Słownik pojęć

A/B testing compares two variants of a solution under controlled conditions to check which better achieves the goal. It allows decisions to be made on the basis of customer behavior…

Acquiescence bias is the tendency of respondents to agree with survey statements regardless of their actual views. It inflates scores and can lead to incorrect business decisions.

Ad post-testing is advertising research conducted after a campaign starts or ends. It assesses whether the message was noticed, remembered, attributed to the brand and understood as intended.

Ad pre-testing (copy testing) assesses the potential effectiveness of a message before it goes on air. It reduces media investment risk by checking clarity, credibility and engagement.

AI-moderated interviews are research conversations conducted by an AI system following a discussion guide. They scale qualitative data collection while keeping a conversational format.

Analysis of variance (ANOVA) assesses whether means in several groups differ more than random variation would suggest. It tests whether segments or message variants respond differently.

The ARIMA model is a classic time series method that forecasts a variable on the basis of its own history. It helps predict sales, demand and other indicators measured over…

Audience measurement is the systematic analysis of who consumes media, how and when. It supports decisions on media planning, advertising sales and content strategy across channels.

The Bass diffusion model describes and forecasts the rate of adoption of new products and technologies in a market. It separates demand into innovators' decisions and those driven by social…

Behavioral segmentation divides a market into groups according to actual behavior toward a brand, product or channel. It shows how consumers really buy, use and respond.

Brand awareness is the degree to which audiences recognize a brand and recall it within a given category. It is the basic measure of a brand's presence in memory and…

Brand equity is the value a brand adds beyond functional features and price. It means the strength of associations, trust and loyalty that translate into market advantage and business results.

The brand funnel is a model of the path from unawareness of a brand to its consideration, choice and loyalty. It allows brand strength to be measured at each stage…

Brand health tracking is the cyclical measurement of brand strength on a fixed set of indicators. It captures changes and trends and assesses whether marketing activity is improving the brand's…

Brand image is the way a brand is perceived by its audience. It is a measurable set of associations, judgments and emotions attributed to the brand that influence decisions and…

A brand lift study measures whether campaign exposure changes how people think, feel or intend to act toward a brand, separating campaign effects from changes that would have occurred anyway.

Brand loyalty is a customer's lasting propensity to choose a brand again despite available alternatives. It combines purchase behavior, attitudes and resistance to competitors' actions.

A business tendency survey is a recurring survey of firms measuring assessments of current conditions and near-term expectations. It signals changes in economic activity early.

A buying center is the group of people and functions influencing a purchase decision in a company. It helps establish who initiates the need, evaluates suppliers and actually decides on…

CAPI interviewing is a face-to-face data collection technique with answers recorded on a mobile device. It is a controlled quantitative measurement carried out in the field.

Case study research examines a bounded real-world phenomenon in depth and in context. It is valuable when market behaviour or decisions cannot be explained by isolated variables alone.

A central location test (CLT) is an evaluation of a product, package or message at a controlled venue. It combines conditions close to an experiment with a realistic decision context.

CHAID analysis is a decision-tree technique that identifies which respondent characteristics are most strongly associated with an outcome, turning survey data into clear segments.

The chi-square test assesses whether differences between observed and expected counts are due to chance. In market research it is used to check associations between categorical variables.

Churn analysis is a structured study of the scale, dynamics and causes of customer attrition. It helps understand not only the churn rate, but also the factors behind the loss…

Cluster analysis is a statistical method that detects homogeneous groups of objects based on similarity of characteristics. It supports segmentation by identifying natural patterns in customer data.

Cognitive interviewing is a qualitative method for testing how respondents understand and answer survey questions. It reveals flaws in wording and flow before a survey is fielded.

Cohort analysis tracks the behavior of groups defined by a shared characteristic, usually the time of first purchase. Instead of a population average it reveals differences between groups over time.

Computer-Assisted Telephone Interviewing (CATI) is a standardized technique of collecting data through an interviewer following a script. It combines process control, fast pace and good quantitative data quality.

Computer-Assisted Video Interviewing (CAVI) is a method in which an interviewer or moderator conducts interviews via live video, combining face-to-face depth with the reach of remote research.

