey questionnaire give it a trial run on people who can give you feedback on such matters as how clear the wording of each question is and whether you’re asking any questions that can’t be answered conclusively.  See how long it actually takes to complete the survey; you’ll probably find that it takes longer than you anticipated, and surveys that take too long often get rushed through or left incomplete.

It’s unlikely that you’ll be surveying every customer unless you have a very small number to work with. Generally a survey is run on just a sample group taken from your total customer base. The number of completed surveys you will need to get a representative opinion depends on several factors but somewhere from 50 to 100 responses can be adequate for a total customer base of from 1,000 to 10,000.  The more you have though, the better.

If you’re going to send out survey forms to be completed and returned by mail then make it easy for them to be returned to you. Use a standard envelope with the postage prepaid; even better is a form that can be folded and sealed so no envelope’s needed at all. Avoid placing any commercial messages on survey forms and envelopes.

If you’re conducting your survey in an email you’ll have to be sure it doesn’t get confused with spam or other unwanted messages. The sender should be clearly identified and the subject line unambiguous – ‘ABC Co. Survey – Please Complete’ for example.

Telephone surveys especially have to be kept brief. Nobody likes being tied up on the telephone answering questions for half an hour and it’s always possible your respondent will be interrupted or called away, thereby leaving your survey incomplete.

It never hurts to offer some sort of incentive to respond to your survey request. A small reward for replying, like a discount off the customer’s next purchase or perhaps an entitlement to be entered in a drawing for a prize, can really increase the response rate.

Confidentiality is becoming increasingly important to many people and this can affect people’s preparedness to respond to surveys. Make it clear that respondents will remain anonymous and don’t have anything that looks like an identifying code on either the questionnaire form or its envelope. Offer a means of contacting your business if they want to verify that the survey is legitimate.

Remember too that people are more likely to give objective answers if they feel they’re talking to a third party instead of someone they’ll be dealing directly with in the future. It might be best to use a firm of survey specialists if the information you want to obtain is sensitive or highly personal.

When the results are in you should try and provide some closure for the respondents by providing them with a report. Or you can put it out to your customers in general. There’s always some advantage to be had from this. If the results are all positive, that’s good – but even if there are negatives you can get up points by promising to address them in the interests of providing an improved level of service.

Until next week,
Mike Reddy