4.4 Market research

Learning objectives

• Define market research, primary research and secondary research.

• Explain why organisations carry out market research and describe the stages of the market research process.

• Explain the advantages and disadvantages of surveys, interviews, focus groups and observations.

• Explain the advantages and disadvantages of market analyses, academic journals, government publications, media articles and online content.

• Distinguish between qualitative and quantitative research.

• Explain quota, random and convenience sampling, and how sampling error and bias arise.

• Calculate the size of each quota in a quota sample.

• Evaluate how an organisation carries out market research, including the ethical issues it raises.

Introduction

Market research is the collection and analysis of information about customers, competitors and markets, used to support marketing decisions. Every business, from a street-food stall to a multinational company, decides what to sell, to whom, at what price and how to promote it, and market research reduces the guesswork in those decisions.

Market research links to all four key concepts. It is a response to change, because tastes, technology and competitors keep moving, and it feeds creativity when it shows needs that no product yet meets. It raises questions of ethics whenever it collects personal data, and it can support sustainability by measuring what customers will pay for greener products before a business commits to making them.

What market research is

A business that knows its customers well can design products they want, set prices they will pay and reach them through the media they use. Market research is the organised way of finding this out. It turns questions such as "Who would buy this?" or "Why are sales falling?" into data that managers can act on.

Some textbooks distinguish market research (research into a particular market) from marketing research (research into any part of marketing, such as the effect of an advertisement). In practice the two terms are used interchangeably.

The information market research provides

Market research answers four broad groups of questions. Each group supports a different set of decisions in the marketing plan.

Large organisations often employ specialist research agencies, such as Ipsos, Kantar or NielsenIQ, which run surveys and tracking panels in many countries. Small businesses usually do their own research, cheaply and informally: a café owner who counts how many people pass the door at lunchtime is carrying out market research.

Why organisations carry out market research

Market research costs time and money, so a business carries it out only when the information is worth more than its cost. The value of research comes from the better decisions it allows. Six main purposes stand out.

To reduce the risk of decisions

Launching a product, entering a new country or changing a price can cost millions. Research reduces the risk of failure by testing the assumptions behind a decision before money is committed. It cannot remove risk completely, because customers may behave differently from what they say, and markets can change after the research is done.

The launch of New Coke in 1985 shows both the purpose and its limits. Coca-Cola changed its formula for the first time in 99 years after taste tests in which, according to the company, nearly 200,000 consumers preferred the new recipe. The tests measured taste but not the emotional attachment people felt to the original drink, and after protests the original formula returned as Coca-Cola Classic on 11 July 1985, 79 days after the launch.

Sources: ABC7/6abc News, "Beverage Backlash: History of New Coke" (2019), quoting The Coca-Cola Company.

To identify customer needs and wants

A business with a market orientation starts from what customers need and then designs products to meet it. Research shows those needs, including ones customers find hard to put into words. It also shows how needs differ between groups, which is the basis for dividing a market into segments and choosing which to target.

To test new products and ideas

Before a full launch, a business can test a concept, a prototype, a name, a price or a package with potential customers. A further step is test marketing: selling the product in one region or one group of stores for a limited time and measuring the results before deciding whether to launch it everywhere.

Test marketing gives evidence of real buying behaviour rather than stated intentions, but it is slow and it shows the product to competitors, who may copy it or react before the full launch.

To monitor competitors

Research into competitors’ prices, products, promotions and market share helps a business decide how to position itself and respond to rivals. Much of this comes from secondary sources such as competitors’ websites, annual reports and media coverage, and some from observation, such as visiting rivals’ stores.

To measure the effectiveness of marketing

After a campaign or a change to the marketing mix, research measures what it achieved: changes in brand awareness, website visits, sales or customer satisfaction. This allows managers to judge whether the marketing budget was well spent and to improve the next campaign.

To forecast and plan

Information on market size and growth, combined with the business’s own sales records, allows managers to forecast future sales. Those forecasts shape production, staffing, cash flow and the marketing budget.

