Types of data

Quantitative data is numerical information. It answers questions such as how many customers buy, how often they buy and how much they spend. For example, a bakery might survey 150 shoppers about a new breakfast deal: 96 say they would buy it once a week, and, separately, 78 say they would spend £4 or more. These are quantitative results.

Benefits

• Quick and easy to analyse: totals, averages and percentages can be worked out, so you can see the main result without reading lots of comments.
• Large samples and wide coverage: a business can ask many people the same questions and collect a lot of information in a short time.
• Patterns, trends and comparisons: summarising the numbers in tables, bar charts or line graphs makes the main findings easier to understand. The bakery could compare how many younger and older customers would buy the deal, or see whether interest changes over several weeks.
• Easy to share: charts and summary figures communicate results clearly to a manager, investor or business partner.

Limitations

• No reasons: numbers do not explain why 54 shoppers said they would not buy it once a week or how they feel about the price. They give limited insight into customers' opinions, thoughts and feelings.
• Sampling bias: if the bakery only surveys shoppers who already buy breakfast there, the sample is not representative of the wider market. A large sample can still be biased if it is chosen badly.
• Outdated quickly: tastes, prices and competitors change, so a survey from several months ago may no longer reflect demand.
• May need further research: the bakery might need interviews or a small discussion group to find out why customers like or dislike the deal before changing its menu.

Qualitative data is information based on opinions, attitudes, reasons, feelings and descriptions rather than mainly numbers. It helps a business work out why customers think or behave in a certain way, rather than just how many of them did something. What matters is the type of data, not how it was collected.

Benefits

• Detailed insight into motivations — customers explain their reasons in their own words, so the business learns what is really behind a choice.
• Reveals the unexpected — open answers can raise issues or ideas that the business never thought to ask about.
• Helps improve products and marketing — knowing the reason lets a business change the right thing to meet customer needs.

For example, a bakery leaves a comments box beside its new lunch deal. Several customers write that the queue is far too slow at half past twelve. The owner had only wanted to know which sandwich people preferred, but the comments point to a serving problem that is losing lunchtime sales.

Limitations

• Harder and slower to analyse — answers arrive as words, so someone must read, group and interpret them instead of simply adding up figures.
• Smaller samples — collecting detailed answers takes time, so fewer people usually take part. The findings may be less representative and harder to generalise to all customers.
• Difficult to compare, measure or analyse statistically — opinions do not fit neatly into averages, percentages or other numerical tests.
• Subjective interpretation and researcher bias — two people may read the same comments differently, and a researcher may notice the answers that match what they already believe.
• No sense of scale — it may not show the size of a market or the scale of demand. Six customers loving an idea does not prove that thousands would buy it.

Combining quantitative and qualitative data means using numbers and reasons together. The two types complement each other: quantitative data shows what is happening, such as how many customers buy a product or how often, while qualitative data explains why. On its own, each gives only part of the picture.

For example, a small bakery launches a vegetarian pasty. Its sales records show 300 sausage rolls sold in a week but only 40 pasties — roughly one pasty for every seven sausage rolls. The numbers reveal that the pasty is not selling well, but they do not say why. The baker asks 15 customers who bought a sausage roll what they thought of the pasty. Several did not notice it on the shelf, and some thought it cost more than a sausage roll. The numbers show the size of the problem; the opinions show the cause — poor visibility and a price misunderstanding. The baker can act on both: move the pasty to a clearer spot, add a price label, then check the sales figures again.

Using both gives a fuller picture and a more informed decision. Numbers help a business judge how big a problem or opportunity is, while opinions help it decide what to change. Numbers alone can show a pattern without explaining it, and opinions alone can reveal a reason without showing how many customers share it.

Collating data means gathering information from different places and bringing it together in one place so it can be used. Market research data is often scattered — a few figures in a till system, some comments on feedback cards, statistics from a trade body. Collating pulls all of it together so it can be compared, checked and acted on.

Data can be collated internally, from the business's own records and systems, or externally, from sources outside the business. Both quantitative and qualitative data can be collated either way, so numbers or opinions can come from inside the business or from outside sources.

For example, a bakery has weekly sales figures in its till system, comments on feedback slips left at the counter, and local population statistics from the council's website. Collating all three into one spreadsheet lets the owner compare them side by side: which days the till is busiest, what customers commented on those days and how many people live nearby. That comparison is only possible once the data is in one place.

Collating matters because scattered data is hard to use. Once it is gathered together, patterns can be spotted, different sources can be compared and the information can be used to make a decision.

External collation means gathering data from sources outside the business — for example government reports, trade association surveys, newspaper articles, competitors' price lists or published industry research. The business did not collect this data itself, and it can be numbers or opinions.

What external data gives you

Because the data comes from beyond the business's own records, it shows a wider picture of the market. Your own sales only tell you about the customers you already have; external figures can show what people are doing elsewhere, what rivals are charging or how a whole industry is changing. For instance, a small bakery in a seaside town might read a national report showing households are buying fewer loaves but more pastries, and use that to widen its pastry range.

It can also give a more independent view. A trade body or government department is not the business itself, so its findings may be taken more seriously than the business's own claims about itself.

What it costs you

External data may be more expensive and slower to obtain. Some industry reports have to be bought, and finding, reading and sorting through them uses staff time that could be spent serving customers. It can also be less specific: a report covering the whole country may say very little about your street, your prices or your shoppers. Because someone else gathered it for their own purpose, it may also need checking before you rely on it.

Judging a source

Before acting on external data, a business should judge the source against three questions.

A survey of shoppers in a large city may be recent and reliable, but it still tells a business in a small market town very little about its own customers.

Choosing the right data means picking the type of market research data that suits the aim of the research and the decision the business needs to make. Market research data is usually quantitative (numbers) or qualitative (opinions, reasons and descriptions). Matching the type to the question means the results actually help the owner decide something.

Match the data to the question

If the business needs to know how many customers there are, how often they buy or how much they spend, quantitative data suits it, because the answer comes as a number the owner can count and compare. If the business needs to understand why customers think or behave in a certain way, qualitative data suits it, because the answer comes as reasons and opinions in the customers' own words.

Worked example

A small bike repair shop is deciding whether to open on Sundays. Asking 300 past customers whether they would book a Sunday repair gives quantitative data: 78 say yes, so the owner can estimate the extra income Sundays might bring and compare it with the staff cost. That type suits this decision because a number is needed to weigh against a cost. Asking a few customers why they prefer certain days would give qualitative data, which suits a different aim.

Choosing a data type is not the same as choosing a research method, or as deciding whether data is primary or secondary; the type should follow from the aim. Time and budget still matter, so for a simple decision one type may be enough.