A step-by-step guide to market research processes—and how to roll them out for your brand. Learn how to get the data, then make the most of it.
Say what you will about McDonald’s, but as a cultural anthropologist, one of the things I respect most about their brand is the international menu concept.
From maple and bacon poutine in Canada and gazpacho in Spain to India’s McPaneer Royale, McDonald’s knows how to give the people what they want.
And how do they give a global brand local appeal? By gaining a deep understanding of the consumers in every market they plan to enter.
If you’re thinking about doing consumer insights research, you should be familiar with market research processes. That will help you set expectations and ensure you’re getting the right people and tools for the job.
First, let’s start with the basics. What is market research and how is different from marketing research?
People often confuse market research and marketing research. Aren’t they just different words for the same thing?
ESOMAR, the global research and data association, and the American Marketing Association would disagree. Here’s the gist:
Market research emphasizes the process of collecting consumer data, while marketing research refers to the product of that information and/or a function within an organization.
Essentially, you might be looking for a marketing researcher to conduct market research. Market research will help you answer questions about your customers, your competitors, or current and potential markets.
To the uninitiated, market research can seem like a mystery.
However, market research processes are quite systematic—well, in theory. In practice, the steps involve exploration, creativity, and abstraction.
Market research is just one of those things you don’t actually get until you do it. But there are a few steps you can follow to make it a bit easier.
Researchers are curious people. That’s why every research project starts with a question.
What is the part of your business you want to know more about? Identifying the problem is the most important step in market research processes. It’s going to determine every step you take in the future—of market research anyways.
Not sure where to start? Here are a few tips:
• Look for marketing challenges or opportunities. Maybe your brand awareness could use a boost, you’ve noticed declining customer loyalty, or you’re considering opportunities in emerging markets.
• Frame it as a question. Why is customer loyalty decreasing? How can we enter the market for luxury hotels? What does our customer’s typical path-to-purchase look like?
• Determine what type of problem you have. In market research, a problem can be ambiguous, clearly defined, or somewhere in the middle. Do you know the variables and factors influencing what you want to measure? This is important as it will influence your overall research design, which is up next.
There are three types of research designs. The design you choose will be informed by how well-defined your problem is.
If you don’t know much about the problem, you need:
Exploratory research. If you don’t know the major variables or factors at play, your research is ambiguous. Exploratory research can help you develop a hypothesis or ask a more precise question.
If you have a vague idea about what’s important to solve the problem, you need:
Descriptive research. Descriptive research does what it says on the tin: it describes a certain phenomenon or the characteristics of a population. It can build on exploratory research but doesn’t give insight into the how, when, or why.
Descriptive research is useful for parsing out market segments and measuring performance. Consequently, you need a pretty good idea of what you’re measuring and how it will be measured. If you want to know how cause and effect are linked, you need:
Causal research. Market researchers conduct causal research when they want to understand the relationships between two or more variables. Simply, causal research helps you understand cause and effect.
Data is the essence of market research. At the end of these market research processes, data is analyzed, interpreted, and turned into information and actionable insights.
Data can be qualitative or quantitative. Simply, qualitative data can take many forms, from descriptions to audio and video. Quantitative data is typically presented in values and figures.
Choosing your sample
First, you must select the population you want to study. A population is a group with some shared characteristic that you’re interested in gathering data from. It can be broad (Canadians) or narrow (independent gym owners in Chicago).
No matter how small or large your population, it’s unlikely you’ll be able to work with all individuals.
The key to choosing a good sample is that it is representative. That means the people you select to participate (the sample) should reflect the larger group you’re studying.
There two forms of data you can collect: primary data and secondary data.
Primary data is gathered specifically for your project. Secondary data has already been collected, either internally or externally through government agencies, consulting or market research firms, websites, on social networks, and so on.
Depending on your research design, you may want to check internally for secondary data. For example, let’s say you’re trying to understand the annual purchase cycle for your business. You would gather sales and reports and company records—that is secondary data.
But of course, secondary data still needs to be prepared for analysis
There are two ways to collect primary data: directly or indirectly. Direct data collection is just that—you are speaking to your participants directly. That can be through surveys, interviews, focus groups, and so on. Indirect data collection typically means observation. Think in-store observation, shelf experiments, or website heatmaps.
