Market segmentation splits a market into groups of people who share needs, attitudes, or buying habits. Surveys help you find those groups and understand how each one behaves. This guide covers example questions for each type of segmentation variable, how to write answer options that work, and how to turn the answers into groups you can act on.
What segmentation surveys are for
A customer survey is a structured set of questions that collects information from customers or potential customers. A segmentation survey goes a step further. It asks questions that show how people differ, so you can compare groups and decide where to focus your marketing or product work.
The four common bases for segmentation
Segmentation is often described in four parts: geographic, demographic, psychographic, and behavioral. The terms overlap, and different textbooks draw the lines in slightly different places, but the four are a useful checklist.
| Base | What it describes | Example question |
|---|---|---|
| Geographic | Where people live or work | “Which region do you live in?” |
| Demographic | Age, household, occupation, and similar traits | “Which age group do you belong to?” |
| Psychographic | Attitudes, values, and interests | “How much do you agree: I research a product thoroughly before I buy it.” |
| Behavioral | What people do, such as how often they buy | “In the past three months, how many times have you bought [category]?” |
Most segmentation studies use at least two bases together. For example, you might compare how often people buy (behavioral) across age groups (demographic).
Demographic questions for a survey
Demographic questions are usually the easiest to answer and the easiest to analyze. Keep them simple and make sure the answer options do not overlap.
- Age: “Which of these age groups do you belong to?” Use ranges such as 18 to 24, 25 to 34, 35 to 44, 45 to 54, 55 to 64, and 65 or older, plus “Prefer not to say.”
- Household: “How many people live in your household, including yourself?” Options: 1, 2, 3, 4, or 5 or more.
- Employment: “Which best describes your current work situation?” Give a short list and include “Other.”
- Income: Income is sensitive. Ask it in bands set for your market’s currency, make it optional, and explain why you are asking.
Put demographic questions near the end of a survey, unless you need them to qualify people. Starting with personal questions can make people stop before they reach the questions you care about most.
Psychographic questions
Psychographic questions ask about attitudes, values, and interests. They are usually written as statements and answered on an agreement scale, as explained in the guide to Likert scale questions.
- “I like to try new brands before my friends do.” (Strongly disagree to Strongly agree)
- “I look for products that are made in a way that is good for the environment.” (Strongly disagree to Strongly agree)
Keep psychographic statements tied to the decisions your business makes. Avoid statements that stereotype a group, because they can produce answers that are hard to interpret. Also avoid self-labels such as “Are you a price-conscious shopper?” People tend to answer in a way that fits how they see themselves, which may not match what they do. Ask about the behavior instead.
Behavior questions
Behavior questions usually produce the most useful segmentation data, because they describe what people actually do. Two points matter most: ask about a clear, recent time window, and give answer options that cover every possibility.
Purchase frequency and recency
- “In the past 12 months, how many times have you bought [category]?” Options: 0, 1 to 2, 3 to 5, 6 to 10, or more than 10.
- “When did you last buy [category]?” Options: within the last month, one to three months ago, three to six months ago, more than six months ago, or never.
Long recall periods tend to produce less accurate answers, because people forget. Where you can, use a window of a few months rather than “ever.”
Where and how people buy
- “Where do you usually buy [category]?” Use a multi-select list: in a store, from an online retailer, from the brand’s own website, from a marketplace, or other.
- “Which device do you use most when you buy [category] online?” Options: phone, tablet, laptop or desktop.
What prompts a purchase
- “What usually prompts you to buy [category]?” Offer a list of likely triggers, such as a replacement, a special occasion, a recommendation, or an advertisement, and include “Other.”
- “Which of these did you consider before your last purchase?” Show brands or options and allow multiple answers.
Writing answer options that work
Good answer options are mutually exclusive and exhaustive. That means each person fits exactly one option, and every person has an option that applies to them.
- Avoid overlapping ranges. “18 to 25” and “25 to 35” overlap at 25. Use “18 to 24” and “25 to 34.”
- Include “Other” or “Prefer not to say” where the list might not cover every answer.
- Keep one idea per question. A question about “price and quality” is double-barreled. Split it into two. The guide to leading and biased survey questions has more examples of this problem.
- Use the same scale throughout. Switching between scales makes results harder to compare.
Turning answers into segments
Once the data is in, the first step is to cross-tabulate. Compare how people in one group answer a behavior question with how people in another group answer. Patterns often appear quickly.
If you want a more formal grouping, cluster analysis can combine several variables into segments. Results depend on the variables you choose and how you scale them, so treat the output as a starting point for judgment, not a final answer. Before you act on any segment, check three things:
- Size. Is the group large enough to be worth targeting?
- Difference. Does the group differ in a way that affects a decision?
- Stability. Does the pattern hold in a second sample or a later wave of the survey?
Name segments by what people do or need, not by stereotypes. A label like “frequent buyers who compare prices” is more useful than a personality type.
Keep satisfaction questions separate
Questions about how satisfied customers are with a product or service belong in a separate survey. Mixing them into a segmentation study makes the survey longer and blurs the results. The customer satisfaction survey questions guide covers CSAT, NPS, and effort questions.
Common mistakes
- Overlapping or missing answer ranges. These make it impossible to place some respondents in a group.
- Asking too many personal questions at the start. People may leave before they reach the behavior questions.
- Long recall windows. “Have you ever bought this?” is less useful than “in the past three months.”
- Using self-labels. “Are you a loyal customer?” tells you less than how often people buy.
- Creating segments that are too small or too similar. A segment you cannot reach, or one that does not differ from the others, does not help a decision.
Frequently asked questions
What is a customer survey?
A customer survey is a structured questionnaire that collects feedback from customers about their experience with a product, service, or brand. Some customer surveys focus on satisfaction, while others, like the ones in this guide, focus on who customers are and how they buy.
What are some good marketing questions to ask?
Useful marketing questions ask how people first heard about a brand, what they considered before buying, what influenced the decision, and where they usually shop. Ask these after you have chosen a clear goal for the survey, so each question has a purpose.