Computer-Assisted Web Interviewing (CAWI) is an online quantitative technique in which the respondent completes the questionnaire alone. It collects standardized data quickly and scales studies well.

A concept test is research run before a product or communication is launched, assessing the potential of an idea. It shows how audiences understand the proposition and whether they see…

A confidence interval defines a range describing the uncertainty of an estimate of a population parameter. It separates a point result from sampling uncertainty and supports a better assessment of…

Confirmatory research checks whether earlier assumptions, hypotheses or models hold up in the data. It assesses which leads are accurate, relevant to the business and possible to generalize.

A confounding variable distorts the apparent relationship between a presumed cause and an outcome. Confounding in research can lead to wrong conclusions about what drives business results.

Conjoint analysis is a method of determining how buyers evaluate different combinations of product or offer features. It allows estimation of which attributes genuinely influence choice.

A consensus forecast aggregates estimates from multiple analysts, experts or models. It reduces the influence of extreme opinions and provides a reference point for business decisions.

A consideration set is the group of brands a customer genuinely weighs before making a choice. It matters more than awareness alone, because it shows a brand's chance of entering…

A consumer confidence index measures how households assess their finances, the wider economy and willingness to make major purchases. It can signal demand shifts before sales data.

A consumer insight is an accurate explanation of why people think, feel and act in a particular way. It links data with decisions by pointing to the mechanism behind customer…

Content analysis is a systematic method for examining transcripts, reviews, ads, social media posts and documents. It turns unstructured material into evidence for business decisions.

A control group is a reference point that allows assessment of whether change results from the action being studied. It separates the effect of a campaign, product or price from…

Correlation analysis assesses whether two variables change together and what the direction and strength of the relationship are. It shows whether, for example, customer satisfaction co-occurs with loyalty.

Correspondence analysis visualizes relationships between categories of qualitative variables on a two-dimensional map. It shows which brands, attributes and segments co-occur more often.

Coverage error occurs when the sampling frame does not match the target population. Some relevant people, households or companies have no chance of being sampled, which can bias results.

Cronbach's alpha assesses whether multiple survey items work together as a consistent scale. It helps determine whether a battery of statements provides a reliable measure for analysis.

A cross-sectional study analyzes a population or group of respondents at a single point in time. It quickly captures the current picture of customer attitudes and needs, without observing change…

Cross-tabulation shows the distribution of answers at the intersection of two or more variables. It makes it easy to spot differences between respondent segments, brands or channels.

Customer Effort Score (CES) shows how much effort a customer puts in to reach a goal, for example getting help or completing a purchase. It rates the ease of the…

Customer journey mapping organizes knowledge about how an audience moves through successive stages of contact with a brand. It shows touchpoints, user goals and the places where consistency is lost.

Customer Lifetime Value (CLV) measures the economic value of a customer across the entire relationship. It helps assess not only current sales, but also the future potential of segments and…

Customer Satisfaction Score (CSAT) is a metric for quickly measuring customer satisfaction with a specific experience or service stage. It shows whether the interaction met expectations.

Data anonymization and pseudonymization reduce the risk of identifying research participants while keeping data usable. The right choice depends on whether records must stay linkable.

Data mining is the process of discovering meaningful patterns, relationships and segments in large datasets. It turns customer, transactional and survey data into evidence for decisions.

Data saturation is the point in a qualitative study when further interviews add no substantially new information. It helps decide when to stop interviewing based on the quality of the…

Data triangulation combines different sources, methods and perspectives in order to verify conclusions more reliably. It separates signal from noise when a single method does not show the full market…

Data weighting is a statistical procedure that corrects the sample structure so that it better reflects the population. It limits the impact of over- or under-representation of respondent groups.

A decision-maker respondent is a study participant who genuinely influences supplier choice or solution rollout. In B2B this makes it possible to analyze decision mechanisms, not just declarations.

The Delphi method is a structured way of aligning expert opinions when complete historical data are not available. It relies on anonymous, multi-round consultations with an aggregated assessment.

Demographic questions describe respondents by personal, household, professional or organisational characteristics. They enable group comparisons, quota control and statistical weighting.

