Not sure of the difference between CAPI and CAWI? Confused by BPTO and Van Westendorp? Perplexed by cluster analysis, factor analysis and key driver analysis? We've compiled this glossary to help you navigate the language of market research. Hope it helps!
Accompanied Shopping
A qualitative research method where participants are interviewed while they shop in a real retail environment. It combines natural observation with in-the-moment questioning to understand decision-making, behaviours and product interactions.
Bar Chart
A visual data display using horizontal or vertical bars to compare values across categories. Commonly used to show differences, rankings or trends in market research data.
Brand Equity Modelling
A data-driven approach to measuring the strength of a brand. Brand equity modelling identifies the components that contribute to brand strength and uses Key Driver Analysis to determine which attributes most strongly influence perceptions, preference and purchase behaviour.
Behavioural Science
An interdisciplinary field drawing from psychology, economics, sociology and anthropology to understand how people think and behave. In market research, behavioural science reveals subconscious motivations, decision biases, value perceptions and choice patterns. See System 1 and System 2 Thinking.
Brand Fit
A metric that evaluates how well a brand’s attributes match what a target segment values. A strong brand fit means the brand performs well on criteria the segment finds most important, helping optimise brand positioning.
Brand Mapping
A projective technique where respondents group competing brands based on perceived similarities or differences across dimensions like value, innovation, service or product range.
Brand Personalities
A qualitative technique where respondents imagine a brand as a person and describe its lifestyle, appearance, attitude and values. Helps uncover emotional and symbolic associations.
Brand Price Trade-Off (BPTO)
A pricing research technique where respondents choose between branded products at progressively increasing price levels. The exercise identifies price elasticity, competitive demand curves and optimal price thresholds.
CAPI (Computer-Assisted Personal Interviewing)
Face-to-face data collection using laptops or tablets instead of paper questionnaires. Improves accuracy, efficiency and interviewer control.
CATI (Computer-Assisted Telephone Interviewing)
A telephone interviewing method where responses are entered directly into a computerised questionnaire script, increasing speed and data quality.
CAWI (Computer-Assisted Web Interviewing)
Online surveys delivered via email lists, customer databases or online research panels. CAWI is cost-efficient and ideal for large, diverse samples.
Census
A study that gathers data from every member of the target population rather than a sample.
Confidence Interval
A statistical range that expresses where the “true” population value is likely to fall. Typically measured at 95% confidence.
Conjoint Analysis
A multivariate research technique used to determine which features, benefits or price points most influence consumer choice. Respondents evaluate different product/service combinations, enabling simulation of real-world “what-if” scenarios and optimal portfolio or pricing strategies.
Closed Question
A structured question format where respondents choose from predefined options, including dichotomous or multiple-choice responses.
Cluster Analysis
A statistical method used to create customer segments by grouping respondents based on similarities. Core to segmentation studies.
Coding
The process of converting open-ended or qualitative responses into numerical or categorical data for analysis.
Confidence Level
The probability that a statistic from a sample reflects the true value for the full population.
Continuous Research / Longitudinal Research / Tracking
Studies conducted at multiple time points to monitor trends, behaviours or performance metrics over time.
Contrived Observation
Observation of participants in a controlled environment, such as a lab or test facility.
Correlation
A statistical measure showing the strength and direction of a relationship between two variables. Correlation does not imply causation.
CX (Customer Experience)
The overall perception customers form across all interactions with a business. A key driver of loyalty, retention and advocacy.
Data Analysis Errors
Non-sampling errors caused by incorrect data entry, formatting issues or transcription mistakes during analysis.
Data Cleaning
The process of checking and correcting data for inconsistencies, errors or missing responses before analysis.
Depth Interview / IDI
A qualitative one-to-one interview designed to explore motivations, attitudes and behaviours in rich detail.
Descriptive Statistics
Summary metrics such as mean, median, mode, range and standard deviation used to describe and interpret large datasets.
Desk Research / Secondary Research
Analysis of existing information from internal or external sources to answer research questions without collecting new data.
Face-to-Face Survey
A direct interviewing method conducted in person using paper, laptops or tablets.
Factor Analysis
A statistical technique that identifies underlying themes or dimensions within large sets of variables. Useful for simplifying long attribute lists and supporting segmentation or modelling.
Gabor-Granger Pricing
A method for estimating price elasticity and optimal price points by asking respondents about purchase likelihood at different price levels.
Hall Tests
Venue-based testing where respondents are invited into a central location to evaluate products, packaging or concepts.
Impact Indices
Scores that quantify how strongly specific variables influence an outcome such as preference or satisfaction.
Key Driver Analysis (KDA)
Analysis that identifies which variables (e.g., brand imagery or product attributes) have the strongest impact on a key metric.
Linear Regression
A statistical model that quantifies how multiple variables contribute to a single dependent outcome.
Machine Learning
An AI technique where models learn from data to classify, predict or automate tasks such as text analysis and customer segmentation.
MaxDiff (Maximum Difference Scaling)
A prioritisation method where respondents identify the most and least appealing attributes from sets, producing a ranked list of importance.
Menu-Based Conjoint (MBC)
A form of conjoint where respondents build their preferred product or service from a menu of options.
Margin of Error
The range within which the true value in the population is expected to fall. Decreases with larger sample sizes.
Natural Language Processing (NLP)
AI techniques that analyse and interpret human language—for example, sentiment analysis and text classification.
Net Promoter Score (NPS)
A loyalty metric based on how likely customers are to recommend a brand. Calculated as Promoters minus Detractors.
Observation
A research method where behaviours are recorded without direct questioning, ideal for real-world behavioural insights.
Open-Ended Question
A question format allowing respondents to answer freely in their own words, generating rich qualitative insights.
Paired Interviews
A qualitative method where two related participants (e.g., partners, friends or parent–child pairs) are interviewed together.
Predictive Analytics
Statistical and machine-learning techniques used to forecast future outcomes based on historical data.
Price Sensitivity Measurement (Van Westendorp)
A pricing technique that identifies acceptable price ranges by asking consumers when a product is too cheap, too expensive, a bargain or acceptable value.
Quantitative Research
Structured research involving large sample sizes to generate statistically reliable insights.
Qualitative Research
Exploratory research designed to uncover motivations, emotions and deeper understanding rather than numerical data.
Regression Analysis
A statistical approach that quantifies relationships between variables and identifies predictors of key outcomes.
Sample
A subset of the target population selected to represent the wider population.
Secondary Data
Previously collected information such as market reports or public datasets used to support research objectives.
Segmentation
The process of dividing a market into distinct groups with shared characteristics. Includes consumer and occasion-based segmentation.
Sentiment Analysis
An AI-based technique that detects emotional tone (positive, negative or neutral) in text responses or reviews.
Share of Voice (SOV)
A measure of a brand’s advertising presence in relation to competitors.
Share of Wallet (SOW)
The percentage of a customer’s total category spending that goes to a specific brand.
Significance Test (Sig Test)
A statistical method used to determine whether observed differences are real rather than due to sampling variation.
Standard Deviation
A measure of data variability indicating how spread out responses are around the mean.
System 1 & System 2 Thinking
Two cognitive systems: System 1 is fast and intuitive; System 2 is slower, deliberate and analytical.
TURF Analysis
A method identifying combinations of products, messages or features that maximise audience reach.
Universe
The full population being targeted or studied.
Verification
The process of checking data accuracy by identifying errors, speeders, straight-liners or implausible responses.
Weighting
Adjusting survey data to correct over- or under-representation to better reflect the target population.
Word Cloud
A visual representation of frequently used words in qualitative data, where word size reflects frequency.