As the names suggest, quantitative research produces numerical data while qualitative research generates non-numerical data — but that’s only a surface-level distinction.
When there’s real uncertainty, or a business problem isn’t yet clearly defined, exploratory research using qualitative techniques is the right tool. When there’s little ambiguity and the problem is already structured, descriptive or causal research using quantitative techniques takes over.
Qualitative Research
Qualitative research gives vital insight into why consumers feel or behave the way they do — identifying underlying opinions, behaviour patterns and motivations rather than measuring a fixed set of variables. It helps formulate hypotheses for deeper exploration or later quantification, and adds genuine richness to what quantitative research measures by understanding rather than counting. In short, qualitative research decodes the less rational, more emotional side of a consumer’s decision-making.
Qualitative studies are particularly useful for market studies assessing interest in a new idea within a specific demographic, new product development aimed at understanding a genuine end-user need, creative development research (what should a brand say, and how should it say it), and diagnostic studies comparing how a company’s category or brand is perceived against a competitor’s.
Focus Groups in Depth
Focus group discussions, also known as group interviews, are the most widely used qualitative research technique. A group of six to twelve respondents is put at ease and asked general questions before the conversation shifts toward the actual research topic, with one or two moderators guiding the discussion — where two moderators are used, they typically adopt deliberately opposing positions to draw out a fuller range of reactions. The group is usually observed from an adjoining room, with the discussion recorded for later analysis.
A well-run focus group follows a fairly consistent process: define the research problem clearly, select an appropriate sample, decide how many separate groups are needed (a minimum of two is standard, since a single group’s dynamic can skew results), arrange the logistics and venue, select and brief the moderators, prepare the discussion material, run the session, and finally analyse the recorded data into a summary report.
| Advantages | Disadvantages |
|---|---|
| Relatively low cost and can be conducted quickly, including as a pilot study. | Not appropriate for gathering quantitative data — results aren’t statistically projectable. |
| Genuinely flexible, and can surface reactions a structured survey would miss. | A dominant, self-appointed group member can end up shaping or restricting other members’ responses. |
| A skilled moderator can read facial expressions and non-verbal cues from respondents who don’t articulate their views well verbally. | Quality depends heavily on the moderator’s own skill, and results can be inconsistent across moderators. |
| Respondents can complete a questionnaire before or after the discussion for additional data. | Recording equipment, if placed poorly, can make respondents self-conscious and affect the naturalness of their responses; the group also may not fully represent the broader population it’s drawn from. |
In-Depth Interviews
Personal, one-on-one interviews are preferred over focus groups in several specific situations: discussing sensitive or confidential topics where a group setting would inhibit honest answers, situations requiring detailed individual probing (automobile purchase psychology, for instance, tends to be best explored one-on-one), situations where a respondent might otherwise be swayed by group opinion (views on a socially sensitive topic, say), and interviews with busy professionals whose schedules don’t suit a group session (getting a doctor’s perspective on a new medical topic, for example).
Projective Techniques
Direct questions don’t always surface a consumer’s real motivations — people aren’t always fully aware of their own reasons, and even when they are, they don’t always articulate them accurately. Projective techniques work around this by extracting emotional responses indirectly:
- Brand personalities: Testing the strength of a brand’s advertising by asking respondents to imagine the brand as a person — what would this brand be like if it were a person?
- Collages: Asking respondents to build a collage representing a brand, to understand what associations and symbols consumers connect with it.
- Word association: Recording the first word that comes to mind when a brand name is mentioned — a brand that consistently produces a strongly positive first association is maintaining a strong brand image.
Quantitative Research
Quantitative research aims to produce numerical data and uncover statistically projectable patterns, drawn from a carefully selected sample representative of the target population. Where qualitative research is inherently subjective, quantitative research is comparatively objective. A 95% confidence level is the standard benchmark in empirical sampling — meaning that if the survey were repeated 100 times across independent random samples, roughly 95 of the resulting confidence intervals would successfully capture the true population parameter.
Quantitative research is most widely used to determine cause-and-effect relationships — testing, for instance, how much revenue might be expected to grow if a marketing budget increases by 15%. If the relationship between the two variables proves statistically strong, the hypothesis holds and the company can justify increased investment with real evidence behind it, not just intuition.
