Frameworks Level: Practitioner
Knowledge areas: what to do in each
The DAM Policy asks for three capabilities: data governance, data cataloguing and data quality. Each rests on work in other data management areas. A catalogue record needs to know where the data lives, who can see it and which system is the authoritative source. A quality plan needs the reference data and the flows that feed it. This guide gives each area the activities that matter for the policy, in the order most organisations need them, down to the steps, outputs and evidence.
In brief
- 39 activities across 10 data management areas, each with steps, outputs, evidence, a measure and who does what.
- Three stages: what to do before your first return, before your second, and as business as usual.
- Every activity says which DAM Policy requirements it supports and which standards apply.
Draft. Draft by this portal for review by the profession that owns each area. Not government guidance. Activities are written fresh by this portal; DAMA-DMBOK is DAMA International's copyrighted work, and only its area names are used here.
Where to start: three stages
Pick the activities for the stage you are at. You do not need every activity for your first return.
- Start: before your first returnWhat the DAM Policy needs now, so the return is complete and true.
- Build: before your second returnMakes the next return easier, and makes critical assets safer to reuse.
- Embed: business as usualKeeps the work going without a project, and prepares critical assets for AI and wider sharing.
All activities at a glance
The DAMA wheel in the data portals compass drills down to every activity here. Governance, cataloguing and quality activities are the three capability playbooks.
Sources
- DAMA-DMBOK knowledge areas (names only, linked, not copied)V: Verified against the primary source
- Sovereign Standards Platform: introduction, user guide and roadmap (cross-government internal draft, version 1.1, November 2026). Used for the business glossary anatomy, the knowledge organisation ladder, controlled vocabularies, the draft data standards classification, crypto-dependency and residency fields. The pack is not published; the standards classification and the standards crosswalk it describes are drafts in review with the Data Standards Authority.O: This playbook's own synthesis of published guidance
Knowledge area · 5 activities
Data architecture
The blueprint of data across the organisation: where data lives, how it flows and which source is authoritative.
Why it matters for the DAM Policy: The policy asks you to designate the source data set behind any distribution you designate (DAM-R12). You can only do that if you know where each critical asset lives, which copy is authoritative and how it flows. Architecture gives the catalogue its structure.O: This playbook's own synthesis of published guidance
Usually led by: Design authority or solution architect. Content owner: Your design authority or architecture community.
Read with the Data architecture and models supplement.
ARC-1 · Start: before your first return
Map where each critical asset lives
Goal: One list that says which systems hold each critical asset, so nothing is designated from the wrong copy.O: This playbook's own synthesis of published guidance
Steps
- Start from the candidate and designation log. For each candidate, ask the steward and custodian which systems hold the data.
- Record each system, the copy it holds (master, replica, extract, archive) and who runs it.
- Mark copies that are spreadsheets, shared drives or personal stores. These are risks, not sources.
- Add the landscape to the catalogue record as related resources, not as separate critical assets.
- You produce
- A data landscape list for critical assets
- Updated candidate and designation log
- Evidence it is done
- Every designated asset has at least one named system and a named custodian.
- Measure
- Share of critical assets with a named system and custodian.
- Who does what
- Accountable: Design authority or solution architect
- Responsible (does the work): Data steward, Catalogue manager
- Consulted: Data custodian
- Informed: Data owner
- DAM Policy requirements it supports
- DAM-R01, DAM-R12
- Capabilities it builds
- Identify and scope critical data assets, Data architecture and data services
- Templates
- Candidate and designation log
- Standards and guidance
- The Technology Code of PracticeV: Verified against the primary source on 2026-11-05
- Develop your data and APIs using a reference architectureV: Verified against the primary source on 2026-11-05
ARC-2 · Start: before your first return
Name the authoritative source for each critical asset
Goal: Everyone who reuses the asset knows which copy to trust and where to report problems.O: This playbook's own synthesis of published guidance
Steps
- For each critical asset, decide which system is the authoritative source. If two systems disagree, the data owner decides.
- Record the source system, the entities it is authoritative for and how other copies are refreshed from it.
- Where you designate a distribution, link it to its source data set in the catalogue (DAM-R12).
