When to use qualitative vs. quantitative research
When planning a market research study, one question comes up almost immediately: should we use qualitative or quantitative research?
That question often leads to even more. Should we run a survey? Conduct customer interviews? Bring people together for focus groups? Or use a combination of methods?
The key factor to understand here is that one methodology isn’t inherently better than another. Qualitative and quantitative research answer different kinds of questions, and the right choice usually comes down to two things: what you need to learn (your research objectives), and what you need to decide (the business decision the research needs to support).
Quantitative research helps you understand what is happening, how often it happens, how much it matters, and among which audiences.
Qualitative research helps you understand why something is happening, how people think about it, and what questions you may not have considered yet.
What is quantitative research?
When you think quantitative, think numbers. Quantitative research uses structured, measurable data to identify patterns, compare groups, and quantify behaviors, preferences, or attitudes. Its findings are typically expressed through numbers, percentages, probabilities, and statistical relationships.
Surveys are probably the most familiar form of quantitative (often called “quant”) research. Quantitative market research can also include more sophisticated approaches such as brand tracking, audience segmentation, and cluster analysis.
While most people think about multiple choice questions as the primary survey question type, there are some advanced survey question types and analysis techniques that researchers can use to answer specific questions, like which features should we prioritize? Or what’s the ideal bundle of products/services and how much should we charge for it? MaxDiff analysis reveals prioritization, or what customers value most. Conjoint analysis shows how buyers make tradeoffs among features, benefits, and price. Pricing research helps determine what customers may be willing to pay.
When should you use quantitative research?
Quantitative research is especially valuable when you need to identify or validate patterns across a larger audience. Imagine, for example, that you want to measure customer satisfaction across a national market.
A survey can show what percentage of customers are satisfied, which audience segments are more or less satisfied, how satisfaction has changed over time, and which factors influence brand loyalty or repeat purchases.
Use quantitative research when you need to:
Measure prevalence or magnitude: How many customers experience a certain problem? What percentage prefers concept A over concept B?
Compare audiences or segments: How do enterprise customers behave differently? Which attributes do first-time buyers value more than repeat purchasers?
Test hypotheses: Does an idea uncovered during customer interviews hold true across a larger target audience?
Prioritize features or attributes: Which capabilities matter most? Are there any that customers are willing to trade for a lower price?
Understand pricing and willingness to pay: Conjoint analysis reveals tradeoffs customers are willing to make between price and features. Van Westendorp analysis identifies an acceptable price range based on customer perceptions, while Gabor-Granger studies measure purchase intent at different price points.
Track changes over time: Are awareness, satisfaction, consideration or perceptions improving after a new campaign or product launch, or after a new competitor enters the market?
One important caveat: “quantitative” does not automatically mean “representative.” Even a survey with thousands of responses can produce misleading results if the respondents do not reflect the population you are trying to understand. Surveying only your most loyal customers, for example, may reveal what they want—but not what your entire customer base thinks. A thoughtfully recruited smaller sample can be more useful than 1,000 poorly selected respondents (which is why at Bixa, we’re so dedicated to making sure you get the right participants in your sample).
To draw conclusions beyond the people who took the survey, researchers need to think carefully about who the population is, how participants are recruited, whether important groups are adequately represented, and what statistical claims the sample can support.
AAPOR, the professional association for public-opinion and survey research, identifies sample construction, weighting, margin of error, questionnaire wording and recruitment methodology among the information necessary to interpret survey results responsibly.
In other words, the value of quantitative research isn’t simply that it includes more people. Its value comes from structured measurement designed around the decision you need to make.
What is qualitative research?
While the heart of quantitative research is measurement, qualitative research hones in on understanding. Qualitative research explores the motivations, perceptions, language, and experiences behind people’s decisions and behavior.
Qualitative methods include in-depth interviews and focus groups, which allow researchers to explore respondents’ opinions and motivations. Other approaches include ethnographic and observational research, which examine how people behave in real-world settings. Researchers might spend time with families in their homes to understand their daily routines or observe shoppers in a supermarket to see how they navigate the store and compare products.
Diary studies, in which participants record their experiences over time, can also provide valuable insights. Other approaches include shop-alongs, where researchers accompany consumers while they shop, and moderated user experience (UX) research. In moderated UX research, a researcher guides participants through a product or service to understand how they use it and where they encounter confusion or friction.
