Qualitative vs Quantitative Research: Which Do You Need?

Two kinds of question, not two competing methods

"Qualitative vs quantitative" sounds like a choice between two competing approaches to the same problem, with one presumably better than the other. It is not. They answer different kinds of question, and the question you are actually asking, often without realising it, usually points clearly to one or the other.

Quantitative research answers questions with numbers: how many, how often, what proportion, is A bigger than B. Qualitative research answers questions with explanations: why does this happen, what does this mean to the people involved, how do they experience it. Neither is more rigorous than the other. They are rigorous about different things.

What quantitative research tells you

Quantitative research measures something across enough people that you can describe the result with a number and have some confidence that the number means something: 62% of users prefer option A, satisfaction scores rose from 6.2 to 7.4, conversion is 3% higher on the new checkout flow. Surveys, structured questionnaires, analytics, and A/B tests all sit in this category.

The value of quantitative research is scale and comparability. It can tell you that something is true for most of your audience, not just the handful of people who happened to take part, and it can tell you whether one version of something performs better than another in a way you can defend with a number. The trade-off is that it tells you very little about why. A survey can tell you that 40% of respondents are dissatisfied with onboarding. It generally cannot tell you what, specifically, is going wrong, or what would fix it. We cover this approach on our surveys and quick insights page.

What qualitative research tells you

Qualitative research goes deep with a smaller number of people: interviews, usability testing, contextual inquiry, focus groups, diary studies. Instead of a number, the output is an explanation, often grounded in direct observation or what people say in their own words, of why something happens, how people think about it, and what is actually going on beneath a behaviour or a metric.

The value of qualitative research is depth and explanation. It can tell you why users abandon a checkout flow at a specific step, what mental model someone has of a product that does not match how it actually works, or what a "good outcome" looks like from a patient's perspective rather than a clinician's. The trade-off is that findings come from a small number of people, so you cannot put a percentage on how common a finding is across your whole audience. Five people describing the same frustration tells you the frustration is real and worth fixing. It does not tell you whether it affects 5% or 50% of your users. We cover this in depth on our in-depth user research and usability testing pages.

Where the wrong choice goes wrong

Treating a small qualitative sample as a statistic. "3 out of 5 users said X" sounds like 60%, but it isn't a percentage of anything beyond those five people. Qualitative findings describe patterns and explanations, not population-level proportions. If a stakeholder asks "but what percentage of our users feel that way?", that is a quantitative question, and the honest answer from a five-person study is "we don't know, but here's what we'd want to measure to find out".

Asking "why" questions in a survey. Open text fields in a survey can surface a few illustrative quotes, but most respondents give short, surface-level answers to open questions when there is no one to follow up with them. If the real question is "why are people doing this", an interview, where a researcher can ask a follow-up, probe an unexpected answer, and notice things the participant did not think to mention, will get you much further than a free-text box on a form.

Running a big study when a small one would answer the question. If the question is "what's confusing about this checkout flow", five usability sessions will surface the major issues quickly (see our guide to how many participants you actually need). Commissioning a 500-response survey to answer the same question is slower, more expensive, and less likely to surface the specific points of confusion than watching five people actually try to check out.

Running a small study to settle an internal argument about scale. If two stakeholders disagree about whether "most" or "a few" customers care about a feature, a handful of interviews will not settle it, because both sides can find a participant who confirms their view. That is a quantitative question, and needs a quantitative answer.

How to choose: start from the decision, not the method

The quickest way to choose is to finish this sentence: "After this research, we will be able to..." If the answer is "...say with confidence how common this is, or whether A performs better than B", you need quantitative research. If the answer is "...understand why this happens and what to do about it", you need qualitative research.

A second useful test: do you currently know the problem exists, but not why, or do you not yet know whether it is a problem at all? "We know drop-off is high at this step, but don't know why" is a qualitative question (go and watch people use it). "We think this might be a problem but aren't sure how widespread it is" is a quantitative question (go and measure it across a larger sample).

Using both together: the most common pattern in practice

In real projects, the two approaches are often sequenced rather than chosen between. A common pattern is qualitative first: a handful of interviews or usability sessions to understand the landscape, generate hypotheses, and identify what is worth measuring. Then quantitative: a survey or analytics review across a larger sample, to find out how widespread those issues actually are.

The reverse order is just as common. Analytics or a survey shows you that something is happening, a drop-off, a low satisfaction score, a feature nobody uses, but not why. Qualitative research then digs into the specific "why" behind the number: a handful of interviews or usability sessions with people from the affected group, to understand what is actually going on.

If you are not sure which end to start from, tell us what you already know and what decision the research needs to support. We will recommend a method, or a sequence of methods, that gets you a real answer rather than just an activity.