The market research process

Good research follows a sequence of steps, each building on the one before. Skipping a step, especially the first, is a common reason for research producing data that answers the wrong question.

The diagram shows six stages:

1. Define the problem and the research objectives. A vague problem ("sales are falling") becomes precise objectives ("find out why customers aged 18–24 visit less often than a year ago").

2. Plan the research. Decide what information is needed, whether existing secondary data can provide it, which primary methods to use, who to ask (the sample), the budget and the timescale.

3. Collect the data. Carry out the survey, interviews, focus groups or observations, or gather the secondary sources.

4. Analyse and interpret the data. Summarise the numbers (percentages, averages, charts) and look for patterns and themes in opinions.

5. Present the findings. Report the results to decision-makers clearly, with their limitations, often with charts or an infographic.

6. Make and review the decision. Managers act on the findings and later check whether the decision worked, which often raises new questions.

Primary and secondary research

Every market research method collects either new information or information that already exists. This gives the first and most important classification of research methods.

Primary research is designed by the business for its own question, so the data is up to date and exactly relevant. It is also expensive and slow, because the business must design the research, find people to take part and analyse the results itself.

Secondary research is quicker and cheaper, and it is usually the first step in any project: there is little point paying for a survey to find a fact that a government statistics office already publishes free of charge. Its weakness is that the data was not collected for this business’s question, so it may be out of date, too general or of unknown accuracy.

A business’s own records are a further secondary source, sometimes called internal secondary data: past sales figures, customer complaints, loyalty card records and website statistics. They are cheap to use and specific to the business, but they say nothing about people who are not yet customers.

Primary research methods

Four methods of primary research are used most often: surveys, interviews, focus groups and observations. Each suits a different kind of question, and many projects combine two or more.

Surveys

A survey asks a number of people the same set of questions so that their answers can be compared and counted. The set of questions is a questionnaire. Surveys can be carried out face to face (in the street or a store), by telephone, by post or, most often today, online through email links, websites, apps and social media.

The quality of a survey depends largely on its questionnaire. A well-designed questionnaire produces answers that are clear, comparable and honest.

Open and closed questions

Questionnaires use two kinds of question. A closed question offers a fixed set of answers (yes/no, multiple choice, a rating scale), so the results are easy to count and compare. An open question lets the respondent answer in their own words, which gives richer detail and reasons but takes longer to analyse.

A well-designed questionnaire is short, starts with easy questions, uses simple and neutral wording, asks one thing per question and is tested on a few people (a pilot) before it is sent out. A leading question, which pushes the respondent towards an answer ("Our new menu is excellent, isn’t it?"), produces results that overstate customers’ satisfaction and so mislead managers.

Evaluation of surveys

Interviews

An interview is a conversation, face to face, by telephone or by video call, in which a trained interviewer asks one person questions. Interviews may be structured (a fixed list of questions, like a spoken questionnaire), semi-structured (key questions with freedom to follow up) or unstructured (an open conversation around a topic). In-depth interviews of 30–60 minutes are common in research with business customers, where a few buyers account for most sales.

Focus groups

A focus group brings together a small group of people, typically six to ten who share some characteristic of the target market, for a discussion led by a moderator. The group may taste a product, look at advertisements or packaging, or talk about how they use a service. The discussion is usually recorded, and the business’s managers sometimes watch through a one-way mirror or a video link.

The main strength of a focus group is the interaction: one participant’s comment prompts others to agree, disagree or add ideas, which can bring out attitudes that a questionnaire would never reach. The main weakness follows from the same feature. A confident participant can dominate, others may go along with the group (groupthink), and a group of eight people cannot be taken to represent a whole market.

Observations

Observation means watching and recording what people actually do, rather than asking them. A retailer may watch which way shoppers turn when they enter a store, how long they spend at a shelf and which products they pick up and put back. Online, website and app analytics record clicks, searches and abandoned shopping baskets, a digital form of observation carried out on a huge scale.

