AP® Business with Personal Finance review sheet from Aim for Five (aimforfive.com/business-finance/units/2/2-3)
Unit 2 · Topic 2.3
2.3 Market Research
Market research is how a business gathers evidence before it spends serious money. It uses cheaper secondary research first, then primary research to test specific hypotheses, chooses the right method for the data it needs and presents results with the right kind of chart.
Key terms
- quantitative vs. qualitative data
- secondary research
- primary research
- business hypothesis
- A/B testing
- biased sample
Why businesses do market research
Market research means collecting detailed information about markets, products and how customers behave, to guide marketing decisions. It produces two kinds of data. Quantitative data is numbers: how many, how much, how often. Qualitative data is descriptions in words or images: why and how.
Before launching a new product, a business uses research to check three things:
- Desirable: customers want it, because it creates value and solves their problem (problem-solution fit).
- Feasible: the business can actually make and deliver it with its resources, technology, expertise and time.
- Viable: it can earn a profit in the market.
Secondary and primary research
Secondary research uses information someone else already gathered: government data, industry reports, academic studies and databases. It can show market size (in dollars and in customers), trends, segments, what drives buying decisions and details about rivals. Businesses usually start here because it's much cheaper than collecting their own data.
Primary research is new data the business collects itself to test a business hypothesis, a specific assumption about customers, a product or a market. Example: teens in our town will pay at least $15 for a phone case made from recycled plastic. Testing the hypothesis before acting saves money if it turns out to be wrong.
Research doesn't stop at launch. Businesses with products already on the market keep researching to see how customers' needs are shifting, spot ideas for improvements and protect or grow their market share.
Choosing a primary research method
An A/B test is a kind of experiment: half the visitors to a website see version A of a page and half see version B, and the business compares results.
| Method | Best when you need | Data you get |
|---|---|---|
| Survey | Lots of responses that reflect a large group | Mostly quantitative |
| Interview or focus group | Deep detail from a few people, with follow-up questions | Qualitative |
| Observation | How people actually behave in a natural setting | Behavior, not opinions |
| Experiment | How people behave in a controlled setting | Behavior, not opinions |
| A/B test | Real customers' reactions to two workable options | Which option performs better |
Avoiding skewed data
Results are only useful if they reflect the real population. Use a sample that's big enough and that matches the people you care about; surveying only your friends gives a biased sample. Ask neutral questions. Don't you agree our new flavor is delicious? pushes people toward yes; How would you rate the new flavor from 1 to 5? doesn't.
Even good research can't predict everything. Businesses often have to act on limited, unclear or contradictory data.
Showing the results
- Bar chart: comparing separate values, like each year's sales or revenue for several companies.
- Stacked bar chart: a total and its parts at once, like yearly sales split by product line.
- Line graph: a trend over time, like customers per year.
- Pie chart: parts of a whole, like each company's share of a market.
Worked examples
Try each one yourself first, then open the solution.
- Example 1Calculator allowed
Reading an A/B test
An online store tests two product pages. Version A (photo of the product alone) was shown to 2,000 visitors and 90 bought. Version B (photo of a person using the product) was shown to 2,000 visitors and 130 bought. Which version should the store use, and what makes this a fair test?
Show the solutionHide the solution
- Step 1: Conversion rate = buyers ÷ visitors.
- Step 2: A: 90 ÷ 2,000 = 0.045 = 4.5%. B: 130 ÷ 2,000 = 0.065 = 6.5%.
- Step 3: B converted 2 percentage points more visitors into buyers.
- Step 4: It's a fair comparison because both versions reached the same number of real shoppers and only one thing (the photo) changed.
Answer: Version B, with a 6.5% conversion rate versus 4.5% for A. The test is fair because the samples were equal and only the photo differed.
- Example 2
Picking the method and the chart
A snack company wants to know why some customers stopped buying its chips, and then to show managers how its sales split among three flavors over the past five years. Which research method and which chart fit best?
Show the solutionHide the solution
- Step 1: Why is a qualitative question, and the company needs detail and follow-up questions, so interviews or a focus group fit.
- Step 2: Showing totals over five years and how each flavor contributed is a whole-and-parts comparison over time, so a stacked bar chart fits.
- Step 3: Trap: a pie chart shows parts of a whole for only one moment, so it can't show five years at once.
Answer: Interviews or a focus group for the why question; a stacked bar chart for flavor sales across five years.
Common mistakes
- Calling a survey of a few friends good evidence. Samples must be big enough and represent the target population.
- Mixing up feasible (can we make it?) and viable (can it make a profit?).
- Using a pie chart for change over time. Use a line graph or a bar chart.
On the exam
- Question 3 (Business Concept Application) asks you to describe how a business did its research, describe what the data show and explain how it could use the finding. Name the method, quote specific numbers and connect the result to a decision.
- Watch for biased survey questions and unrepresentative samples in stimulus material; questions often ask what's wrong with a study.
Connected topics
Videos
Check yourself: 2.3 Market Research
4 questions on 2.3 Market Research. Pick an answer to see if you got it, and why.
| Version | Emails sent | Customers who clicked | Customers who bought |
|---|---|---|---|
| Subject line A: "Save 20% this weekend" | 5,000 | 600 | 150 |
| Subject line B: "Your new favorite hoodie is here" | 5,000 | 450 | 180 |
Results of an email test by Northline Apparel, a hypothetical clothing brand. Customers were randomly assigned one version.
Northline's email test is best described as
What percent of the emails with subject line B led to a purchase?
Which conclusion is best supported by the data?
Which business hypothesis was Northline most likely testing?
0 of 4 answered