Dependent and independent variables are core concepts for testing relationships in research. They show whether factors such as price or advertising are linked to outcomes like purchase intent.

Descriptive statistics organise, summarise and present data so that patterns in a sample can be understood. They turn survey and transaction data into clear indicators of market structure.

Desk research is the analysis of data that already exists in credible sources. It allows knowledge about the category, competition and market environment to be organized quickly before primary research.

Discourse analysis examines how language shapes meaning and attitudes around a brand, product or category. It reveals not only what people say, but how they frame issues and use metaphors.

A discussion guide is a research interview outline that organises objectives, topics and questions. It keeps interviews comparable while leaving the moderator room to probe key issues.

A double-barreled question asks respondents to evaluate two or more issues in a single survey item. One answer may reflect different opinions, producing ambiguous and unreliable findings.

Dyadic research analyzes a relationship or decision process from the perspective of two connected parties. It is useful where one side's opinion does not explain the full dynamics of purchase…

An econometric model is a quantitative description of relationships between market phenomena, based on data and statistics. It explains what drives demand, sales and prices, and serves to build forecasts.

Effect size measures the magnitude of a difference, association or change in data. Unlike statistical significance alone, it shows whether a result matters for a business decision.

An empathy map organizes knowledge about the customer through the lens of their perspective, emotions and decision context. It shows what the customer says, thinks, feels and does, supporting offer…

An employee engagement survey measures how emotionally and cognitively connected employees are to their work, team and organisation, and what supports effort and retention.

Employee Net Promoter Score (eNPS) measures employees' willingness to recommend their employer as a place to work, helping organisations monitor employer advocacy and track changes over time.

An ex-ante forecast estimates future outcomes using only information available at the time. An ex-post evaluation compares it with actual results to measure forecast error and improve models.

The Experience Sampling Method (diary method) repeatedly records a respondent's experiences and emotions close to the moment they occur. It captures the gap between declarations and real behavior.

Expert interviews are a qualitative method of obtaining knowledge from people who understand how a market works. They help to grasp the business context and to verify research hypotheses.

Exploratory research serves to identify a problem or a market when there is too little knowledge to state hypotheses. It aims at discovering and refining questions rather than confirming relationships.

Exponential smoothing is a time series forecasting method that gives greater weight to the most recent observations. It captures the direction of change quickly without reacting to short-term noise.

Eye tracking is a method of measuring visual attention that shows what a viewer looks at, in what order and for how long. It moves the analysis from declarations to…

Factor analysis uncovers the hidden dimensions behind many observed variables. It helps to understand which deeper attitudes and choice criteria organize respondents' answers.

Fieldwork is the execution stage of a study: contact with respondents and data collection. It covers recruitment, measurement and oversight of compliance with the design, determining the quality of the…

A Focus Group Interview (FGI) is a qualitative method based on a moderated discussion in a small group. It shows how people form judgments and give meaning to brands and…

Forecast error (MAPE) shows by what percentage a forecast differs on average from the actual value. It allows the quality of models to be compared and forecast accuracy to be…

Grounded theory builds models through systematic analysis of data rather than by testing hypotheses. It works well for poorly described phenomena and newly emerging decision mechanisms.

The halo effect is a cognitive bias in which one salient feature of a brand or product influences the assessment of others. It can inflate or deflate ratings and distort…

Hard-to-reach respondents are individuals or companies that are difficult to access in a study. They involve higher cost, longer timelines and recruitment risk that affects data quality.

The Hawthorne effect is a change in participants' behavior resulting from the awareness of being observed. It distorts quantitative measurements and qualitative observation, inflating what respondents declare.

HCP research examines the knowledge, attitudes, clinical decisions and unmet needs of healthcare professionals, providing evidence for pharma, medtech and healthcare organisations.

A household panel is a permanent group of households that regularly report purchases and consumption. It shows changes within the same households over time and real buying patterns.

An Ideal Customer Profile (ICP) is an operational description of the segment with the highest value and the best fit to the offer. It structures targeting, lead prioritization and offer…

An In-Depth Interview (IDI) is a qualitative method used to understand a respondent's motivations, attitudes and language. It reaches meanings that closed-ended answers cannot capture.