Data analysis is a process of looking for patterns in data and trying to understand why those patterns exist. Data can be analyzed quantitatively or qualitatively.
Quantitative data analysis is a process more complicated than can be described here. Unless you’re a math whiz, you’ll probably just use a data analysis software like SPSS or StatCrunch.
Qualitative data analysis typically involves coding—but not the computer programming kind, don’t you worry. This type of coding can be done by hand or using software such as NVIVO. It involves looking for themes, concepts, and words that repeated throughout the data.
Interpretation involves answering the question:
What does the data tell me about what I wanted to know?
That’s where themes and patterns come in. You can describe trends and present them using figures or descriptions drawn from your participants.
Part of interpretation is using what you know about customers, businesses, or markets to provide recommendations for how to move forward. These data-driven suggestions should offer a solution to the initial problem. The results of the research can also bring to light a problem you weren’t even aware you had.
Market researchers are able to draw on a large toolbox of market research methods. Typically, they fall into the category of qualitative or quantitative because of the type of data they produce.
Qualitative market research methods offer the insights into the why behind the behavior. Here are a few of the most common:
Focus group. A market research technique that involves a group discussion about certain topics led by a moderator.
Best for: Exploratory research
In-depth interviews. Interview conducted with an individual aimed at getting deeper insights about attitudes, motivations, or experiences.
Best for: Descriptive research
Ethnography. Also known as participant observation, it involves spending time with participants in their natural environment (as opposed to a lab setting). Best for: Descriptive research Observational. Carefully watching people to understand what they’re doing. It allows you to learn about consumer or employee behavior, but not the motivation behind.
Best for: Exploratory research Discourse analysis. This is a fancy way of saying ‘analyzing what people say’. Social listening is a form of discourse analysis. Examining customer reviews, help transcripts, social media comments, and more, are all forms of discourse analysis.
Best for: Exploratory or descriptive research
Quantitative market research techniques typically yield quantitative data: facts, figures, statistics. Here are some you might consider:
Surveys. Surveys are the crux of market research. They involve collecting facts, figures, and opinions using a questionnaire. Surveys may seem simple, but there are a lot of factors that can turn good intentions into bad data—be sure to read our tips on the right question types to ask.
Best for: Descriptive research. Surveys can also yield qualitative data if participants write out answers.
Structured observation. Observation research can also be quantitative if you are observing participants without direct involvement and assigning values to certain behaviors.
Best for: Descriptive research
A/B Testing. Also called split testing, this is a way to compare responses to a variation of a single variable to see which performs better. For example, presenting users with to versions of an ad to see which gets more clicks.
Best for: Descriptive research Experiment. Marketing experimentation typically involves manipulating a variable to see how it influences behavior. They can be conducted in a lab or in the field. Best for: Causal research
Time to put this into practice. Let’s look at market research examples of various types of research designs. Here goes.
Mobile phone company HTC wanted to understand how they could improve the user experience of their phones. This problem requires exploratory research because there wasn’t a specific feature they wanted to test. They simply wanted to learn more from their customers.
With ethnographic research, they observed how participants interacted with their phones. They looked for challenges people had with everyday usage. After analyzing these pain points, they added new functions to their next model that made the phones easier to use.
Company ABC wanted to understand how large the market for vegan cheese is in Canada. They have a somewhat defined research problem: What is the potential market share for vegan cheese?
In order to provide an answer, market researchers will have to describe various characteristics: who the customers are, why they buy vegan cheese, competitor market penetration, and potential opportunities.
This requires mixed-method research. The researchers might collect secondary data on the number of vegans in Canada or how much vegan cheese is sold in the country and through which companies. They may also conduct focus groups to understand what motivates people to buy vegan cheese.
Once complete, they will be able to present statistics on vegans in Canada and estimate Company ABC’s potential market share.
Causal research requires keeping on variables and conditions the same, save for the one you are testing. German marketing and sensory research company isi is a company that runs both field and lab experiments.
They worked with a chocolate bar company to design an experiment that tested 12 different chocolate bar recipes.
The consumers sequentially tested the recipes and provided ratings (quantitative data) and descriptions (qualitative data) of each one. The result was that consumers were most satiated by “a firm, tough texture and a higher amount of caramel and peanuts.”
One thing to remember is that market research is an iterative process. You can keep using what you learned to conduct better studies, evaluate the changing market conditions, and the whims of consumers.