Market research

Why a business does research at all

Market research is the collecting and analysing of information about customers, competitors and the market, so that decisions are made on evidence instead of on what the owner assumes.

The reason it earns its cost is risk. Launching a product, choosing a price or opening a second branch commits money that cannot easily be got back. Research does four jobs:

• Reduces the risk of failure. Finding out before launch that customers will not pay $40 is cheaper than finding out afterwards.
• Finds gaps in the market. A group of customers nobody is serving well is where new products come from.
• Shapes the marketing mix. It tells the business what price to set, where people expect to buy, and which promotion reaches them.
• Tracks what is already happening. Falling sales are a symptom; research finds out whether the cause is price, quality or a new rival.

Research does not remove risk. It narrows it, and a business still has to decide.

Primary research

Primary research collects new information directly, first hand, for this business and this question. It is also called field research.

Observation is the one to get right, because it answers a different kind of question. A questionnaire asks whether someone would buy a healthier snack and 70% say yes. Observation counts how many of them pick the healthy snack off the shelf, and the answer is usually far lower. When a question asks which method suits a business, the deciding point is often whether the business needs stated opinions or real behaviour.

Primary research in general is up to date, kept private from competitors and aimed exactly at the question the business is asking. It is also slower and more expensive than using data that already exists.

Sampling

A business cannot ask everybody, so it asks a sample: a smaller group chosen to represent the whole target market.

Sampling is useful because it cuts the cost and the time of research to something a business can afford, while still giving results that can be scaled up to the market as a whole. The trade-off is that the sample decides how much the results are worth. Too small a sample and a few unusual opinions distort the whole picture. A sample drawn from the wrong people, such as surveying weekday shoppers about a product aimed at office workers, produces confident numbers about the wrong market.

Secondary research

Secondary research uses information that already exists, collected by somebody else for some other purpose. It is also called desk research.

A sensible business uses both. Secondary research first, to size the market and see what rivals do, then primary research to answer the questions the existing data cannot.

What makes research data accurate

• Sample size. Twenty responses is an anecdote. The larger the sample, the less one odd answer matters.
• Who was asked. The sample has to come from the target market. Asking existing customers whether they like the brand tells you nothing about the people who never buy it.
• How the questions were worded. "Do you agree our prices are fair?" invites a yes. A neutral question does not suggest its own answer.
• Who did the asking. People soften opinions face to face, particularly about price and about anything embarrassing.
• When it was collected. Results gathered in a holiday period, or two years ago, describe a different market.
• Honesty of the answers. People overstate how much they exercise, recycle and would pay, and understate how much they spend on treats.

An answer that names one of these and then says what it does to the decision is doing the analysis the marks are for: a small sample means the business could set its price from the views of thirty people and price the product wrongly for the other thousands.

Reading simple research data

Questions give results as a small table, a bar chart or a pie chart, and the marks are for what the numbers mean for the business, not for reading a number off the page.

Worked example. A business surveyed 200 people about the most they would pay for a new sports drink.

Turn the counts into percentages, because a percentage can be compared with anything.

(90 / 200) x 100 = 45% (70 / 200) x 100 = 35% (40 / 200) x 100 = 20%

Those add to 100%, which is the check worth doing every time.

Now group them to answer the actual decision. The number who would pay $30 or less is 90 + 70 = 160, and

(160 / 200) x 100 = 80%

So a price above $30 loses four out of five potential customers, while a price under $20 is acceptable to everybody but throws away the 55% who would have paid more. The data points towards a price in the middle band, and the sentence that earns the analysis mark is the one explaining why: it keeps 80% of the market while charging more than the lowest group would pay.

Two habits protect these answers. Quote the figure, not the shape of the bar, because "the second bar is tallest" is not evidence. And say what the business should do about it, since a question about research data almost always ends in a decision.

Common mistakes

• Calling secondary research "second-hand opinions". It is existing data, whoever collected it.
• Listing a method without saying what it is for. A focus group and a footfall count answer different questions.
• Saying primary research is always better. It is more relevant and more expensive, and for sizing a whole market it is usually the wrong tool.
• Claiming research removes risk rather than reducing it.
• Saying a sample is "a small number of people" without saying it must represent the target market.
• Ignoring sample size when judging results, then treating the percentages as fact.
• Reading a chart and stopping. The marks are for what the figures mean for this business's decision.
• Writing percentages that do not add to 100%, or dividing the wrong way round.
• Answering a question on accuracy by saying "the data might be wrong", with no reason attached.
• Recommending a method without saying why the other methods were rejected, which is what a justified recommendation needs.

Not the topic you were looking for? Describe what you are stuck on in your own words and we will take you to the notes that answer it.