An in-home use test (IHUT) checks how consumers use a product within their everyday routine. It reveals the role of convenience, repeat use and comparison with brands used so far.

Incidence rate is the percentage of screened people who meet the eligibility criteria for a study. It affects recruitment feasibility, fieldwork timelines, sample sources and costs.

Internal and external validity describe the quality of research conclusions: whether an effect can be attributed to its cause, and whether the finding applies beyond the study setting.

The interviewer effect is a distortion of answers caused by the presence and behavior of the person conducting the interview. It is a major source of error, shifting response distributions…

The Interviewer Quality Control Programme (PKJPA) is a Polish industry standard for verifying interviewer fieldwork, helping to ensure interviews are conducted correctly and data are reliable.

Jobs to Be Done (JTBD) analyzes customer choices through the lens of the job they want to get done in a given situation. It focuses on the customer's goal and…

The Kano model identifies how product or service features influence customer satisfaction, separating basic expectations, performance drivers and unexpected benefits to set priorities.

Key driver analysis identifies which elements of an offer have the strongest impact on an outcome such as satisfaction or loyalty. It distinguishes highly rated attributes from those that actually…

The laddering technique is a qualitative interviewing method that moves from product attributes to consequences and personal values, revealing the deeper drivers of customer choice.

Latent class analysis (LCA) identifies unobserved groups of respondents with similar patterns in answers, behaviours or preferences, supporting data-driven market segmentation.

A leading question is phrased in a way that suggests a particular answer or judgement. In market research, it distorts results by measuring the wording's influence, not just opinions.

Levels of measurement define how respondents' opinions and behavior are converted into data for analysis. The choice determines which statistical tests are permissible and how differences are interpreted.

The Likert scale is a popular tool for measuring opinions and attitudes in surveys. It turns respondents' subjective declarations into ordered data, comparable across segments and periods.

Logistic regression predicts the probability of a binary event: buy or not buy, churn or stay. It indicates which characteristics and behaviors increase the chance of it occurring.

A longitudinal study analyzes change over time through repeated measurement of the same units. It helps distinguish real change in behavior and attitudes from one-off fluctuations in measurement.

Market concentration describes how large a share of sales is held by a small number of companies. It helps assess whether a category is fragmented or dominated; it is often…

A market forecast is a structured assessment of future changes in demand, sales or prices, based on data and models. It estimates the most likely scenarios and the factors behind…

Market monitoring is the repeatable tracking of changes in demand, supply, prices and competition. It is a continuous process of observing data that allows faster response and a grasp of…

Market penetration determines what share of potential buyers actually bought a brand or product in a given period. It is a basic measure of offer reach and a starting point…

A market research online community (MROC) is a moderated group of respondents on a dedicated digital platform. Participants take part in a series of tasks and discussions spread over time.

Market scenarios are structured descriptions of the possible ways the market environment may develop. They do not predict one future but help assess how to respond to different sets of…

Market share shows what portion of category sales is generated by a brand, company or product. It is the basic tool for assessing competitive position against the size of the…

A marketing experiment checks whether a given action really influences audience behavior. It relies on a controlled change of one factor and observation of its effects.

Marketing mix modeling (MMM) estimates how marketing activities and external factors contribute to sales or other outcomes over time, helping to improve budget allocation across channels.

MaxDiff analysis is a preference measurement method that establishes the relative importance of features based on "most" and "least" preferred choices. It produces a reliable ranking of priorities.

Measurement reliability is the degree to which an instrument produces stable results free of random error. It is a condition for trusting the data; without it, decisions based on results…

Measurement validity determines whether a research instrument really measures the intended phenomenon rather than a superficially similar one. It is a basic condition for credible interpretation of data.

A Minimum Viable Product (MVP) is the simplest version of a solution that verifies whether the proposed value is genuinely needed by the market. It allows key assumptions to be…

Missing data and imputation cover identifying, assessing and treating absent values in a dataset. How gaps are handled can materially affect survey findings and business decisions.

