The central distinction
Large organisations have invested heavily in data governance, catalogues, lineage, critical-data-element registers, quality dimensions, scorecards and stewardship. Those capabilities are valuable. The problem begins when the existence of those artefacts is treated as evidence that the underlying data journey is controlled.
A catalogue can tell an organisation what a field means, who owns it, where it is expected to originate and which systems are expected to consume it. It cannot, by itself, prove that every expected record moved today, that mapping logic preserved meaning, that rejected records were investigated, or that the downstream process used the intended controlled output.
The catalogue is a map. The operating control environment must prove the journey.
What is going wrong in large organisations
1. They govern nouns, while failures happen in verbs
Governance concentrates on datasets, domains, owners, systems and critical elements. Integrity fails during extraction, transfer, mapping, transformation, filtering, enrichment, aggregation, exclusion and consumption.
2. They confuse lineage with proof
Lineage shows an intended or discovered route. It does not establish that the expected population travelled intact, on time and under the approved transformation logic on a particular run.
3. They measure dimensions rather than control outcomes
Completeness, accuracy, validity and timeliness are useful lenses. A percentage score is not enough when the missing or distorted fraction may contain the records that matter most to a regulatory, customer or risk decision.
4. Controls are periodic rather than continuous
Monthly dashboards, quarterly attestations and point-in-time testing detect issues after downstream decisions have already been made. Material transfers and transformations need proportionate automated monitoring.
5. Accountability ends at organisational boundaries
Source, engineering, platform, control and business teams may each own a component. No one is clearly accountable for proving integrity across the complete chain and acting when evidence breaks.
6. Programmes produce artefacts rather than evidence
A catalogue, glossary, council, lineage diagram or dashboard can demonstrate programme activity. Operating evidence must show what was tested, which population was covered, what failed, who acted and whether remediation restored integrity.
What catalogues do well—and where the boundary lies
Catalogues are important for discoverability, shared meaning, metadata, ownership, policy visibility and lineage navigation. Those are prerequisites for control design because an organisation cannot control a journey it cannot describe.
But even a well-implemented catalogue normally relies on data-quality and control signals produced elsewhere. Actian's own guidance makes the distinction explicitly: a catalogue is not a full Data Quality Management tool and should not replace controls in source systems or transformation flows.
- A catalogue can expose an owner; it cannot prove that the owner responded to today's exception.
- A catalogue can display lineage; it cannot prove that the full population followed that route without loss.
- A catalogue can display a quality score; it cannot determine whether a small failure affected the most decision-critical cases.
- A catalogue can hold rules and metadata; it cannot prove that the correct rule version executed at the correct point.
DAMA is a foundation—not an executable end-to-end control system
DAMA-DMBOK® is a globally recognised body of knowledge spanning governance, data quality, metadata, architecture, integration and other core disciplines. The fair criticism is not that DAMA ignores these subjects. It is that a body of knowledge does not, by itself, create a prescriptive operating architecture that continuously proves a decision-critical journey.
An organisation can be strongly aligned to DAMA principles and still lack automated reconciliations at material hand-offs, correctness checks after transformation, evidence of control execution, exception ownership and end-to-end accountability for a defined outcome.
The missing layer: continuous control monitoring across the journey
Continuous control monitoring does not mean placing every conceivable rule everywhere. It means identifying the material data touches where completeness, correctness, timing, transformation or consumption can change the outcome—and making the detective control proportionate, repeatable and evidenced.
Define the expected population
Establish what should move, when, in what form and under which approved scope or exclusion.
Control each material transfer and transformation
Use counts, values, checksums, schema validation, mapping checks, reference integrity, drift detection and consumption evidence as appropriate.
Monitor control execution—not only data scores
Detect non-execution, stale thresholds, recurring exceptions, control degradation and unresolved ownership.
Retain evidence through remediation
Show the exception, impact, decision, corrective action, retest and sustainable closure.
Seven questions that test whether the journey is truly controlled
- Can the organisation define the complete expected population supporting the decision?
- Can it prove that population arrived at every material hand-off, intact and on time?
- Can it prove that mappings, transformations and reference values preserved intended meaning?
- Can it identify which control version executed, against which population, and when?
- Can it show all exceptions, authorised exclusions and unresolved gaps?
- Is one accountable owner able to connect evidence across organisational and provider boundaries?
- Can the organisation reconstruct how the final output or decision was produced and defended?
If several answers depend on assumptions, screenshots, manual explanations or disconnected artefacts, the environment may be well documented without being fully controlled.
Why this remains a live institutional problem
Framework adoption and tool investment have not removed the implementation gap. The Basel Committee's January 2026 update continued to identify governance, data lineage, cross-border complexity, emerging technology and compensating controls among the challenges surrounding BCBS 239 implementation.
The EDM Association's 2026 benchmark similarly describes data management as an established organisational discipline whose implementation remains uneven, with most organisations still operating in developmental or defined capability ranges across many components.
The issue is therefore not a shortage of terminology, frameworks or platforms. It is the conversion of those foundations into operating proof.
What stronger organisations add
They do not discard the catalogue, glossary, lineage platform or data-quality framework. They connect those assets to an end-to-end control architecture:
- Business outcomes and decision-critical journeys determine control scope.
- Catalogue metadata and lineage inform where controls must be placed.
- Automated monitoring proves completeness and correctness at material data touches.
- Exceptions are tied to accountable owners, impact decisions and time-bound remediation.
- Control execution and remediation evidence are retained so the outcome can be reconstructed.
The practical takeaway
A mature data catalogue can make an organisation much easier to understand. A mature data-integrity control environment makes its decision-critical outcomes easier to trust and defend.
The question is not whether the data is catalogued. The question is whether the organisation can prove the chain.
Official reference points
These sources support the distinctions and current implementation context discussed above. The DQIntegrity analysis and conclusions remain an independent interpretive position.