Home / Courses / Principles of Marketing / Unit 4

Principles of Marketing Unit 4: Marketing Research and Data

Good marketing decisions depend on good information. This unit explains how firms collect, analyze and use data about customers and markets, and how to judge whether research results can be trusted.

Covers: Marketing information systems and big data, The marketing research process, Exploratory, descriptive and causal research, Secondary versus primary data, Observation, surveys and experiments, Sampling plans, Analytics, CRM and research ethics

Study guide

A marketing information system (MIS) is the people and procedures that assess information needs, develop the needed information and help decision makers use it. Information comes from three main places. Internal databases hold data the firm already has, such as sales records, customer service logs and web analytics. Competitive marketing intelligence is the systematic gathering of publicly available information about competitors and trends, such as rival websites, job postings and trade shows. Marketing research is the systematic design, collection, analysis and reporting of data about a specific marketing situation. Today firms also work with big data, the huge and complex data sets produced by digital activity, and use analytics to find patterns in it.

The research process has four steps: define the problem and research objectives, develop the research plan, implement the plan, and interpret and report the findings. Defining the problem is often the hardest step, because a symptom such as falling sales can have many causes. Objectives come in three types. Exploratory research gathers preliminary information to define the problem and suggest hypotheses, often with focus groups or interviews. Descriptive research describes things such as market size or customer attitudes, usually with surveys. Causal research tests cause-and-effect hypotheses, such as whether a 10% price cut raises sales enough to pay off, usually with experiments.

Researchers use secondary data, information that already exists because it was collected for another purpose, and primary data, collected for the specific purpose at hand. Secondary data, such as census figures or industry reports, is quicker and cheaper but may be outdated, too general or biased. Primary data is more relevant but costs more. Primary data can be gathered by observation (watching behavior, including ethnographic research in people's natural settings), surveys (asking questions, best for descriptive information) and experiments (changing one factor while controlling others, best for causal information). Contact methods include mail, phone, personal interviews and online, each trading off cost, speed, control and response rate.

Because researchers rarely study a whole population, they draw a sample, a segment chosen to represent it. A sampling plan answers three questions: who is to be studied (the sampling unit), how many (sample size) and how they are chosen (sampling procedure). In probability samples, every member of the population has a known chance of selection; simple random, stratified random and cluster samples are examples, and they let researchers estimate sampling error. Nonprobability samples, such as convenience, judgment and quota samples, are cheaper but cannot measure sampling error. Larger samples shrink the margin of error, but only in proportion to the square root of the sample size, so quadrupling a sample only halves the error.

Interpreting results requires judgment. Correlation in survey or web data does not prove causation; only controlled experiments, such as A/B tests that randomly show two versions of a web page, can show cause and effect. Bias can come from leading questions, a sample that does not match the population, or low response rates. Customer relationship management (CRM) systems combine data from every customer touch point to build a full picture of each customer, which supports targeting and loyalty programs. With this power come ethical duties: protect consumer privacy, get consent, store data securely and never misuse research as a disguised sales pitch, sometimes called sugging.

Key ideas

• A marketing information system draws on internal databases, competitive intelligence and marketing research.
• Defining the problem is the first and often hardest step of the research process.
• Exploratory research suggests hypotheses, descriptive research describes, and causal research tests cause and effect.
• Secondary data is cheaper and faster; primary data is more specific to the current problem.
• Experiments are the best approach for causal information because they control other factors.
• Probability samples allow sampling error to be measured; nonprobability samples do not.
• Margin of error falls with the square root of sample size, so precision gets expensive.

Formulas and rules

Common mistakes

• Choosing a survey to prove cause and effect when only an experiment can do that.
• Calling information primary because it is useful; primary means collected for the current purpose.
• Assuming a large convenience sample is representative; size cannot fix a biased selection method.
• Thinking that doubling the sample size halves the margin of error; it takes four times the sample.
• Confusing a quota sample (nonprobability) with a stratified random sample (probability).

Before the test, make sure you can

• Name the three sources of information in a marketing information system.
• List the four steps of the research process and write clear research objectives.
• Match exploratory, descriptive and causal objectives to focus groups, surveys and experiments.
• Classify data as primary or secondary and weigh their pros and cons.
• Distinguish probability from nonprobability samples and name examples of each.
• Calculate response rates, margins of error and A/B test lift.
• Explain why correlation does not prove causation and identify ethical problems in research.

Words to know

Marketing information system (MIS)

People and procedures for assessing information needs, developing the needed information and helping decision makers use it.

Big data

The huge and complex data sets generated by today's information technologies, such as web, mobile and sensor data.

Internal databases

Collections of consumer and market information obtained from data sources within the company network.

Competitive marketing intelligence

The systematic monitoring, collection and analysis of publicly available information about consumers, competitors and developments in the market.

Marketing research

The systematic design, collection, analysis and reporting of data relevant to a specific marketing situation.

Exploratory research

Research to gather preliminary information that will help define the problem and suggest hypotheses.

Descriptive research

Research to better describe marketing problems, situations or markets, such as market potential or consumer attitudes.

Causal research

Research to test hypotheses about cause-and-effect relationships.

Secondary data

Information that already exists somewhere, having been collected for another purpose.

Primary data

Information collected for the specific purpose at hand.

Ethnographic research

A form of observational research in which trained observers watch and interact with consumers in their natural environments.

Focus group

A small group of people invited to talk with a trained moderator about a product, service or organization.

Experimental research

Gathering primary data by selecting matched groups, giving them different treatments, controlling unrelated factors and checking for differences in responses.

Sample

A segment of the population selected for marketing research to represent the population as a whole.

Probability sample

A sample in which each population member has a known chance of being included.

Convenience sample

A nonprobability sample in which the researcher selects the easiest population members from which to obtain information.

A/B testing

An online experiment that randomly shows users two versions of a page or message to see which performs better.

Customer relationship management (CRM)

Managing detailed information about individual customers and carefully managing customer touch points to maximize loyalty.

Sample practice questions

30 more questions like these, from level 1 to 5, are in the practice set.