Mixed methods research combines qualitative and quantitative techniques within a single project, designed so that they complement one another and explain the market phenomenon under study more fully.

A mixed-mode survey combines two or more data collection modes, such as online, telephone and face-to-face interviews. It expands reach but requires control of data comparability.

Monadic and sequential monadic testing are designs for evaluating concepts, products or ads. A monadic test shows one stimulus per respondent; a sequential test shows several in order.

Multidimensional scaling (MDS) represents perceived similarities and differences between brands, products or services on a visual map, revealing competitive structures hidden in ratings.

Narrative analysis treats people's stories as a source of data on how they understand experiences and decisions. It shows not only what a respondent declares, but how they order events…

Needs-based segmentation divides a market by customer expectations, problems and motivations rather than demographic characteristics. It allows more accurate offers and customer experience to be designed.

Net Promoter Score (NPS) is a loyalty metric showing customers' willingness to recommend a brand. Its value grows when the score is placed in the context of segments and the…

Netnography is the qualitative analysis of consumer behavior, opinions and language in online environments. It shows how people spontaneously talk about brands in a natural digital context.

Neuromarketing applies knowledge from neuroscience and cognitive psychology to analyze consumer reactions to stimuli. It reveals not only declarations but also responses at the level of emotions.

Non-response bias is a distortion of results that occurs when those who do not take part in a study differ from respondents. Even a large sample can mislead if the…

Nowcasting estimates what is happening in the economy or market now, before complete official data are available, combining recent, high-frequency signals to close the information gap.

An omnibus survey is a quantitative study in which several clients share one measurement and buy their own block of questions. It gives fast access to data at a lower…

Open-ended and closed-ended questions are the two basic question types in research instruments. The choice between them affects data quality and whether a study measures scale or explains causes.

Operationalization in research translates concepts such as satisfaction or loyalty into measurable indicators. It determines whether a study really measures what it should and whether results are credible.

Oversampling, or a sample boost, deliberately recruits more respondents from a subgroup than its share of the population. It strengthens subgroup findings but requires proper weighting.

A packaging test assesses whether packaging is noticeable, understandable and consistent with buyers' expectations. It verifies design decisions before launch or a change of visual identity.

Paper-and-Pencil Interviewing (PAPI) is a method in which an interviewer records answers on a printed questionnaire. It remains relevant where technology access or field conditions are limited.

Participant observation is a qualitative method in which the researcher learns about people's behavior by taking part in their environment. It captures what respondents do, not only what they declare.

A perceptual map shows how audiences perceive brands and products in relation to selected attributes. It translates respondents' opinions into a clear picture of the market, supporting brand positioning.

A pilot study is a controlled check of the design before the actual measurement. It detects errors in instruments, recruitment and questionnaire logic before the study runs on the full…

Price elasticity of demand describes how strongly demand changes in response to a change in price. It is the basic tool for assessing customer price sensitivity in pricing and promotional…

Primacy and recency effects describe how an answer's position in a response list influences its selection. They are a material source of survey bias that requires response-order control.

Primary data are collected directly for a specific research problem rather than taken from existing datasets. They provide information from the people and situations relevant to a decision.

Principal component analysis (PCA) reduces a large set of correlated variables to fewer uncorrelated components, revealing the main patterns in data while retaining most of the variation.

Product-market fit is the state in which an offer answers a real and widespread need within a segment. It means demand has been confirmed and buyers are ready to purchase.

Projective techniques reveal hidden associations, emotions and barriers that respondents do not express directly. They show how people interpret brands, messages and purchase situations.

Purchase intent is a customer's declared readiness to buy a product or brand within a given time horizon. It is an indicator of potential demand, not a sales forecast.

Purposive sampling is a non-probability method in which participants are deliberately selected because they meet criteria relevant to the research objective, not for representativeness.

Qualitative coding turns statements and textual material into analytical categories that can be compared. It makes it possible to move from raw data to usable conclusions.

Qualitative research explains respondents' motivations, meanings and decision barriers. It answers the "why" and "how", supporting problem exploration and offer development.