Two forms of observation are widely used in marketing. In ethnographic research, researchers spend time with people in their homes, workplaces or shops to understand how a product fits into their daily lives. In mystery shopping, a researcher poses as an ordinary customer to observe and rate the service a business, or its competitor, provides.

Observation is valuable because people often cannot or will not describe their own behaviour accurately. A shopper may say they always compare prices, while observation shows they pick the first brand they see. LEGO’s research for its Friends range relied heavily on watching girls play at home, which showed preferences that the girls themselves did not put into words.

YouTube

Comparing the primary methods

No primary method is best in every situation. The right choice depends on the research objective: measuring how many people hold a view calls for a survey; understanding why calls for interviews or focus groups; and finding out what people actually do calls for observation.

Secondary research methods

Five main sources of secondary data are open to businesses: market analyses, academic journals, government publications, media articles and online content. They differ in cost, reliability, how up to date they are and how closely they match a business’s needs.

Market analyses

A market analysis is a report on a whole industry or market, usually produced by a specialist research company such as Mintel, Euromonitor International or GlobalData and sold to businesses. A typical report gives the size and growth of the market, the market shares of the leading firms, the main trends in consumer behaviour and forecasts for the next few years.

Market analyses save a business the cost of collecting market-wide data itself, and they are prepared by specialists. They can, however, cost thousands of dollars each, they are available to competitors too, and they cover a whole market rather than one business’s particular customers.

Academic journals

Academic journals publish research by universities and research institutes, checked by other experts before publication (peer review). Journals such as the Journal of Marketing or the Journal of Consumer Research report studies on how consumers make decisions, respond to prices or react to advertising. Their findings are rigorous, but they are often written for specialists, may be several years old and may study general behaviour rather than a particular market.

Government publications

Governments and international organisations publish large amounts of free data: population censuses, household spending surveys, earnings, inflation, trade and internet use. National statistics offices, such as the Office for National Statistics in the UK, the Kenya National Bureau of Statistics or the US Census Bureau, and international bodies such as the World Bank and the International Telecommunication Union (ITU) publish these figures. Such data is usually reliable, based on very large samples and free, but it is general rather than product-specific and is often published a year or more after it was collected.

The chart shows how a business can use free secondary data before it plans primary research. A company planning an online survey of consumers in Kenya can see that only about 35% of the population used the internet in 2024. An online survey would therefore reach mainly younger, urban and higher-income people, and the results would be biased unless the business also used face-to-face or telephone methods.

Media articles

Newspapers, magazines, trade journals, television and radio report on companies, markets and consumer trends. Trade magazines, written for people working in one industry (grocery retailing, hotels, technology), are especially useful for news about competitors, new products and changes in regulation. Media articles are cheap and up to date, but journalists may simplify, and some articles reflect the views of the publisher or of the companies that supplied the story.

Online content

The internet holds a huge range of secondary information: competitors’ websites and prices, company annual reports, customer reviews, social media posts, blogs, forums and search data such as Google Trends, which shows how interest in a search term changes over time. Online content is fast and often free, and reviews and social media show what customers say about products in their own words.

The weakness of online content is reliability. Anyone can publish online, reviews can be fake, and people who post comments are rarely typical of all customers, since the most satisfied and the most annoyed are the most likely to write. A business using online content checks who published it, when, and whether other sources agree.

Evaluation of secondary sources

Comparing primary and secondary research

Primary and secondary research are usually used together. Most projects begin with secondary research, which is cheaper and gives the background (market size, trends, competitors), and then use primary research to answer the questions that existing data cannot.

Qualitative and quantitative research

The second classification of research concerns the type of data collected, not where it comes from. Some questions need numbers (how many, how often, how much); others need explanations (why, how, what people feel).

Quantitative research usually uses closed questions and large samples, so its results can be expressed as percentages, averages or charts and compared over time. It answers "what" and "how many" questions well, but it gives little insight into the reasons behind the numbers.

Qualitative research answers "why" questions. It can show what customers like about a brand, what puts them off a product or how they describe a problem, in their own words. Because samples are small and answers are interpreted by the researcher, its findings cannot be generalised to a whole market with confidence.