Together, these methods allow researchers to probe initial answers, follow unexpected lines of conversation, observe behavior, and understand how participants describe their experiences in their own words.
Ultimately, qual is especially powerful when you don’t yet know what you don’t know.
Suppose a survey shows that customers who use one part of your software are significantly less satisfied than other users.
That finding is useful, but the number alone won’t tell you why. The interface might be confusing, a particular customer segment may be disappointed by the feature’s capabilities, users may be unable to find something they need, or the product may behave differently from what they expected.
An interview or usability study can uncover the explanation behind the pattern.
Use qualitative research when you need to understand:
Why people behave the way they do
The motivations and barriers behind a decision
How customers think and talk about a category
Pain points you haven't identified yet
How people navigate a complex customer journey
Reactions to early ideas, messaging, or concepts
How a product fits into someone's real-world behavior
The language customers naturally use to describe a problem or need
Pay special attention to the last point about language. A survey built around internal terminology and assumptions may provide responses that fail to reflect how customers actually think or talk.
Even a small number of well-designed qualitative conversations can reveal unexpected motivations, concerns, terminology, and decision criteria. Researchers can then turn those insights into hypotheses that quantitative research can test across a broader audience.
Qualitative vs. quantitative research: What's the difference?
The primary difference between qualitative and quantitative research is the type of question each method answers. Qualitative research provides depth and explains why people think or behave a certain way. Quantitative research measures how common, important, or widespread those attitudes and behaviors are.
| Qualitative research | Quantitative research | |
|---|---|---|
| Primary purpose | Explore and understand | Measure and compare |
| Usually answers | Why? How? | What? How many? How often? |
| Data | Words, observations, behaviors | Numbers and structured responses |
| Typical sample | Smaller and purposive | Larger and structured |
| Common methods | Interviews, focus groups, diary studies, ethnography, usability testing | Surveys, MaxDiff, conjoint analysis, segmentation, pricing studies, brand tracking |
| Greatest strength | Depth, nuance, and discovery | Scale, measurement, and statistical analysis |
| Common role | Generate or explain hypotheses | Test or quantify hypotheses |
Keep in mind: Neither column is more rigorous, valuable, or informative than the other. The right methodology depends on the question at hand—not an organization’s preference for numbers, interviews, or a familiar research platform.
The third option: Mixed-methods
Combining qualitative and quantitative research is known as mixed-methods research. As with any custom research program, the business problem should guide the approach—including which method comes first.
Let’s take a look at some common scenarios.
When should qualitative research come first?
A common mixed-methods design begins with qualitative research and follows with quantitative validation. Researchers call this an exploratory sequential design: first explore the issue in depth, then measure it across a larger audience.
As an example, imagine that you’re developing a product for an audience you don’t yet understand well. You could immediately send 25 questions to 500 people—but what should those questions ask, and what language should they use?
Starting with interviews, observational research, or diary studies may first reveal what customers are actually trying to accomplish, what alternatives they're currently using, what frustrates them, which features they value and even how they think about the category itself.
Those insights can then inform a stronger quantitative survey.
You might discover five potential purchase drivers in interviews, for example, and build a survey to determine which ones matter most, across the market.
The qualitative phase identifies the possibilities. The quantitative phase measures which ones matter most, and how broadly they apply.
When should quantitative research come first?
Sometimes the first thing you notice is a pattern. Customer satisfaction may fall unexpectedly among one cohort. Churn may increase at a particular point in the customer journey. One concept may dramatically outperform another. Or a segment may behave in a way your team didn’t expect.
Quantitative research tells you that something is happening. Qualitative research can help explain why.
Researchers call this an explanatory sequential design. Quantitative research identifies the pattern, and qualitative research investigates the reasons behind it. Instead of speculating about the numbers, you speak directly with the people those numbers represent.
Sometimes: qual → quant → qual
For complex questions, two phases may not be enough. You might begin with qualitative discovery, build a quantitative study around what you learn, and then return to qualitative research to explain surprising findings or bring important customer segments to life.
For example:
First: Interview prospective customers to understand how they evaluate a category.
