IIBA-CBDA key terms

A focused, answerable question that states exactly what the analyst needs to learn in order to address a business problem and direct the data work.

A concise description of the gap between the current and desired business situation, including who is affected and why it matters.

A statement of what the analysis is meant to accomplish, expressed with measurable outcomes so that success can later be judged.

A testable, provisional explanation or prediction about the data that the analyst will confirm or reject using evidence.

The boundaries of an analysis, stating which questions, data, time periods, and stakeholders are included and which are explicitly excluded.

Something accepted as true for planning purposes without supporting evidence, which must be validated before results are relied upon.

A structured method for tracing a visible symptom back to the underlying causes so that analysis targets the real problem rather than its effects.

Pre-agreed, measurable conditions that define whether the research answered its questions well enough to support a decision.

An examination of how processes, data, and performance work today, providing a baseline against which future changes are compared.

A document describing the questions, data sources, methods, timeline, roles, and deliverables for an analysis effort.

Data collected directly from original sources, such as surveys, interviews, or observations, specifically for the question at hand.

Data that already exists because someone else gathered it for another purpose, such as published reports, sales records, or government statistics.

Any system, person, document, or repository from which data is obtained, together with its owner and access conditions.

The characteristics used to judge whether data is fit for use, commonly completeness, accuracy, consistency, timeliness, validity, and uniqueness.

A technique of examining data structure, values, and patterns to reveal gaps, duplicates, invalid entries, and unexpected distributions before analysis begins.

A reference that defines each data element's meaning, format, allowed values, owner, and source so that terms are used consistently.

The documented path data travels from origin through transformations to reporting, used to trace errors and demonstrate trustworthiness.

The complete set of items, people, or transactions about which the analyst wants to draw conclusions.

Selecting a subset of a population to analyze when examining everything is impractical, together with the method used to choose that subset.

Data that can single out an individual, directly or in combination with other data, and therefore requires protection and lawful handling.

The authoritative source that holds the official, master version of a data element when several systems contain similar values.

Summary measures such as totals, averages, percentages, and spread that describe what is in a dataset without generalizing beyond it.

Methods that use sample data to estimate population values or test hypotheses, producing confidence intervals and probability-based conclusions.

A measured relationship in which two variables move together; on its own it does not show that one causes the other.

Comparing values over time to identify the direction, rate, and pattern of change, including seasonality, so future performance can be anticipated.

Dividing a population into meaningful groups sharing characteristics so that patterns and needs hidden in aggregate totals become visible.

A value far from the rest of the observations, which may signal a data error, a rare event, or an important exception worth investigating.

A measure of how unlikely an observed result would be if no real effect existed; it supports but does not prove a conclusion.

Correcting or removing inaccurate, duplicated, incomplete, or badly formatted records so that analysis rests on reliable input.

Representing data graphically, such as with charts or dashboards, so that patterns are seen faster than through tables alone.

Explaining what the analysis results mean in business terms, including what changed, by how much, and what that implies.

A factual, evidence-based statement produced by the analysis, stated without judgment or proposed action.

A finding connected to context and consequence that explains why something happened and what it means for the business.

Presenting results as a coherent narrative that links evidence, meaning, and recommended action for a specific audience.

The acknowledged weaknesses of an analysis, such as data gaps, assumptions, or sample bias, that bound how far its conclusions can be trusted.

A specific, justified proposal for action that follows from the analysis and states the expected benefit, the owner, and the conditions for success.

An insight specific enough and timely enough for a decision maker to act on, naming the affected area and the likely effect of acting.

Comparing the expected costs of an option with its expected benefits, often in monetary terms, to judge whether pursuing it is worthwhile.

A table that scores options against weighted criteria so that alternatives can be compared transparently and the chosen option justified.

A defined, repeatable measure used to track whether a chosen action delivers the intended business outcome over time.