Market research gives marketers one of their most valuable assets: evidence. But evidence is only useful when it is analysed with discipline. Too often, research is treated as a reporting exercise. A survey is run, a deck is built, charts are presented, and everyone leaves the room feeling slightly better informed. That is not enough.
The purpose of market research analysis is not to describe what customers said. It is to improve the quality of marketing decisions.
A marketer analysing research should begin with the decision, not the dataset. Are we deciding which segment to target? Is a positioning credible? Which product feature matters most? How much are customers willing to pay? Why is consideration weak? The analysis changes depending on the question.
Quantitative tells you how much. Qualitative tells you why
There are several options available. Quantitative analysis is useful when you need scale: how many people think, feel, know or do something. It can show differences between groups, reveal patterns in behaviour, test associations between variables and help prioritise opportunities. Cross-tabs, driver analysis, segmentation models, conjoint, MaxDiff and tracking data all have a role when the question demands measurement.
Qualitative analysis does a different job. It is best for meaning. Interviews, groups, ethnography, customer diaries and open-text responses help marketers understand language, motivation, tension, context and category behaviour. Thematic analysis can reveal repeated patterns. Journey analysis can show where friction occurs. Language analysis can expose the words customers naturally use, which is often more useful than the phrases marketers invent internally.
The strongest analysis often combines both. Use qualitative research to explore, then quantitative research to size. Or use quantitative data to spot the pattern, then qualitative work to explain it. The mistake is asking one method to do the other’s job. A focus group cannot tell you market prevalence. A survey cannot fully explain why people behave as they do.
Good analysis means judging the evidence, not just reporting it
Good analysis follows a clear process. First, check the quality of the evidence. Who was sampled? Were they the right people? Were the questions neutral? Was the sample large enough? Were important customer groups missing? Poor inputs create poor outputs, however sophisticated the analysis looks.
Second, separate signal from noise. Not every difference matters. A statistically significant finding may still be commercially trivial. A striking verbatim may be powerful, but unrepresentative. The analyst’s job is to judge weight, not just find drama.
Third, triangulate. Look across multiple sources: customer research, sales data, search behaviour, complaints, market share, brand tracking, competitor activity and internal knowledge. One piece of data can mislead. Patterns across sources are harder to ignore.
The biggest danger is finding the answer you wanted
This is where analysis can go badly wrong. Research is often used to confirm what the business already wants to believe. Teams over-read small samples, chase the most exciting quote, ignore inconvenient findings or confuse stated intent with real behaviour.
The Levi’s Tailored Classics case is a useful warning, as featured in The MiniMBA in Marketing product module. In the 1980s, Levi’s wanted to move beyond jeans into higher-priced menswear. Research identified a “Classic Independent” segment: 21% of the market, responsible for 46% of wool-blend suit purchases, with a preference for traditional suits, independent stores and high-quality fit. On paper, it looked attractive. But the deeper research told a different story. These men associated Levi’s with jeans, not suits. They were uncomfortable with the idea of wearing a Levi’s suit to work and doubtful that a standard-fit garment could deliver the tailoring they wanted. Rather than listen, Levi’s pushed ahead with Tailored Classics. The range later achieved only 65% of its modest sales target.
The Levi’s example was not a failure of research collection. It was a failure of interpretation. The evidence was there, but the business focused on the parts that supported its ambition and discounted the parts that challenged it.
That is where market research analysis often goes wrong. Data is made to serve an internal opinion, rather than sharpen an external understanding of the market.
Analysis should sharpen judgement, not replace it
The best market research analysis is commercially sharp and methodologically humble. It does not pretend research has all the answers. It asks better questions, tests assumptions, exposes trade-offs and improves judgement.
Data tells you what was found. Analysis tells you what it means. Only then should strategy formulation begin.