It’s quite obvious why you should consider your customer’s feedback. You want to take the right decisions and you want to know if you’re still on the same page as your users. Great! Of course there are a ton of ways how to gather your customer’s feedback: it can range from a survey to in-depth interviews to product usability tests.
For ways how to test your product, e.g. your website, check out our recent post on online testing tools. Assuming you do not have a high-digits budget to conduct market research through a service provider, here are five practical steps how you can receive structured customer feedback. (Market researchers please do not continue to read…)
1. Define clearly what you want to do
Maybe you know this situation: something happens which changes your direct environment like dropping numbers, a competitor’s move, a new technology. The situation could be complex, like is there a need to change your product features, the appearance, the price? So you decide to ask some customers which could show you the direction to go, because they’re your customers, right? They know what they want.
In order to have a significant statistical result which “indicates” you a direction, you would need quite a lot of usable answers which you can only get by online or telephone polls. And it depends on your industry, your customer base or the incentive, but don’t expect more than 5-10% of reached customers to respond. On the other hand, if you are asking just a few random customers for their opinion, you will face the issue how to interpret the responses.
So before you are about to pick up the phone, be clear on the following:
- The Why: Describe in one sentence why you want to get customer feedback. If you do not know why you are asking for it, most likely the feedback gathering will explore many interesting insights in many different directions which leaves you more clueless than before for the actual issue at hand.
- The Stakeholders: Define for whom you want to have feedback. If it’s for your product team, explicit non-standardised answers give the highest insights. If it’s to prove ROI and to share with management, a statistical significant number-based result is essential.
- The Who: Define from whom you want to have feedback. Is it your total customer base, so the surveyed group should have the same structure or a sub-segment which can only be reached through a specific communication channel.
- The What: Define what you would ask customers. Is there any other way how you can get the information, e.g. by user data or transactional data. Maybe you can avoid asking your customer something obvious or what he does not know actively.
- The What Then: What will you do with the results? Even though this is the last step, it is important to know what you will do with the results. Are the results important for a specific milestone? Do you need immediate indication or can you wait 3-4 weeks until a regular survey is set up, conducted and analysed? Are the results input to a concept? Do you require more open answers in order to generate ideas and broader understanding?
- The Impact: What is the impact of the customer feedback? Will it facilitate a high-budget project or might affect a few only? This helps to identify the budget and efforts you want to put into gathering and analysing the feedback.
Having answered all these questions leaves you with a framework which avoids any distraction from the issue to be solved and which will help you to define the following steps.
2. Define what to survey
In general you can survey the overall product or service, specific features, appearances, usage and so on (usability and live tests excluded here). Within the conceptual phase you might want to survey the current status quo to identify gaps or improvement potential. You can conduct quantitative or qualitative, online or telephone/in-person surveys in order to learn more about your customer’s behaviour which is not shown in the data. You can explore the “why” customers are using or not using your products or services. In the product development phase you have already set your targets, so you might want to survey customers in order to check their perception on mock-ups or on the relevance of features for your prioritisation.
You could survey what customers think about your launch communication idea: the message, the appearance, the name or logo of what you want to promote. In my past, I conducted such surveys only for bigger projects. There are plenty of reasons why responses can be more positive, helpful or the opposite: Does the customer know you and your brand from previous usages which influences his response or maybe it is a groundbreaking innovative idea and customers cannot imagine yet how it will be.
But what you should definitely take into consideration to survey on an ongoing basis or regularly are:
- Brand and image, awareness and perception of specific values you are aiming towards
- Customer satisfaction, overall or with specific areas
- Concept tests in order to define which are most appealing benefits to focus on in your communication
- Packaging or pricing perception
There are many more examples which can be in other areas like your operations, your sales or HR teams etc. Just bear in mind how often you want and should conduct surveys and if you can manage not to “spam” your customers.