Data collection for quantitative research typically runs through surveys and questionnaires, delivered by mail, phone, in person, or online — and increasingly, mobile-first online surveys have become the dominant channel given how much of daily browsing now happens on a phone.
Mail surveys are inexpensive and preserve anonymity, but offer no chance to clarify a confusing question. Telephone and face-to-face surveys reduce the risk of error since an interviewer is available to assist, though that same interviewer can inadvertently bias responses. Online surveys reach a wide audience quickly and cheaply, but without some incentive, response and completion rates can be disappointingly low.
A hybrid approach — commonly a Telephone-Mail-Telephone (TMT) method, where respondents are briefed by phone and then sent the survey to complete at their own convenience — can combine the strengths of more than one channel.
The right medium depends on budget, available time and the complexity of what’s being asked: mail surveys suit a tight budget, online surveys suit a tight timeline, and personal or telephone surveys suit situations genuinely requiring real-time interaction.
Choosing Between the Two
Where budget allows, qualitative and quantitative research work best together rather than as a strict either-or choice — they offer genuinely different perspectives and complement each other.
Some research designs, such as usage-and-attitude studies, deliberately combine both in a single project. A qualitative report tends to read as a narrative account with direct quotations from respondents; a quantitative report tends to read as a statistical summary of correlations, significance levels, means and other hard figures.
Newer Techniques Worth Knowing
Alongside the traditional techniques above, two newer approaches have become increasingly common as more customer behaviour happens digitally:
| Technique | Type | What It Adds |
|---|---|---|
| Ethnographic / observational research | Qualitative | Observing customers in their natural environment — at home, in a store, using a product — rather than asking them to describe their own behaviour, which sidesteps the gap between what people say they do and what they actually do. |
| Digital analytics and A/B testing | Quantitative | Measuring actual behaviour at scale directly from website or app data, and testing two versions of a page, offer or message against each other with a real audience rather than relying solely on stated survey preferences. |
Both reflect the same underlying shift: as more of a customer’s actual behaviour leaves a digital trace, researchers increasingly have the option to observe real behaviour directly rather than relying entirely on what a respondent reports about themselves — though self-reported research still matters, since it’s often the only practical way to understand the reasoning and motivation behind a behaviour, not just the behaviour itself.
In practice, most well-run research functions today treat digital analytics and A/B testing as an always-on layer sitting underneath more deliberate, periodic qualitative and quantitative studies — the analytics tell you what’s happening in near real time, while the deeper qualitative and quantitative work explains why it’s happening and how confidently that pattern can be generalised beyond the specific customers observed. Neither layer fully substitutes for the other.
For a student or a working manager choosing a research approach for a specific business question, the most useful first question usually isn’t ’qualitative or quantitative’, but ’how well-defined is the problem already, and how confident do I need to be in the answer before acting on it’ — the honest answer to that question does most of the real work of choosing the right method, or the right combination of methods, for the situation at hand.
Frequently Asked Questions
Q: When should a researcher use qualitative research instead of quantitative research?
A: When there’s real uncertainty or the business problem isn’t clearly defined yet — qualitative, exploratory research is the right tool for understanding underlying motivations before a more structured, quantitative study can be designed.
Q: Why is a minimum of two focus groups recommended rather than just one?
A: A single group’s dynamic — a dominant participant, an unusual mix of personalities — can skew results in ways that aren’t representative, so running at least two groups helps confirm whether a finding is genuine or just an artifact of one particular group.
Q: What are projective techniques used for in market research?
A: To surface emotional or subconscious associations that direct questions often miss, since consumers aren’t always fully aware of, or able to articulate, their real motivations — brand personality exercises, collages and word association are common examples.
Q: What does a 95% confidence interval actually mean in quantitative research?
A: That if the same survey were repeated 100 times, the same response would be obtained roughly 95 times out of those 100 — a standard way of expressing how reliable a given sample size’s results are likely to be.
Q: Can qualitative and quantitative research be used together?
A: Yes, and where budget allows they often work best combined — qualitative research uncovers the underlying why, and quantitative research confirms how widely that finding holds across a larger, representative sample.