The New Coke taste tests were quantitative: they measured how many people preferred one taste to another. What they missed was qualitative, the feelings and loyalty attached to the original brand. A smaller programme of interviews or focus groups about what Coca-Cola meant to its drinkers might have shown that changing the formula carried a risk that taste scores could not measure.

Sampling

A business can rarely ask every customer or potential customer for their views. It asks a smaller group and uses their answers to draw conclusions about everyone. How that smaller group is chosen decides how far the conclusions can be trusted.

Population, sample and sample size

The population is the whole group the research is about, such as all the customers of a bank, all teenagers in a city or all the restaurants in a region. It is often called the target population.

A sample is the part of the population that is actually asked or observed. Studying a sample is quicker and cheaper than studying the whole population, which is called a census. A national census of every household, run by a government every ten years or so, shows how expensive studying an entire population is.

The sample size is the number of people in the sample. A larger sample usually gives more reliable results, but each extra respondent costs time and money, and beyond a certain point the gain in accuracy is small. The way the sample is chosen, its sampling method, matters at least as much as its size.

To choose a sample, researchers often need a sampling frame: a list of the members of the population, such as a customer database, a school register or an electoral roll. If the list leaves out part of the population, every sample drawn from it will too.

Random sampling

In random sampling, every member of the population has an equal chance of being selected. Researchers give each member of the sampling frame a number and use a random number generator, or a computer, to pick the sample. Because chance alone decides who is chosen, the researcher cannot favour any group.

Quota sampling

In quota sampling, the population is divided into groups (by age, gender, income or location, for example) and the researcher is given a quota: a fixed number of people to find in each group, usually in proportion to the group’s share of the population. Within each group, the interviewer chooses who to ask, often by stopping people in the street until the quota is filled.

Quota sampling is the method most commercial street and online surveys use. It is cheaper and faster than random sampling and needs no list of the population, only knowledge of how the population divides into groups. Because the interviewer, or the willingness of people to join an online panel, decides who is chosen within each group, it is not random and can be biased.

The UK general election of 2015 shows the risk. Most opinion polls, run by phone and online with samples matched to the population by age and gender, predicted a near tie between the two main parties; the Conservatives won by 36.9% to 30.4%. An independent inquiry found that the samples had systematically over-represented Labour voters, because the people who were easiest to reach and most willing to take part were not typical of all voters, even within each quota.

Sources: BBC News, "Election 2015: Polling failure caused by unrepresentative samples", 19 January 2016.

Convenience sampling

In convenience sampling, the researcher asks whoever is easiest to reach: friends and family, people passing a stall, the business’s social media followers or the first people to click on a survey link. It is the quickest and cheapest method and is widely used by small businesses and in early, exploratory research.

The figure shows the population of 40 people, of whom 12 (30%) are under 30. The random sample happens to include only 2 people under 30, a reminder that chance can still produce an unbalanced small sample. The quota sample matches the population’s age split exactly, but the interviewer filled it with the first people reached. The convenience sample is the cheapest but, because the researcher stood where younger people gathered, half of it is under 30.

Other sampling methods

Three further methods are used in market research. They extend the ideas of random and quota sampling.

Stratified sampling divides the population into groups, as quota sampling does, but then selects people randomly within each group. It combines the representativeness of quotas with the fairness of random selection, but it needs a full list of each group.

Cluster sampling is used when the population is spread over a wide area. The researcher randomly selects a few areas, such as towns, schools or stores, and then surveys people only within them. It saves travel costs, but the chosen clusters may not be typical of the whole population.

Snowball sampling starts with a few suitable respondents and asks each of them to recommend others. It is useful for reaching small or hidden groups, such as owners of rare classic cars or specialist professionals, but the sample consists of people who know each other and may share the same views.

YouTube

Sampling error and bias

Results from a sample are an estimate of the population’s views, and an estimate can be wrong for two different reasons: chance and bias.