- Record the decision in the decision log, with the date and who made it.
- You produce
- Authoritative source named in each catalogue record
- Decision log entries
- Evidence it is done
- Designated distributions link to their source data set; decisions are dated and owned.
- Measure
- Share of critical assets with a named authoritative source.
- Who does what
- Accountable: Data owner
- Responsible (does the work): Design authority or solution architect, Catalogue manager
- Consulted: Data steward, Data custodian
- DAM Policy requirements it supports
- DAM-R12, DAM-R02
- Capabilities it builds
- Data architecture and data services, Reference data, master data and authoritative sources
- Templates
- Critical data asset catalogue record, Entity and relationship list, Decision log entry
- Standards and guidance
- Develop your data and APIs using a reference architectureV: Verified against the primary source on 2026-11-05
- Data Catalog Vocabulary (DCAT) version 3V: Verified against the primary source on 2026-11-05
ARC-3 · Build: before your second return
Draw the main data flows between systems and organisations
Goal: You can say where critical data goes, so sharing, quality and security risks have a place on the map.O: This playbook's own synthesis of published guidance
Steps
- For each critical asset, list the flows in and out: source, target, method (API, file, manual), frequency and owner.
- Mark flows that cross organisational boundaries and link each to its data sharing agreement.
- Mark manual re-keying and spreadsheet hand-offs. These are where quality usually drops.
- Share the flow list with the data quality lead so quality rules sit where data enters and leaves.
- You produce
- A data flow list for each critical asset
- Links to data sharing agreements
- Evidence it is done
- Every external flow of a critical asset has a sharing agreement or a recorded exception.
- Measure
- Share of external flows with an agreement.
- Who does what
- Accountable: Data owner
- Responsible (does the work): Design authority or solution architect, Data steward
- Consulted: Data custodian, Data quality lead or analyst, Data protection officer
- DAM Policy requirements it supports
- DAM-R11, DAM-R15
- Capabilities it builds
- Data architecture and data services, Cross-boundary collaboration, Data protection and lawful sharing
- Templates
- Data sharing agreement
- Standards and guidance
- API technical and data standardsV: Verified against the primary source on 2026-11-05
- ICO data sharing code of practiceS: Found by web search; the site blocked automated checking, so open it before relying on it
- Digital Economy Act 2017V: Verified against the primary source on 2026-11-05
ARC-4 · Build: before your second return
Set a target data architecture and record decisions
Goal: Changes move critical assets toward open, portable designs instead of new legacy.O: This playbook's own synthesis of published guidance
Steps
- Write down the principles for critical data, for example one authoritative source, open formats and documented interfaces.
- Record each significant choice as an architecture decision record with the options, the choice and the reason.
- Check new systems and contracts against the principles before approval, including how data leaves at exit.
- Review the target each year with the governance board.
- You produce
- Data architecture principles
- Architecture decision records
- Evidence it is done
- New systems holding critical data have a decision record and an exit route for the data.
- Measure
- Share of new systems holding critical data with a recorded exit route.
- Who does what
- Accountable: Data governance board
- Responsible (does the work): Design authority or solution architect
- Consulted: Commercial lead, Service or product owner
- Informed: Chief data officer
- DAM Policy requirements it supports
- DAM-R16, DAM-R04
- Capabilities it builds
- Data architecture and data services, Avoiding future legacy: open standards, portability and exit
- Standards and guidance
- The Technology Code of PracticeV: Verified against the primary source on 2026-11-05
- Open standards for governmentV: Verified against the primary source on 2026-11-05
- Managing technical lock-in in the cloudV: Verified against the primary source on 2026-11-05
- Choosing technology: an introductionV: Verified against the primary source on 2026-11-05
ARC-5 · Embed: business as usual
Make architecture review a gate in change
Goal: Every change that touches a critical asset updates its landscape, source and flows before it goes live.O: This playbook's own synthesis of published guidance
Steps
- Add a question to your change and assurance gates, such as "does this change a critical data asset?".
- If yes, require an updated landscape, flow list and catalogue record before go-live.
- Feed spend control and service assessment evidence from the same records rather than a new pack.