3. Choose your tool to gather feedback and set up the survey
As per theory you need to define whether you need a quantitative or qualitative approach, depending whether you want statistically more relevant results or individual responses.
Quantitative approaches need comparable answers, i.e. the usage of closed questions with rating scales, check boxes whereas qualitative approaches use additionally open questions within individual response fields or additional questions depending on the answer.
Tools could be basic emails including online tools like Google Forms and SurveyMonkey. The latter might be useful to facilitate the reporting as it generates data files or graphs as per configuration. You can also choose to invite your customers in or call them by phone. You can either address the survey individually to customers or make it an open survey via your website.
Online survey checklist:
- Create your list of questions which is as convenient to fill as possible, but make sure all necessary questions and explanations are reflected
- Choose wisely the number of response options, e.g. a feature list of 30 lines will scare respondents off
- Choose wisely the type of response depending what you want to read out
- Set up logics, so respondents do not see questions which are not relevant to them
- Most surveys are anonymous, so you might want to add a few general questions to understand the replies better e.g. in B2C like gender, age, income, qualification or in B2B like corporate form, country, revenue, business position
- Offer an incentive to take part in the survey
Make sure all information around the survey are communicated: until when the survey is open, who is the sender of it, why you conduct it, what happens with the data etc.
Personal interview checklist:
- Schedule interviews ahead
- Prepare a questionnaire which acts rather like an interview guideline
- Try if possible to ask each respondent the same
- Capture the responses immediately as usually personal interviews produces a lot of individual opinions
- Don’t capture your conclusion right away, just the actual response
- If possible store in your database the information whom you have contacted
A few hints on common challenges:
- Don’t use your survey to capture everything you ever wanted to know. Better: It increases the response rate if you keep the survey as short as possible and if you guide the customer through one single topic.
- Asking questions or use specific tonality to hear the expected answer. Better: double check if you really need the question and if your customers can truthfully respond to it, consider rating scales or further in-depth questions.
- Assuming the customer has the same knowledge of the product than you have. Better: explain terminologies used and keep it simple.
4. Define the contacted customers
Online tools allow you to address a large base of respondents easily and fast. If you want to gather feedback reflecting your whole customer base you can either ask your whole customer base or you make a subset of your customers.
In order not to bias your outcome, you should select a limited customer group which reflects your total base. So, for instance if your customer base is 50% male and female each, 70% coming from cities >100k and so on, you might apply the same ratio to the selected group.
Also take into consideration if you need a quota: e.g. you know that 80% of your customers are mobile users which are essential for your survey outcome, ensure that only 80% of mobile users responses are included in your results. Surveys from which you expect significant results should at least have 100 complete responses, so the invited base will need to be at least 10-20x times higher .
If you choose to call your customers or invite a few to your company, make sure you still select according to a few criteria reflecting your survey goal, selecting more than you need and to save the information who you have contacted. Be aware that you cannot reach that many, telephone or in person interviews last usually 20-45min per respondent. Then you need time to reach them and to capture the individual responses.
5. Finalise and what next
When all is set, make a short internal test to ensure your one-shot contact will be successful. For online surveys allow at least a week’s response time.
Jumping to conclusions immediately after the first responses are in, is very tempting. Nevertheless I strongly recommend to properly put all responses together and document it in a summary:
- Are all responses useable and comparable? Just consider whether all respondents have understood the questions and answer options rightly configured.
- Is there strange behaviour? I was so happy once when seeing 95% satisfaction rate in all my survey questions in India, until I realised that it was a telephone interview and all respondents have been very polite. Check significant positive or negative results if there could be other reasons for the result.
- When sharing results, make sure to add as a side note the total base of your respondents, the selection filters, the survey method, so your fellow colleagues can work with the results.
I might have omitted scientific approaches and other examples, but I still hope it is an indication where to start if you have never worked with surveys before. For conducting surveys you might want to check out the following tools:
What have been your experiences on getting customer feedback?