market research midterm

Ernest dichter

founder of focus groups

JOHN wanamaker

pricetage, money back guarantee (return), price tag, ½ and full page ad newspaper, advertising, truth in advertising, "half the money i spend on advertising is waster, trouble is i dont know which half!"

ICEBERG PRINCIPLE

10% SYMPTOM, 90% DEEPER PROBLEM

10% The Symptom (Visible Part)

This is what you see — the obvious issue or sign that something is wrong.

SYMPTOM Example

"Sales have dropped," "Customer satisfaction is low," or "Website traffic decreased." These are symptoms, not the real problem.

90% The Deeper Problem (Hidden Part)

This is what lies beneath the surface — the underlying causes or true reasons behind the symptom.

DEEPER PROBLEM Example

Poor product quality, Ineffective advertising or targeting, Changes in customer preferences, Competitor innovations

BIG DATA

answers what not why

VOLUME

size of data within database

VELOCITY

the speed of the data

VERACITY

how much trust we have in the information

VARIETY

type of data we collect, unstructured (images), structured

VOLATILITY

(example: how relevant is former/outdated data?)

VULNERABILITY

(example: how secure is the data storage?)

VALUE

Represents the ability to turn data into meaningful insights that drive business growth, improve operations, and enhance decision-making.

Secondary data

data that already exists

Advantages of secondary data

redefines, cost efficient, uncovers research challenges, supports primary research, acts as a solution, provides sample frame

Disadvantages of secondary data

Lacks availability (can't find what you need), Inaccurate, Insufficient or irrelevant, or Outdated

Consumer data law starter

california

GDPR

Initial reason you started to see some of those pop Ups on sites for privacy/cookies/, transparency, control, enforcement

(European Union)

General Data Protection Regulation

CCPA

california consumer protection act, 2020 enacted

Federal law (industry/demo specific only)

COPPA, HIPPA, FCR, Gramm- Leach-Biley

COPPA

children's online privacy protection act (under 13

Gramm-Leach-Biley

financial services Fair credit reporting

First Party Data

Stems from a company's own database (the company's own data collection;)

Second Party Data

One company "shares" their first party data with another company

Third Party Data

Company "using" data is not the originator of the data (see Lotame article)...an external source

Zero Party Data

(consumer very intentionally provides additional information to company...example: online quiz about preferences

Qualitative

smaller sample,Can try to get to "why"consumers behave a certain way and "how they feel", Often exploratory in nature (as brands look to have consumers "open up"), May drive further quantitative research

Quantitative

larger sample size, more statistical, often explains "how much/ why, what)

Focus group fundamentals

Representative Participants, Skilled Moderators ,Well Designed Discussion Guide (Client Focused), Appropriate Resources / Setting

Advantages of focus groups

group dynamic, observation capabilities, immediacy for results.

Disadvantages of focus groups

group think/influence, artificial environment, rush results

In depth interviews advantages

Eliminate Group Pressure / Dynamic, Sensitive Topic (health, divorce, finance,etc), Allows For Unstructured Flow of Questions, Desire for Hands on / Personal Usage, Need to Speak with Unique Set of Participants (example: CEOs)

Disadvantage of in-depth interviews

time, cant leverage group energy & momentum, less direct client observation

Projective techniques

Word Associations, Sentence & Story Completion, Cartoon Tests, Third Person Technique

Word Association

. In this test, a participant is given a word and is asked to speak the first thought that comes into his mind after hearing the word.This method bypasses conscious, rational defenses and provides insight into deep-seated attitudes and values that might be difficult to articulate through direct questioning. it can produce superficial or misleading data due to the influence of group dynamics and interpretation bias

a third-party technique

is a projective technique that asks respondents to describe the thoughts, feelings, or behaviors of a hypothetical third person (like a neighbor or a friend) instead of themselves. This indirect approach helps to uncover deep-seated or sensitive opinions that people may be unwilling to admit to holding directly, as they project their own views onto another person, allowing for more honest and uninhibited responses

The mode

the type of channel/method used to conduct the survey

Examples of modes

online surveys, mobile surveys, Mall intercept, social media, mail surveys, telephone surveys, door to door.

What influences what survey mode a research uses

time, quality of date & sample, BUDGET, length/structure of questions, respondent requirements (&incidence rate)

Incidence rate

percent of people in a sample that qualify for a study

Translation

to what degree can researcher reach the intended target for the survey

Best survey practices

use sensible incentives, have a thoughtful introduction, use "other" and open end options, shorter + better, pretest

Research errors in survey research

random sampling, systematic error/bias, sample design, measurement.

Random sampling error (chance variation)

does your sample represent your target, this error can not be totally eliminated but it can be reduced

Sample design errors

frame error, population specification error, selection error.

Frame error

when the list (the sampling frame) you're using to pick your sample does not fully match the actual population you want to study. (telephone directory (use phone book = leaving out people without landlines)

Population specification error

Happens when the researcher defines the wrong population to begin with — they're not clear about who should be studied.

Selection Error

Happens when people are chosen (or choose themselves) in a biased way, so your sample doesn't represent the population properly. (messy house, in house survey, households refuse to let you in because their house is messy, excludes people who might differ from those who agreed

Measurement errors include

surrogate info, questionnaire bias, input error, non response bias, response bias

Surrogate Information Error

Collecting information that doesn't truly measure what you want to know — you're using the wrong "stand-in" (surrogate). (kellogs, wants to know healthy peoples eating habits but asks how many cereal bowls u eat a week, eating cereal is surrogate for healthy eating

Questionnaire Bias

Poor wording, order of questions, or leading phrasing that pushes respondents toward a certain answer. The way the questions are written changes the answers. The words matter.

