Data maturity assessment
Assess the decisions your data can support reliably today.
Maturity is not the number of tools, models or dashboards in place. It is the repeatable ability to produce evidence, govern its meaning and use it in a consequential decision.
The actual problem
Capability exists, but nobody can state where it is dependable.
Generic maturity models reward the presence of platforms, roles and processes. That can make a technically sophisticated organisation appear mature even when leadership still reconciles metrics manually or cannot connect analysis to action.
A useful assessment evaluates a bounded set of decisions end to end. It asks whether the required evidence is available, understood, timely, owned and embedded in a review cycle.
- Strong infrastructure, weak trustModern tooling exists, but teams still validate key numbers outside the governed environment.
- Uneven capabilityOne domain has reliable metrics and ownership while another depends on manual intervention.
- Activity as progressRoadmaps count migrated sources and delivered reports without measuring decision use.
- Unclear next investmentEvery capability gap appears important because business consequence is not part of the assessment.
- Assessment without evidenceScores reflect stakeholder perception but are not tested against records, logic or observed routines.
- No target stateTeams pursue the highest maturity level even where a simpler, controlled capability is sufficient.
Method
Assess maturity through representative decision pathways.
The assessment uses observable evidence rather than self-reported confidence alone. Scores are diagnostic labels, not precise measurements, and should remain visible by capability rather than being hidden inside one enterprise average.
- Decision selection
- Choose a small set of recurring decisions with material value, risk or coordination cost across different domains.
- Evidence trace
- Follow each decision from source records and definitions through analysis, review, action and outcome feedback.
- Capability rating
- Evaluate decision linkage, metric governance, quality control, delivery reliability, ownership and adoption against explicit evidence anchors.
- Variation analysis
- Compare domains and decisions to distinguish local practice gaps from shared platform or governance constraints.
- Risk weighting
- Prioritise weaknesses by decision consequence, frequency, detectability and the cost of operating around them.
- Proportionate target
- Define the minimum capability required for each decision instead of assuming every domain needs the same level of control.
Evidence required
What turns a maturity opinion into a defensible assessment.
Interviews reveal perceived constraints. Artefacts and observation establish what happens in practice. Both are needed because formal process and actual operating behaviour frequently differ.
| Evidence | Capability tested | Question |
|---|---|---|
| Decision forums and recent examples | Decision linkage and adoption | What information changed the choice or timing? |
| Metric definitions and production logic | Semantic governance | Can the headline measure be reproduced and interpreted consistently? |
| Quality incidents and workarounds | Control and resilience | How is a material defect detected, escalated and corrected? |
| Delivery history and backlog | Prioritisation and throughput | Are requests connected to decisions and delivered when useful? |
| Roles, ownership and escalation | Operating model | Who can resolve meaning, access, quality and priority? |
| Outcome review records | Learning loop | Does the organisation compare expected and actual effects? |
Outputs
A maturity profile tied to action and consequence.
The output should explain not only where capability is weak, but why the gap matters, where it appears and what sufficient improvement looks like.
- Decision pathway assessmentsEvidence, capability ratings and constraints for each selected decision.
- Capability heatmapVariation across domains without obscuring weaknesses inside a blended score.
- Evidence registerArtefacts supporting each finding, with uncertainty and unverified claims identified.
- Risk-ranked gapsWeaknesses ordered by business consequence, recurrence and current workaround cost.
- Proportionate target statesThe control and capability level required for each decision class.
- Improvement sequenceNear-term operating changes, enabling work and measures of actual capability improvement.
Worked example
An average score can hide the constraint that matters.
A retention decision has reliable analytical delivery but weak definitions, controls and outcome review. The proposed weighting reflects the consequence of each capability for this specific decision.
| Capability | Rating | Decision weight | Observed evidence |
|---|---|---|---|
| Decision linkage | 2 / 4 | 25% | Decision named, but no reversal criteria. |
| Metric governance | 1 / 4 | 25% | Two retention definitions remain in use. |
| Quality control | 1 / 4 | 20% | Late events are corrected manually. |
| Analytical delivery | 3 / 4 | 15% | Cohort analysis is reproducible and timely. |
| Ownership and feedback | 1 / 4 | 15% | No owner reviews expected versus actual outcome. |
The arithmetic is transparent, but the composite should not be treated as a scientific measurement. Its purpose is to show why additional analytical capacity is not the first priority: the stronger delivery capability is constrained by competing definitions and an absent feedback owner.
The first improvement should stabilise the retention contract and review responsibility, then test whether the existing analytical capability produces a more consistent decision.
Limits
What maturity ratings should not be used to claim.
Maturity is contextual and evidence-bound. It cannot be reduced to a universal benchmark without losing the decisions, risks and operating constraints that give it meaning.
- Not a certificationThe assessment covers selected pathways and evidence, not every system, team or control.
- Not an industry league tableDifferent organisations legitimately require different capability levels for different decisions.
- Not precise interval measurementOrdinal ratings support comparison and discussion; a score of four is not twice a score of two.
- Not a substitute for implementationThe assessment identifies constraints and sequence; accountable teams must deliver and sustain the change.