Sampling error cannot be avoided completely, but it gets smaller as the sample gets larger, and for a random sample it can be estimated. A survey of 1,000 randomly chosen people gives a much more precise estimate than a survey of 100, as long as both samples are chosen without bias.

Bias is a systematic error that pushes results in one direction. It comes from the way the research is designed or carried out, so a larger sample does not remove it.

• Selection bias: the sample leaves out or under-represents part of the population, for example an online survey in a country where most people are offline.

• Non-response bias: the people who reply differ from those who do not, for example when only the most loyal or the most dissatisfied customers answer.

• Response bias: respondents give inaccurate answers, because of leading questions, a wish to please the interviewer or embarrassment.

• Interviewer bias: the interviewer’s manner, wording or choice of whom to approach influences the results.

The best-known example is the 1936 US presidential poll by the magazine The Literary Digest. It mailed about 10 million mock ballots, mostly to its readers and to car and telephone owners, and received about 2.4 million replies, which predicted a clear win for Alf Landon. Franklin D. Roosevelt won in a landslide. Later research found that non-response bias was the main cause, because voters who disliked Roosevelt were much more likely to reply, while George Gallup predicted the winner correctly with a far smaller sample.

Sources: Squire, P. (1988), "Why the 1936 Literary Digest Poll Failed", Public Opinion Quarterly 52(1); Freedman, D., Pisani, R. and Purves, R. (2007), Statistics, 4th edition.

Choosing a sampling method

The right method depends on what the research is for and on the resources available. Random sampling is most accurate but most demanding; convenience sampling is least demanding but least reliable; quota sampling is the usual compromise in commercial research.

The ethics of market research

Market research collects information about people, often personal, and sometimes without their full awareness. This makes ethics central to how it is carried out. Research that deceives or harms people damages the trust between a business and its customers, and in many countries it is also illegal.

Privacy and data protection

Businesses must protect the personal data they collect, keep it securely, use it only for the purpose stated and delete it when it is no longer needed. In the European Union and the UK, the General Data Protection Regulation (in force since May 2018) sets these rules and allows large fines; many other countries, including Brazil, India and South Africa, have similar laws. Ethical researchers also collect only the data they need and anonymise results so that individuals cannot be identified.

Informed consent

People taking part in research should know who is collecting the data, what it is for and how it will be used, and should agree freely to take part. This is informed consent. It matters most when the data is sensitive (health, finance, religion), when people are observed or tracked online, and when the participants are children, whose parents or guardians must give consent.

The Cambridge Analytica scandal shows what happens when consent is missing. A researcher’s personality quiz app collected the Facebook data of up to 87 million people, including friends of quiz users who had never downloaded the app, and some of the data was used for political advertising. In October 2018 the UK Information Commissioner’s Office fined Facebook £500,000, the maximum under the law then in force, for allowing developers access to people’s information "without sufficiently clear and informed consent".

Sources: BBC News, "Facebook fined £500,000 for Cambridge Analytica scandal", 25 October 2018.

Honesty and the use of findings

Two further ethical issues concern honesty. Selling under the pretence of research ("sugging"), where a call or email presented as a survey turns into a sales pitch, misleads people; fundraising in the same way is called "frugging". The Market Research Society’s code of conduct bans both. Research findings can also be misused: a business may publish only the favourable results of a survey, or use a tiny sample to support an advertising claim. ESOMAR, the global association of research professionals, publishes a similar international code that requires honesty, consent and the protection of participants.

Sources: Market Research Society, Code of Conduct (2023), rule 4.

Ethics is only one of the key concepts that market research touches. The table summarises how each of the four applies.

Evaluation: why and how organisations carry out market research

Market research gives managers evidence rather than opinion, but evidence of a particular kind: what a sample of people said or did at a particular time. Its value depends on how well it is designed and how carefully its results are interpreted.

Benefits

• Reduces the risk of costly failures by testing ideas before money is committed.

• Keeps the business close to its customers’ changing needs, supporting a market orientation.

• Provides evidence that can persuade investors, lenders and senior managers to back a plan.