Input Error

Technical or clerical mistakes when recording, entering, or processing responses.

Non-Response Bias

The people who don't respond to your survey might be systematically different from those who do respond. The missing voices are different in a meaningful way from those you captured.

Response Bias

When respondents' answers are inaccurate — whether intentional or not. Lying to look better, memory issues, misreporting for better outcomes.

Advantages of moving to online surveys

speed, ability to reach hard to reach targets, less expensive, personalization capabilities, large sample possibilities.

Disadvantages of moving to online surveys

quality concerns, call back/recontact challenge, may be poor approach to reaching niche sample

Open panel recruitment

anyone can join, may be solicited through ads, can build panel quickly, lacks control of who joins, does it lead to professional survey takers?

Closed panel recruitment

by invite inly, may require research company to partner with brands and companies that have large databases to recruit from and/or careful sampling methods employed to build sample

Mobile survey considerations

increased response rate, consumers can provide content easily, speed can help costs, may bring broader reach, mobile only internet users, geo location capabilities, may be preferred in general by consumers, (need to keep questions short/easy)

What is observational research

recording behaviors (or patterns of behaviors) through observation, Does NOT normally include questioning consumer, can be done through machine or human observation

Brands consider the following to determine if observational research "makes sense" to do. Often

can "it" be observed?, can the behavior be observed in a relatively short time frame?

Approaches to observation research

natural or contrived, open or disguised, human vs machine, direct vs. indirect observation

Natural

natural environment (looking/counting consumers in "real" stores)

Contrived

simulation, artificial (simulated environment, example, a lab)

Open

consumer knowns they being observed

Disguised

consumer is unaware they are part of a study

Human vs machine

who is doing the observing, machine/tech or humans?

Direct

current behaviors, can be monitored/counted

Indirect

may be past behavior, analysis of past items, photos etc.

Advantages to observation

no survey/ interviewer bias, no RECALL required by participant! (BIG ADVANTAGE)

Disadvantages to observation

often can NOT get an understanding of emotional drivers, present behavior does not always predict future behavior, can be costly, can be time consuming

Ethnography

can show how when (+explore why) people shop/buy, can show how a consumer actually uses a product, can tap into consumers ideas + creativity for a product, can reveal a products problems .I think of it as a mini documentary view on consumer behavior

Why ethnography can't be in focus groups

usually used for broad exploratory research, innovation, raw IDEAS, etc, converse focus groups may tend to be a bit more tactical and a bit further down the exploratory line

What are experiments?

-A research strategy,Determine IMPACT / EFFECT (CAUSE AND EFFECT)- Researcher is an active PLAYER (unlike observations or surveys)

Causal Research

shows that a change in one area predicts/causes a change in the other In order to infer cause, must demonstrate.... THREE ELEMENTS!!

1. Correlation

A must vary together with B in some "predictable" fashion, ●Correlation alone does NOT prove causation, ●Can be positive or an inverse correlation

2. Appropriate Order of Occurrence

Order Matters: Did A occur before B?

3. *Control/Eliminate Causal Factors*

Did some other factor help change B (other than A)?, ●(example: our weather and Taylor Swift example)

Lab Experiments...

-Allows for Control across "other factors", BUT...does NOT allow for "real" marketplace environment; unnatural, Provides Real Marketplace Environment (example: test market), BUT...can't control other influences (example: weather)

Field Experiments...

Provides Real Marketplace Environment (example: test market), BUT...can't control other influences (example: weather)

Experimental Design Terms

independent variable, dependent variable, subjects

●Independent Variable - what is changed

Example: price, promotion, etc.

Subjects

- who participates in the study

Dependent Variable

- what is measured Example: Change in Sales, Awareness, etc.

After-Only w/ Control Group (NO PRE-MEASUREMENT

Random assignment of test and control group, Measures Difference: in TEST GROUP COMPARED TO CONTROL

An A/B test

(also called a split test) is an experiment where two versions of something — version A and version B — are compared to see which one performs better.

A brand lift study measures

how a marketing campaign changes people's perceptions or attitudes toward a brand.

How it works (brand lift)

There's a test group that sees the ad and a control group that doesn't.

After exposure, both groups are surveyed on questions like

"Do you remember this brand?" (ad recall) "How likely are you to buy this product?" (purchase intent)"What's your opinion of this brand?" (brand favorability)

TEST MARKET RESEARCH

Any research that tests a new product/price/promo, etc. in a certain geographic market/region.

Why do Test Marketing?

Good estimate of sales under "live" conditions, Can identify weakness of product, ads,etc.

But there are major factors to consider(test marketing)

Cost vs. Success, Speed and competitive forces, Cost of roll-out locally vs. nationally, Risk of Test

Mean/Average

Add up all data points + divide them by the total # of data points

Mode

The most frequent: what 'shows up' the greatest number of times.

Median

The middle number/mid-point: order all the data points + pick the data point in middle

Measurement levels

nominal, ordinal, interval, ratio

very common approach basic but very effective, answers fit in one and only category, no true number value; no sense of order/rank, mode is possible, no average or medium can be performed in analysi