Then: Survey a larger sample to quantify the major decision drivers and identify distinct audience segments.
Then: Conduct additional interviews or diary research with people from those segments to understand what the statistical profiles actually mean in the context of their lives.
This intentional integration is the foundation of mixed-methods research. The NIH's widely cited mixed-methods research guidance defines the approach not simply as conducting qualitative and quantitative studies next to one another, but as integrating the two so they produce a more complete understanding of the problem than either could provide alone.
A real-world Bixa example: Discovering what customers value, then measuring it
Pricing and product development illustrate how qualitative and quantitative research can work together.
For Eko Health’s digital stethoscopes, Bixa conducted qualitative research with healthcare buyers alongside quantitative methods including MaxDiff and conjoint analysis. Each method had a different role in answering the larger business question.
Interviews with industry professionals helped us understand the needs, context, and buying considerations surrounding a medical device. Quantitative analysis then allowed us to measure the relative value customers placed on particular attributes and tradeoffs.
The research produced an especially consequential finding: Bixa’s conjoint analysis identified a specific product design change that could increase clinicians’ willingness to pay by $120 per stethoscope!
That's the advantage of connecting depth with measurement. Understanding why a feature matters is useful. Knowing how much the market values it can change the entire product and pricing conversation.
Mixed methods can also make segmentation much more useful
Audience segmentation is another area where quantitative analysis alone may not be enough.
Quantitative research can identify statistically distinct groups based on attitudes, behaviors, needs, or motivations. Cluster analysis may reveal that an audience isn’t one homogeneous market, but four meaningfully different customer groups. Qualitative research helps teams understand the people and behaviors behind those statistical segments.
But a statistical segment is only useful if teams can understand the human behavior behind it.
We’ve used this combination in research ranging from national audience segmentation to brand and marketing strategy. In one nonprofit project, we combined qualitative interviews with elected officials with a large national survey and cluster analysis to identify four psychographic personas among women considering political leadership.
The quantitative research created defensible segments. The qualitative research provided the context needed to understand the motivations, barriers, and journeys behind them.
This is what mixed-methods research does best: not simply producing more data, but making different kinds of data more useful together.
Is mixed-methods research always better?
No. Mixed-methods research is valuable only when both forms of research are necessary to answer the business question.
A company that needs to track brand awareness every quarter probably doesn't need to conduct 20 interviews every time it fields its tracker. Similarly, a product team trying to understand why users can't complete a specific workflow may learn far more from a focused usability study than from a 1,000-person survey.
Mixed-methods research requires additional time, budget, and expertise. Simply adding interviews to a survey doesn’t make a study better; the value comes from integration.
The qualitative phase should influence what happens quantitatively, or vice versa. Findings should be considered together while contradictions should be investigated rather than averaged away. Each method should have a clear job within the larger research design.
Any researcher at Bixa would tell you: both components must be rigorous and deliberately connected. One weak method can’t be rescued simply by adding another. The goal isn’t to use as many methodologies as possible. It’s to build the right level of evidence for the decision at hand.
Why a custom market research approach matters
This is also where working with a boutique market research firm like Bixa can be different from buying a survey, a focus group or a panel as a standalone product.
Business problems don’t arrive with methodological instructions: “For the right answer, please conduct 15 interviews followed by a 400-person MaxDiff study.”
They begin with decisions:
Should we launch this?
Which customers should we target?
Why are subscribers churning?
How should we position this product?
Which features should we build?
What should we charge?
What high-value offering can this audience not get elsewhere?
A strong research partner recommends quantitative research when the decision requires measurement, qualitative research when it requires depth, and an integrated approach when neither method can provide enough information on its own.
And because mixed-methods research can be sequential, the design doesn't necessarily have to be frozen from day one.
Something uncovered in interviews may deserve additional testing in the survey. A surprising survey result may warrant a follow-up conversation. An apparent segment may turn out to represent a behavior nobody inside the organization had considered before the project began.
That kind of flexibility is harder to achieve when each methodology sits with a different vendor, team, or software platform. In Bixa’s experience, research is more effective when qualitative and quantitative methods are designed as one cohesive program from the beginning.
The goal isn’t mixed methods for the sake of mixed methods. It’s research designed around the decision you need to make—and the evidence you need to make it confidently.