• Measures the results of marketing spending, so budgets can be moved to what works.

• Shows competitors’ strengths and weaknesses, helping a business find a distinctive position.

Limitations

• Cost and time. Good primary research is expensive and slow; by the time results arrive, the market may have moved or a competitor may have launched first.

• Reliability. Results are only as good as the sample and the questions. Biased samples, leading questions and small groups produce confident but wrong conclusions.

• Stated versus actual behaviour. People often say they will buy a product, or pay more for an ethical one, and then do not.

• Past views may not predict future behaviour. Research describes current or past views; it may not predict reactions to a product that is genuinely new, because customers find it hard to imagine what they have never seen.

• Interpretation. The same data can support different conclusions, and managers may use research to confirm a decision they have already made.

What the value of market research depends on

Whether research is worth its cost, and which methods suit, depends on the situation. The most important factors are:

• The size of the risk. A multinational launching a new product line can justify years of research; a market stall testing a new flavour can simply try it for a week.

• The budget and size of the business. Small businesses rely on secondary sources, their own records, convenience samples and observation; large ones can afford random samples, agencies and ethnographic studies.

• The type of question. "How many?" needs quantitative research with a representative sample; "why?" needs qualitative methods.

• The type of market. In business-to-business markets with few customers, in-depth interviews with the main buyers may be enough; consumer markets need larger samples.

• How fast the market changes. In fast-moving markets such as fashion and technology, quick, cheap research repeated often may be more useful than one slow, thorough study.

• How new the product is. Research tests improvements to existing products well; for radical innovations, observation of behaviour and test marketing tell managers more than asking people what they want.

In most cases the strongest approach is a combination: secondary research for the background, qualitative research to understand customers and shape the questions, and quantitative research on a well-chosen sample to measure how far the findings apply. Managers still have to judge what the results mean, so the businesses that gain most from research treat its findings as evidence to weigh alongside their own experience, with the limits of the sample and the method in mind.

Case study: LEGO Friends and research into how girls play

By the late 2000s the LEGO Group, the Danish toy maker, had recovered from a financial crisis in the early 2000s by focusing on construction sets that mainly appealed to boys, such as LEGO City and LEGO Star Wars. Previous attempts to attract girls had either failed or been only modestly successful, and LEGO’s managers wanted to understand why before trying again.

How LEGO researched the market

Starting in 2007, LEGO sent small teams of designers, marketers and outside consultants to shadow girls and interview their families over a period of months in Germany, South Korea, the UK and the USA. The approach was mainly ethnographic rather than based on focus groups. The teams watched girls play in their own homes and talked to them and their parents about their toys and routines. Including design and testing, the project took four years.

The research produced qualitative findings that a questionnaire would have been unlikely to show. Girls did like to build, but differently from boys: they preferred "stops along the way", starting to tell stories and rearrange the model before it was finished, rather than building straight through to match the picture on the box. Many wanted a figure they could identify with, which the square LEGO minifigure did not offer. And they cared about beauty, which LEGO’s researchers broke down into harmony, friendlier colours and a high level of detail.

The greatest concern for girls really was beauty.

— Hanne Groth, LEGO market research manager, quoted in Bloomberg Businessweek, "Lego Is for Girls", December 2011

Each finding shaped the product. The Friends sets, set in a fictional town called Heartlake City, came with pieces bagged so that girls could start playing before finishing the whole model, six new colours (bright pink was already in the LEGO palette), and a new "mini-doll" figure, taller and more detailed than the minifigure, with five named main characters and backstories. LEGO launched 23 products in the UK on 26 December 2011 and in the USA on 1 January 2012, after the Christmas rush, so that retailers would give the range its own shelf space, supported by a US$40 million global marketing campaign.

The results

LEGO Friends sold far more than forecast. By mid-2012 LEGO had sold twice as many Friends sets as it expected for the first half of the year and had to increase production, and in its 2012 annual results it said the range had performed twice as well as expected. In the USA, sales of the range were about three times higher than LEGO had expected. Group revenue rose by 25% in 2012, with LEGO naming Friends as one of the main reasons alongside LEGO Star Wars, LEGO City and LEGO Ninjago.