Quantitative research measures attitudes and behaviors in a structured and comparable way. It delivers numerical data for assessing the scale of a phenomenon, segmentation and data-driven decisions.

A quasi-experiment assesses the effects of a campaign or business change when random assignment to test and control groups is not possible. It supports causal inference, used with caution.

Quota sampling is non-random recruitment in which the structure of the sample is controlled by predefined quotas for selected characteristics. It reproduces the profile of a population when random selection…

Randomization is the controlled introduction of randomness into the research process, limiting systematic errors. It increases the credibility of measurement and neutralizes order effects.

Recall bias occurs when respondents do not accurately remember past events, behaviours or exposures. It can distort estimates of purchasing, media consumption and advertising impact.

Regression analysis estimates the relationship between a dependent variable and explanatory variables. It shows which factors are most strongly linked to customer decisions and supports forecasting.

A representative sample is a subset of a population that allows reliable inference about the whole with respect to the characteristics studied. It determines whether results can be read as…

A research brief organizes the objectives, questions, scope and conditions of a study before methods and a supplier are chosen. It reduces the risk of misframing the problem and speeds…

A research design is the logical and operational plan of a project: what data to collect, from whom and in what order. It determines whether results will be useful for…

A research executive summary is the condensed part of a report that translates results into the key decision-making conclusions. It allows quick understanding of business implications and priorities.

A research hypothesis is a testable assumption about a relationship or mechanism between market phenomena. It structures the project, indicates what data to collect, and separates observation from guesswork.

A research panel is a managed database of respondents who have agreed to take part in studies. It allows fast reach to precisely defined groups and repeatable quantitative and qualitative…

A research persona is a data-based description of a representative type of customer or decision-maker. It is not an invented character, but a tool for organizing knowledge about segments and…

A research question is the precisely formulated problem a project sets out to answer. It defines the scope of analysis, the choice of methods, operationalization and the logic of interpreting…

A research questionnaire is an ordered set of questions that obtains data from respondents in line with the study objective. It determines measurement quality, comparability of answers and credibility.

A research recommendation is a practical indication of action arising from the data and the analysis of the material. It links market research results with a business, product or organizational…

A research screener is a selection tool that decides who can take part in a study and who is excluded. Its quality affects sample accuracy, the credibility of conclusions and…

A respondent incentive is a benefit offered in return for a participant's time and knowledge. Well-designed survey incentives increase participation but must be proportionate and ethical.

Respondent recruitment is the process of identifying, selecting and inviting people who meet the study criteria. Its quality determines whether the data really answers the research questions.

Response rate indicates what share of invited respondents took part in a study. It is the basic measure of fieldwork quality and of the risk of non-response bias.

A retail audit is the systematic observation and measurement of what happens at points of sale. It verifies availability, display, prices and compliance with trade standards.

Retail scanner data are sales records captured at the point of sale. They show what was sold, where, when, at what price and under which promotion, supporting market measurement.

RFM analysis segments customers by recency, frequency and monetary value of purchases. It helps distinguish high-value, active customers from those whose relationship with a brand is weakening.

Sample size is the number of units covered by a study, on the basis of which inferences about the population are made. It determines estimation precision, segmentation analysis and the…

Sampling error is the difference between the result obtained on a sample and the result for the whole population, arising from studying only part of it. It affects the accuracy…

A sampling frame is an ordered list of units from which a sample is drawn. Its quality determines the representativeness of results, the size of the sampling error and the…

Sampling in research is the methodological decision about whom data will be collected from and how accurately it describes the population. Correct sampling limits the risk of wrong business conclusions.

Secondary data analysis is the use of existing information to answer research questions without measuring from scratch. It helps to understand the market and to design one's own studies better.

Selection bias is a systematic distortion that occurs when the people included in a study differ meaningfully from the population it is meant to describe, leading to misleading conclusions.

The semantic differential is a technique for measuring associations and attitudes by rating an object on a scale between opposing terms. It captures how respondents perceive a brand.

A semi-structured interview combines the order of a script with the flexibility of an in-depth conversation. It allows answers to be compared while letting the moderator explore important threads.

