Market research

Market research is the systematic process of gathering, recording and analysing data about a market — including its customers, competitors and trends — in order to make better-informed marketing decisions and reduce the risk of those decisions.

Primary (field) research collects NEW, first-hand data for the firm's specific purpose, for example a customer survey or focus group. Secondary (desk) research uses EXISTING data already collected by others, such as government statistics or a published market report. The key difference is that primary data is original and tailored, while secondary data pre-exists and was gathered for another purpose.

Two methods are: (1) a customer survey or questionnaire, which gathers responses from many customers at low cost; and (2) a focus group, in which a small group of target customers discusses their views to give qualitative insight.

A start-up typically has very limited finance, and secondary research (such as free government statistics, online data and published market reports) is much CHEAPER and FASTER to access than primary research, which requires paying to design and run surveys or focus groups (application: a start-up with little cash cannot fund a large primary study). Secondary research also gives a quick overview of market size and competitors, helping the founder make an early decision within a tight timeframe. However, because secondary data was collected for another purpose and may be out of date, the start-up would ideally add a small amount of targeted primary research once resources allow.

Mean = (6 + 7 + 7 + 9 + 11) ÷ 5 = 40 ÷ 5 = 8. Mode = 7 (the most frequent value). The MEAN of 8 tells the business the average response across all customers, using every value, which is useful for an overall summary. The MODE of 7 tells it the single MOST COMMON response, which is useful for identifying the most popular option (for example the most common rating or price point). Together they show the typical customer view, though the mean can be pulled up by the high value of 11.

The sampling method determines how REPRESENTATIVE the sample is of the whole target population, which drives reliability. A random or stratified sample gives each part of the population a fair chance of selection — stratified sampling in particular chooses proportionally from groups such as age bands — so the results are more likely to reflect the true market and can be generalised with confidence (application: a firm needing a dependable demand estimate would prefer these). By contrast, a convenience sample of whoever is easiest to reach, or a quota sample where the researcher hand-picks individuals, introduces selection BIAS: certain groups are over- or under-represented, so the findings may mislead. For example, surveying only passers-by outside one shop over-represents that store's existing customers. Sample SIZE also matters — a larger sample reduces the effect of chance. Overall, more rigorous sampling raises reliability but costs more money and time, so firms must balance accuracy against their budget.

The bookshop's decision — whether to add a café — is high-cost and risky, so market research is worthwhile, but the method must suit a small firm with a LIMITED BUDGET and no research department.

A sensible first step is SECONDARY research: free local council data on footfall and demographics, and observation of nearby cafés' prices and busyness. This is cheap, fast and gives an early sense of demand and competition. Its limitation is that the data is not tailored to the bookshop and may be out of date, so it cannot confirm whether the shop's OWN customers want a café.

To answer that, PRIMARY research is needed. A short in-store survey of existing customers (quantitative — how many would use a café, how often, at what price) plus a small focus group or informal interviews (qualitative — why they would or would not) would directly test demand. This is affordable and highly relevant. However, surveying only existing in-store customers is a CONVENIENCE sample: it is self-selected and excludes potential new customers a café might attract, so it risks over-stating demand and cannot be generalised to the wider market. A larger, more representative sample would be more reliable but is beyond the shop's budget and time.

On balance, the MOST APPROPRIATE approach is to COMBINE cheap secondary research with a small, targeted in-store primary survey and a focus group, staged by cost. This fits the bookshop's resources and answers both 'how many' and 'why'. The key judgement is that, because the primary sample is small and biased towards current customers, the owner should treat the findings as INDICATIVE and perhaps pilot the café on a small scale (e.g. a few tables) before a full commitment — using real trial data to reduce the residual uncertainty that no low-cost research can remove.

At first sight an 85% approval rating looks like strong evidence to launch, but the RELIABILITY of the research must be examined before the start-up commits its scarce resources.

The main strength is that the research is PRIMARY, cheap and fast: surveying followers cost almost nothing and gave a quick signal, which suits a cash-constrained start-up. The 85% figure and the sample of 50 provide quantitative data that is easy to act on.

However, the research has serious VALIDITY and BIAS problems. First, the sample is a CONVENIENCE, self-selected group of EXISTING followers — people already predisposed to like the brand — so the result is biased upward and does not represent the wider target market the firm hopes to sell to. Second, social-media followers skew towards a particular (often younger, online) demographic, further limiting how far 85% can be generalised. Third, a sample of 50 is small, so chance alone could distort the figure. Fourth, 'loved the design' measures enthusiasm, not willingness to PAY — customers may admire a product yet not buy it. For all these reasons the 85% cannot safely be read as '85% of the market will buy'.

Therefore the start-up should rely on this research only as an ENCOURAGING EARLY INDICATOR, not as a basis for a full launch. The most appropriate next step, proportionate to its budget, is to add a more representative piece of research — for example a stratified survey across age groups, or a small test-market/pre-order campaign that measures actual purchases rather than opinions. The justified conclusion is that acting on the biased 85% alone would be reckless; used alongside cheap follow-up research that corrects the sample bias, it can still usefully inform the launch decision while keeping risk low.

For a new product launch a business needs to answer both 'how many customers will buy?' and 'why (or why not)?', so the usefulness of quantitative and qualitative research depends on which question matters most at each stage.

QUANTITATIVE research (e.g. a large closed-question survey, secondary sales statistics) is useful for SIZING demand: it produces numerical data — percentages, average willingness to pay, forecast sales — that can be summarised with statistics such as the mean and mode and compared over time or against competitors. This helps the firm decide price and production volume and supports objective, evidence-based forecasting. Its weakness is that numbers do not explain the REASONS behind them: a survey may show 40% would buy but not why the other 60% would not, and poorly worded or leading questions can distort the figures.

QUALITATIVE research (e.g. focus groups, in-depth interviews) is useful for UNDERSTANDING motives: it reveals attitudes, emotions and objections, which is invaluable when refining a new product's design, branding or messaging. Its weaknesses are that it is subjective, harder to generalise, and usually drawn from small samples, so it cannot reliably estimate market size and can be influenced by a dominant voice in a group.

The strongest approach is therefore not to choose one but to COMBINE them, using each for what it does best — a common technique is to run qualitative research first to shape the product and generate hypotheses, then quantitative research to test demand at scale. The most appropriate balance depends on the firm's OBJECTIVE, BUDGET and TIME: a cash-rich firm launching a major product can afford both, whereas a small firm may prioritise cheap qualitative insight plus limited quantitative testing. The justified conclusion is that quantitative and qualitative research are COMPLEMENTARY: relying on either alone leaves a blind spot, so combining them gives the most reliable basis for a launch decision, provided the firm remains alert to sample bias in each.