A qualitative survey explained: what it is, the kinds of qualitative survey question with examples, and how to read every open response, not a sample.
Definition
A qualitative survey is a survey whose questions are mostly open-ended — it gathers answers in words rather than numbers. It is still a survey: sent to many people, answered on their own time. What makes it qualitative is the kind of question inside it.
What it is
Instead of rating scales and multiple choice, a qualitative survey asks "what changed for you?", "what got in the way?", "what would you keep?" The respondent writes a sentence or a paragraph. The data is text, read by coding into themes.
What it captures
A qualitative survey reaches the things a fixed-option question cannot — the barrier you did not list, the reason behind a low score, the suggestion you would never have thought to offer. It trades the precision of a number for the meaning a number leaves out.
How it differs from a quantitative survey
A quantitative survey uses closed questions and produces numbers you can total. A qualitative survey uses open questions and produces words you read for meaning. Most real surveys mix the two — for that, and for the question "is a survey qualitative or quantitative?", see qualitative and quantitative survey.
The questions
A qualitative survey is only as good as its questions. The five types below each open a different door; the examples are written to be reused. A good open question is specific, neutral, and asks for one thing.
| Question type | What it opens | Example |
|---|---|---|
| Experience | What the respondent actually went through. | "Describe your experience with the program in your own words." |
| Reason | The reason behind a rating or an outcome. | "What is the main reason for the score you gave?" |
| Change | What shifted, and how. | "What can you do now that you could not do before?" |
| Barrier | What got in the way. | "What made it hard to take part, if anything?" |
| Suggestion | What the respondent would change. | "What is one thing you would change, and why?" |
Qualitative survey questions, by program type
Customer experience
Reason
"What is the one thing that would make you more likely to recommend us?"
Barrier
"Was there a moment our service let you down? Tell us what happened."
Each answer explains a score the dashboard cannot.
Training
Change
"What can you do now, after the training, that you could not do before?"
Suggestion
"What part of the training would you change for the next cohort?"
Each answer names what the program should keep or drop.
Scholarships and grants
Experience
"Describe how the application process felt from your side."
Barrier
"Was any part of the application unclear or hard to complete?"
Each answer points to a step worth fixing before the next cycle.
Reading the responses
Running a qualitative survey is the easy part. The work is in the reading — and a qualitative survey that gets skimmed instead of read speaks for a handful of people, not the group. Three practices keep every response counted.
Practice 01
Decide the themes you are listening for before you read — and let new ones be added as they appear. Coding every answer against one defined scheme is what turns a pile of text into a finding you can report.
Why it matters
Without a scheme, two readers code the same survey two different ways.
Practice 02
The barrier that matters often shows up in a small share of answers. Skimming for a quotable line misses it. Reading all of them is what makes the survey representative.
Why it matters
A theme is only evidence when you know how many people raised it.
Practice 03
Keep every open answer attached to the person who wrote it — and to whatever number you also collected from them. A theme tied to one record can be read against a score, a cohort, a demographic.
Why it matters
An anonymous quote cannot be correlated with anything.
Coded against a scheme, read in full, and tied to the respondent, a qualitative survey becomes evidence rather than anecdote. For how those themes are then read alongside the numbers, see qualitative and quantitative analysis.
Who this is for
A qualitative survey is most valuable to the teams who already send one — and only ever skim the answers. For each, reading every response changes a different cost.
Customer experience
The team adds a "why" box to its satisfaction survey. The comments come back by the hundred and get read for a quotable line, if at all.
What reading in full changes
Every comment is coded, so the team sees how many customers raised a barrier — not just that one of them did.
Training
The team runs an end-of-course qualitative survey. The open reflections are rich, but reading every one by hand never fits the reporting deadline.
What reading in full changes
Every reflection is read and themed, so the report names what the cohort actually valued — not a sampled impression.
Scholarships and grants
The team surveys applicants on the process. The open answers point to friction in the application, but the volume means they are skimmed.
What reading in full changes
Every answer is coded, so the team can see which step lost applicants and fix it before the next cycle.
Works the same way for fellowship reviews, accelerator cohorts, and grant cycles — the same qualitative survey, different respondents.
FAQ
A qualitative survey is a survey built mostly from open-ended questions — it gathers answers in words rather than numbers. It is still a survey, sent to many people and answered on their own time; what makes it qualitative is that respondents write in their own words. The data is text, read by coding into themes.
Qualitative survey questions are open-ended: they ask a respondent to answer in a sentence or a paragraph rather than pick from fixed options. Common types are experience questions, reason questions, change questions, barrier questions, and suggestion questions. A good one is specific, neutral, and asks for one thing at a time.
Examples include: "Describe your experience with the program in your own words" (experience), "What is the main reason for the score you gave?" (reason), "What can you do now that you could not do before?" (change), "What made it hard to take part?" (barrier), and "What is one thing you would change, and why?" (suggestion). Each opens a different door.
A quantitative survey uses closed questions — rating scales, multiple choice — and produces numbers you can total and compare. A qualitative survey uses open questions and produces words you read for meaning. One measures how much; the other explains why. Most real surveys mix both kinds of question.
Use a qualitative survey when you do not yet know what to measure, when a number has moved and you need the reason, or when you want respondents to raise things you did not think to ask. It is the right tool for exploring, explaining, and surfacing the unexpected — and a weak tool for measuring how widespread something is.
Analyze a qualitative survey by coding: decide the themes you are listening for, then read every response and tag it against that scheme, adding new themes as they appear. Report each theme with how often it occurred and an example answer. Coding every response — not skimming a sample — is what turns open text into evidence.
There is no fixed number. A qualitative survey can be useful with a few dozen responses, because the goal is range of experience, not statistical scale. The practical limit has always been the other direction: the more responses you collect, the harder they are to read in full by hand — which is the problem coding against a scheme is meant to solve.
Yes, and most do. A qualitative survey is defined by its open questions, but pairing each one with a closed question — a rating, a yes/no — is good practice: the closed answer gives scale, the open answer gives the reason. Keeping both attached to the same respondent is what makes the pair useful.
A good qualitative survey question is specific, neutral, and single. Specific: it asks about a real moment, not a vague impression. Neutral: it does not lead the respondent toward an answer. Single: it asks for one thing, so the response is codable. "What made the application hard?" beats "How was your overall experience and what would you change?"
No. Both gather qualitative data, but a qualitative survey is written and self-completed, sent to many people at once, with fixed prompts. An interview is a live conversation that can follow up and probe. A qualitative survey scales further and is more consistent; an interview goes deeper. They are complementary methods.
In research, a qualitative survey is used to explore a question before measuring it, to explain a quantitative result, or to gather a wide range of experience without the cost of interviewing everyone. It produces text data that is analyzed by coding, and it is often run alongside a quantitative instrument so the study has both scale and reason.
You make qualitative survey results comparable by coding every response against the same defined scheme, so the same answer is read the same way each time and across cohorts. Consistent coding is what lets you say a theme rose this cycle or appears more in one group than another — turning a qualitative survey from a set of quotes into a comparable measure.
Reading every response used to mean many hours of manual coding, which is why large qualitative surveys got skimmed. Coding each response against a fixed, versioned scheme — applied the same way to every answer as it arrives — is what makes reading the whole survey feasible, so the result speaks for everyone who answered, not the few that were sampled.
Bring a qualitative survey
A working session, not a demo. Bring a real qualitative survey — a batch of open-ended responses you have never had time to read in full. We code every one against a scheme you define and show the themes, live, with the count behind each.
Live walkthrough · 30 min · with Unmesh Sheth, Founder & CEO