Sensory analysis measures how people perceive a product through sight, smell, taste, touch and sound. It turns subjective product experiences into evidence for development and positioning.

Sentiment analysis automatically recognizes the emotional tone of textual statements. It allows large sets of reviews, comments and social media content to be organized quickly.

Shopper ethnography is a qualitative method based on observing consumer behavior in the real purchase environment. It shows not declarations, but actual actions and choice mechanisms.

Shopper research focuses on how consumers make decisions at the real or digital point of sale. It explains how the choice unfolds and which stimuli change shopper behavior.

A shopping mission is the specific goal a consumer enters a store or an app with. It explains why a purchase is to happen now and in this particular channel;…

Snowball sampling works by having each respondent point to further people who meet the criteria. It works well with hard-to-reach populations where no reliable sampling frame exists.

Social desirability bias is a distortion of answers that occurs when a respondent wants to look better. It lowers the validity of declarations, especially around norms, status, health and finances.

Social listening is the systematic analysis of public online statements about a brand, category and competition. It helps to understand the context, tone and topics that shape purchase decisions.

Statistical significance assesses whether an observed difference or relationship results from more than random sample variation. It helps distinguish an analytical signal from chance fluctuations.

Stratified sampling divides a population into analytically relevant strata and draws a sample within each of them. It increases control over the structure of the study and the quality of…

Structural equation modeling (SEM) tests how observed variables and latent concepts such as satisfaction or loyalty are related within one model, using quantitative research data.

Student's t-test assesses whether a difference between means is statistically significant. It is used to compare two groups, two measurements or a group against a reference value.

Syndicated data are market measurements collected with a single, repeatable methodology and made available to many clients. They enable comparisons over time and across categories without a dedicated study.

Synthetic respondents are AI-generated profiles that simulate the answers of specific customer groups. They support the exploration of hypotheses but do not replace real respondents.

Text mining is a set of methods for automatically extracting meanings and patterns from textual data. It turns unstructured customer statements into organized analytical conclusions.

Thematic analysis is a method for organizing and interpreting qualitative data. It translates statements, transcripts and comments into clear patterns of meaning that support decisions.

A time series is a sequence of observations of a single phenomenon ordered in time. It allows change, trend and seasonality in market processes to be studied and signals useful…

Time series decomposition separates change over time into trend, seasonality, cyclicality and a random component. It helps assess whether a change is a lasting tendency or a one-off disturbance.

Total survey error (TSE) is a framework for identifying and reducing the combined errors that affect survey findings, from coverage and sampling to measurement and data processing.

A touchpoint is any situation in which a customer comes into contact with a brand, product or communication. It can occur before, during and after purchase, shaping how the brand…

A tracking study is a repeatable measurement at fixed intervals, following brand, communication and customer experience indicators. It shows trend and dynamics rather than a single result.

Trade research is market research among distributors, wholesalers, retailers, dealers and other channel partners. It shows how a sales channel works and how partners assess suppliers.

TURF analysis selects a set of products or messages that reaches the broadest audience without counting the same people more than once. It helps maximize the reach of a portfolio.

Type I and Type II errors are two mistakes in hypothesis testing: detecting an effect that does not exist or missing one that does. They shape how market research results…

An unmet need is an important customer need that current offers do not satisfy sufficiently. It serves to identify space for innovation and gaps in the market.

Usability testing checks whether users can use a product, site or prototype effectively and without unnecessary effort. It shows where the design supports the goal and where it gets in…

Usage & Attitudes research describes the consumer's relationship with a category, brand or product. It shows who buys and uses, as well as how, why and how often a purchase…

A vector autoregression (VAR) model analyses and forecasts several time series that may influence one another, treating each variable as both an outcome and a potential driver of the others.

Voice of Customer (VoC) is a systematic process of collecting and interpreting customer opinions and needs. It is a standing mechanism for translating their perspective into product and sales decisions.

Web scraping is the automated collection of data from public websites for analysis. It allows prices, assortment and reviews to be observed faster and at a greater scale than manually.

Willingness to Pay (WTP) is the maximum price a buyer accepts for a product or a set of benefits. It serves to assess the pricing potential of an offer and…