Short answer
Data analysis turns a defined question into a defensible conclusion by selecting relevant data, checking limitations, and connecting findings to the decision at hand.
About Data analysis
Data analysis is the systematic examination of data to answer questions and support defensible conclusions. It links evidence to a stated question without overstating what the data can show.
Use this competency for
- Roles that investigate questions using existing datasets and explain what the evidence supports.
- Work that requires comparing results, finding patterns, or diagnosing changes before a decision.
Do not use this competency for
- Use statistical reasoning when the main evidence is the choice, validity, or interpretation of statistical methods.
Important distinctions
Statistical reasoning
Statistical reasoning evaluates methods, assumptions, and uncertainty, while data analysis organizes evidence to answer a specific question.
Data visualization
Data visualization focuses on visual representation, while data analysis focuses on the reasoning that produces a conclusion.
Expectations by level
IC1
IC1: Scoped analysis
Answers well-defined questions with guidance, using a known dataset and method. Checks basic validity and states what the result does and does not support.
Observable behaviors
- Filters and aggregates data according to the documented question.
- Checks row counts, missing values, and unexpected ranges before interpreting results.
- Records the query, assumptions, and conclusion in a reviewable artifact.
Examples
- For a weekly drop in sign-ups, compares the affected period with the agreed baseline and flags incomplete event data.
- Given a defined retention question, reproduces the approved cohort calculation and explains the observed difference.
IC2
IC2: Independent analysis
Independently frames and completes ambiguous team-level analyses. Chooses relevant comparisons, tests alternative explanations, and connects findings to a practical decision.
Observable behaviors
- Translates a business question into measurable analysis steps and acceptance checks.
- Compares plausible explanations before presenting a conclusion.
- Explains limitations and recommends what evidence would reduce remaining uncertainty.
Examples
- When conversion changes, separates traffic mix from within-channel performance before recommending action.
- For a support backlog question, combines volume and resolution-time data and identifies where the queue accumulates.
IC3
IC3: Analysis direction
Leads complex analyses that affect multiple teams, defines a defensible approach, and reviews how others connect evidence to decisions.
Observable behaviors
- Defines analysis plans for questions with competing metrics or incomplete data.
- Reconciles conflicting findings across datasets and documents the resolution.
- Reviews team analyses for unsupported claims, missing comparisons, and decision relevance.
Examples
- Before a pricing decision, aligns product and finance on measures, compares segments, and documents sensitivity to exclusions.
- When two dashboards disagree, traces definitions and joins, resolves the discrepancy, and publishes the decision-ready result.