The chart shows LEGO’s revenue and net profit in the years around the launch. The 2012 figure is shown as first reported; LEGO’s 2013 report restated 2012 revenue as DKK 23,095 million. The research did not create the growth on its own, but it allowed LEGO to enter a segment it had failed to reach several times before, at a lower risk than a launch based on guesswork.

Criticism and limits

The research-led design was also criticised. Campaign groups said the pink and purple sets, the beauty salon and the café reinforced gender stereotypes, and a US petition by the SPARK movement asking LEGO to change its marketing gathered more than 50,000 signatures. The criticism points to a tension in research-led design. By designing around what girls in 2007–2011 said and did, LEGO risked reinforcing the patterns that earlier toy marketing had helped to create.

The case shows the strengths of qualitative, observational research. Watching children play in their own homes showed needs that girls could not have described in a survey, and it gave designers specific, usable ideas. It also shows the limits: the research was slow and costly, it described current behaviour rather than how attitudes might change, and good research did not settle the ethical question of whether the business should simply give customers what they appear to want.

Sources: Bloomberg Businessweek, "Lego Is for Girls", 14 December 2011; The Guardian, "Lego’s ‘sexist’ Friends range for girls spurs 35% profit rise", 31 August 2012; LEGO Group interim results press release, August 2012; The Copenhagen Post, 21 February 2013; LEGO Group press releases on its annual results for 2011 (1 March 2012), 2012 (21 February 2013) and 2013 (27 February 2014); Associated Press, 28 February 2013.

• Explain why LEGO chose ethnographic observation rather than a large survey to research how girls play.

• Explain one limitation of LEGO’s research in drawing conclusions about girls in all of its markets.

• To what extent does the success of LEGO Friends show that market research reduces the risk of launching a new product?

Exam practice

Market research appears in Paper 1 and Paper 2 at SL and HL, usually as short structured questions (define, explain, distinguish, calculate a quota) and as part of 10-mark extended responses on marketing decisions, where you are asked to evaluate or recommend research methods for the business in the case.

Definitions earn full marks when they are precise and not circular: "a focus group is a group that focuses on a product" earns nothing.

A calculation like this is often followed by a part asking why the quota sample might still be unrepresentative.

Distinguishing questions ask for the difference between two concepts, with both sides made explicit.

Summary

• Market research collects and analyses information about customers, markets and competitors to reduce the risk of marketing decisions.

• It follows a process: define the problem and objectives, plan, collect, analyse, present, then decide and review.

• Primary research collects new data (surveys, interviews, focus groups, observations); secondary research uses existing data (market analyses, academic journals, government publications, media articles, online content).

• Quantitative research measures how many; qualitative research explains why. Source (primary or secondary) and type (quantitative or qualitative) are separate classifications.

• A sample represents a population. Random sampling gives everyone an equal chance; quota sampling fixes the number from each group; convenience sampling takes whoever is easiest to reach.

• Sampling error comes from chance and falls as the sample grows; bias is systematic and is not cured by a larger sample.

• Ethical research respects privacy and data protection, obtains informed consent and reports findings honestly.

• The value of research depends on the size of the risk, the budget, the question, the market and how new the product is; combining methods is usually strongest.

Self-check

• To do:

  I can define market research, primary research and secondary research.

• To do:

  I can explain why organisations carry out market research and describe the market research process.

• To do:

  I can explain the advantages and disadvantages of surveys, interviews, focus groups and observations.

• To do:

  I can explain the advantages and disadvantages of the five secondary sources.

• To do:

  I can distinguish between qualitative and quantitative research and give an example of each.

• To do:

  I can explain quota, random and convenience sampling, and the difference between sampling error and bias.

• To do:

  I can calculate the quota for each group in a quota sample.

• To do:

  I can evaluate the research methods an organisation uses, including their ethical issues, and recommend a suitable combination.