How to do Customer Surveys right
Once you know your customer pretty well and have adapted your Basics and Business Model after the Observations, Interviews or Focus Groups, you can start with customer surveys. Until now, your interactions with potential target customers were mostly centered around their problems, pain points, lifestyles, processes, fears, frustrations, dreams and happiness points... Now it's time to introduce your solution and make some real validations with a larger number of individuals through surveys.
What are Customer Surveys?
Surveys are lists of close-ended and open-ended questions. They are best if you'd like to make some statistics. They can be done both online and offline. The surveys can be about different topics similar to the ones in the Persona Framework but they can also introduce your startup's solution.
Â
Why are Customer Surveys important?
The first three reasons surveys are important are similar to the ones from Customer Knowledge. But there are additional ones: number 4 and 5.
1. Cheaply validate or disprove an existing problem, your target customers, your solution and potentially your revenue model: Instead of investing time and money directly into the solution, find out first whether your solution solves a customer need.
2. Problem solution fit: as Albert Einstein once said:
"If I had an hour to solve a problem, I'd spend 55 minutes thinking about the problem and 5 minutes thinking about the solution".
The better you know your customers' problems and pain points, the better you can formulate a creative and needed solution.
3. Marketing & Communication: it allows you to create better marketing and communication strategies and spend less money.
4. Market data: It can be that some data you're looking for was not available through desktop research, so surveys are your chance to gather some statistics.
5. Investors: compelling customer feedback and data from market research such as surveys can strengthen your position to get investments by convincing them that there is a need.
Â
How to create & conduct Customer Surveys?
1. Hypothesize
Similarly to Observations, Interviews and Focus groups, you should start by hypothesizing what you'd like to learn. The Lean Methodology recommends to approach building startups like experiments. Therefore, before starting to execute, formulate your concrete hypotheses with the help of Startegyzer's Testing and Learning cards Framework: Learning Cards.
It's important to understand what you want to learn to formulate better questions! The topics you want to learn about might include:
Occurrence, Frequency, and Strength of the Problem
Occurrence, Frequency and Strength of specific frustrations, fears, motivations, happiness, specific interactions, search habits or communication channels (from the Framework: Persona)
Typical baskets and basket sizes
Expenses spent to solve the problem / to use similar products
Product usage data
Attractiveness of your proposed solution
Consumer behavior
Brand choices
Marketing choices
And much more...
2. Prepare
You'll need to prepare two things:
Your survey questions
How you will get people to fill in your survey
Survey questions
You should aim at a survey that doesn't take more than 10 min to fill in. People are busy, so the shorter the survey is, the easier it will be to get participants. You can easily set-up a survey on Survey Monkey.
Survey monkey also provides a good overview at the type of questions
Multiple choice questions: select one or more options from a list
Rating scale questions: questions where answers display a scale of a selected range
Likert scale questions: "do you agree or disagree" types of questions
Matrix questions: they can be used for questions that have the same answer choices, so as to put it together in a table
Dropdown questions: long list of multiple choices, similar to multiple choice questions
Open-ended questions: require the participant to type an explanation to answer
Demographic questions: questions about demographic data (age, gender, income level...)
Ranking questions: participants can order answers by preference
Image choice questions: display images as answers to that participants can select
Click map questions: allows participants to click on a selected spot of an image
File upload questions: allows participants to upload files
Slider questions: similar to rating scale but with a slider
We recommend mostly using close-ended questions with choices and maybe one or two open-ended questions. But don't concentrate on open-ended ones because they take too long to answer. Try to make it as simple and as fast as possible to answer. From your interviews, you have already figured out possibile answers, so rather use these as choice options with one choice called "Other".
One other thing, you might have different types of target customers. Make sure to have a couple of questions to identify which group the participant belongs to.
Finally, similar to interviews, there are certain question rules:
DO NOT ASK WHETHER YOUR BUSINESS IS A GOOD IDEA, most people would say "yes", "I would buy it" but it doesn't mean anything. Rather, make them perform an action for your (Call to action), for example: "Leave your email to get a 10% discount at our release", "Leave your email to get our weekly newsletter"... or a small commitment, "Refer the survey to a friend" etc.
Be as specific as possible, we want their answers to be based on facts, past experiences, their lives
How you will get people to fill in your survey
Once you start with the surveys, you should already have a pretty good pictures of your target customers and their persona profiles. In the persona profiles, there are 3 important sections for this part:
Interaction: Where (online & offline) does the persona interact?
Information: Where does the persona search for information when they search for something relating to the topic?
Communication: What is your persona's preferred way of communication?
Based on this information, identify
Where specifically you will distribute your survey. You can use the same channels as for the observations and interviews
Family, Friends and friends who know friends => select the people
Contacting "random" people on LinkedIn => select the people
Online social media groups of like-minded people => identify specific groups
Offline associations, clubs, forums local gatherings, etc. => identify specific groups
Organize an event
"Get out of the building" and ask random people. Here, you have to find a place where your target customers hang out. For example, people who have pets go to specialty shops for pets, you can just stand in front of that shop and ask to talk to customers.
How many people should fill in the survey. Usually, at least 100 people should fill in a survey but the more the better. Especially, if you'd like to infer statistics out of it.
3. Test
Test the survey with a couple of friends to correct unclear questions, typos or mistakes and get their ideas, maybe they have suggestions on what else you should be asking.
4. Execute
Start inviting people to fill in the survey and publishing it in identified channels. To make some conclusions you should have at least 100 participants. However, this is the very minimum, a good number would be 300-500 participants and a great number: 1'000 participants.
5. Analyze
Finally, analyze your results. You can export your data to an excel file and start making analyzing of the answers. Then, go back to the learning cards and write down the conclusions, what you have learned and what you will do about it.
You've now talked to a few potential customers and learned quite a new deal of information. Based on these learnings, go back to your Problem, Target Customer, Solution, Value proposition, Summary and Revenue model to make adjustments.
It can easily happen, that you "pivot" at this point. A pivot is simply a change in one or multiple parts of your basics and business model (problem, target customer, solution, value proposition or revenue model). If you do, this is a very positive thing! It means that instead of building something no one would buy, you already pivoted at the beginning without wasting time and money!
Â
Supporting Tools & Additional Resources
Testing & Learning Cards: Startegyzer
Tool to create surveys: Survey Monkey
Explanations on the types of survey questions: type of questions
Â
By the way, you might not be busy all the time with this part since you'll be waiting on the answers. If that's the case, then move on to Market Analysis.
We hope that you've gathered as many participants as possible and can confirm or disconfirm your hypotheses!