Once you’ve decided to start asking your customers for feedback via a survey you really took a big leap, great! Apart from a customer-centric approach though, what you really want to get from it is reliable data, not to mention you don’t want the survey to backfire and annoy your customers for some reason. So what should you avoid if you want to craft a proper customer satisfaction survey?
While working on improving customer surveys with our clients at Casbeg (and being on the receiving end of customer surveys many times before 😉 ), I’ve listed the most common mistakes that seem to be made over and over again, regardless of the industry or company size.
1. No recipient segmentation or wrong segmentation
Keeping in mind the main goal of surveying your customers – reliable data and feedback – don’t go and send your questions wherever and to whomever. Segment your customer base and decide which segment is actually READY to provide you with answers. Set a milestone of delivering your service or using your product and use it as a trigger to reach out. Next think of any other criteria that might shed more light on the replies you’ll receive – if your customer base is rather mixed, try segmenting it by what is important for your business, like industry, size, client/ user type and so on.
Many companies ignore their churned clients while asking for feedback. While this segment of course will be less responsive, it’s still very important to get separate feedback from them. Limit your recipient list according to their last activity and survey those that still remember doing business with you.
I’ve seen some companies make the mistake of segmenting their customers not by objective market criteria, but rather by their own sentiment, excluding either the ones that they “know will give positive feedback” or “already know they’re unhappy”. Don’t be that company and let the truth hit you.
2. No personalisation
The more replies you get, the better the chances for collecting reliable data and making the right decisions for your business you have. Avoid sending template questionnaires with no feeling of brand familiarity, add at least colours your customers will recognise and preferably your logo, too.
First though you need to do your best to turn recipients of your survey to become actual respondents. In most cases, using their first name or even their company name in either CTA or even invite title can work miracles, so show them you care about their answer and use the data you already have at hand to personalise your surveys with custom fields.
3. Asking too many questions
A perfect survey, according to last year’s report, is between 2 and 6 questions. Short and sweet, that’s all you need. Don’t create a monster survey your clients will get scared off by, same goes for pagination – it’s better to send a concise survey with the most important questions you need answered, designed to take just a moment of your customer’s precious time. Respect their time and effort and don’t present them a colossal document, this will also ensure a better response rate.
It’s also a good practice and good manners to tell your customers how long the survey will take – this increases your chances for a better response rate because you’re telling them exactly how much time they’ll need to give you and most of them will respond right away.
If you, however, for some reason, really do need to ask a long list of questions, you could opt for cutting the list into sections and send those to random samples of your customer base, as long as it’s big enough to ask one set of questions to a valid sample of about 100 customers.
4. Asking questions you already know the answer to
Since you have to limit your survey’s length anyway, try to make the best of it and get rid of any questions you should know the answer to, otherwise your customers will not only most likely ignore your survey and close the browser window, but will also feel ignored themselves.
Typically such questions involve the recipient’s personal data or other information regarding the industry they work in, etc. – to avoid asking those questions, segment your customer base first to properly interpret answers to valid questions you have to keep.
5. Ambiguous or leading questions
The questions you ask should always be objective, singular and clear. Double-check with someone on your team if all makes sense if the wording is understandable and doesn’t sound suggestive or misleading. The most common example of wrongly asked questions I encounter are the ambiguous options, like “How satisfied are you with our services/ cooperation?”. Which one is it then? The recipients shouldn’t have to second guess, so make sure they don’t.
The second case I often see are the leading questions, where the respondent is somewhat pressed into giving their answer in a particular way, like What would you like to thank us for? In this case, the company not only assumes positive feedback only but is also asking for this kind of question to backfire.
There is only one situation that comes to mind when it is not only allowed to ask a leading question, but actually recommended to do so – What should we improve?
Being humble and open to suggestions will help you grow, and customers will appreciate that.
6. No open-ended questions
Regardless of whether you opted to send out a score-based customer satisfaction survey with only one question or perhaps a list of few questions, you should always ask your customers to share some open feedback and give them a chance to justify their score answers.
Sometimes though analysing long paragraphs of a large customer base might get a bit overwhelming, so just use a simple trick and ask for a one-liner, like Please explain your answer in one sentence.
7. Asking too often
Unless you’re aiming for the highest score in annoying your customers, don’t overdo it and schedule your feedback rounds timely. It’s really enough to reach out every few months and not weeks or days – give your customers time to build more experience of working with you and be patient. Sending satisfaction survey too often might be harmful to both your company image and the actual cooperation with the respondent, nobody likes being spammed.
8. No thank you page
Survey participants give you their time and invaluable feedback, so be nice, show you appreciate their effort and say thank you. This will end the process in a clear way and let your customers know that their feedback has been collected, plus give your customers some positive experience with your business. Most tools (even the evergreen Google Forms), let you very easily set up and customise a thank you page – use it!
9. Random scales for score questions
Before you send out anything, read up on customer satisfaction survey methodologies and stick to them. You want reliable data and you want to be able to interpret the results in a correct way afterwards, so make sure your scales are accurate and you know how to calculate the final score.
The two most common score questions related to customer satisfaction are CSAT and NPS and even though there’s a ton of information on both of them online, many companies still make mistakes while using them or try to somehow mix them up. Don’t get creative here, survey methodologies work for a reason.