- Report gate results to the governance board each quarter.
- You produce
- A gate question and checklist
- Quarterly gate report
- Evidence it is done
- Changes to critical assets show updated records at go-live.
- Measure
- Share of changes to critical assets with records updated at go-live.
- Who does what
- Accountable: Data governance board
- Responsible (does the work): Programme office and assurance, Design authority or solution architect, Programme or project manager
- Consulted: Catalogue manager
- DAM Policy requirements it supports
- DAM-R04
- Capabilities it builds
- Integration with programmes, spend control and assurance, Data architecture and data services
- Standards and guidance
- Service assessments and applying the Service StandardV: Verified against the primary source on 2026-11-05
- Set up a commercial or digital and technology spend controls pipelineV: Verified against the primary source on 2026-11-05
- Government Functional Standard GovS 005: DigitalV: Verified against the primary source on 2026-11-05
Knowledge area · 4 activities
Data modelling and design
Defining the structure, meaning and relationships of data.
Why it matters for the DAM Policy: A catalogue record describes a critical asset; a data model describes what is inside it. Stewards need a model to agree definitions, quality leads need it to write rules, and users need it to judge whether the asset fits their purpose.O: This playbook's own synthesis of published guidance
Usually led by: Design authority or solution architect. Content owner: Your design authority, with data stewards.
Read with the Data architecture and models supplement.
MOD-1 · Start: before your first return
List the entities and relationships in each critical asset
Goal: A plain list of what the asset is about (people, cases, payments, places) and how they connect.O: This playbook's own synthesis of published guidance
Steps
- With the steward, name the main things the asset records, in business words.
- For each, list its identifier and the few attributes people rely on most.
- Record how they relate, for example "a case has one or more payments".
- Attach the list to the catalogue record and keep it with the asset, not in a project folder.
- You produce
- Entity and relationship list
- Evidence it is done
- Each critical asset links to an entity list that its steward has checked.
- Measure
- Share of critical assets with an entity list.
- Who does what
- Accountable: Data owner
- Responsible (does the work): Data steward, Design authority or solution architect
- Consulted: Catalogue manager
- DAM Policy requirements it supports
- DAM-R13, DAM-R02
- Capabilities it builds
- Data architecture and data services, Data cataloguing and metadata management
- Templates
- Entity and relationship list
- Standards and guidance
- Data Catalog Vocabulary (DCAT) version 3V: Verified against the primary source on 2026-11-05
MOD-2 · Build: before your second return
Agree definitions for the attributes people rely on
Goal: The same word means the same thing in every system and report that uses the asset.O: This playbook's own synthesis of published guidance
Steps
- Pick the attributes that drive decisions, payments or published figures.
- Write one plain-English definition for each and name the steward who owns it.
- Check the definition against cross-government standards and the business glossary before writing a new one.
- Link each attribute to its glossary term, so a change to the definition reaches every model that uses it.
- You produce
- Attribute definitions linked to glossary terms
- Evidence it is done
- Key attributes link to an approved glossary term with a named owner.
- Measure
- Share of key attributes linked to an approved term.
- Who does what
- Accountable: Data steward
- Responsible (does the work): Design authority or solution architect
- Consulted: Catalogue manager, Data quality lead or analyst
- DAM Policy requirements it supports
- DAM-R13
- Capabilities it builds
- Interoperability, data standards and identifiers, Data ownership and stewardship
- Goes with
- META-1
- Standards and guidance
- SKOS: Simple Knowledge Organization System referenceV: Verified against the primary source on 2026-11-05
- Data Standards AuthorityV: Verified against the primary source on 2026-11-05
MOD-3 · Build: before your second return
Publish a data dictionary with the catalogue record
Goal: Users can see the columns, types and allowed values before they ask for access.O: This playbook's own synthesis of published guidance
Steps
- Export the column names, types, formats and allowed values from the authoritative source.
- Add the definitions from MOD-2 and mark personal or sensitive fields.
- Publish the dictionary as a machine-readable file (for example CSV on the Web) linked from the catalogue record.
- Regenerate it from the source when the structure changes, rather than editing by hand.
- You produce
- Machine-readable data dictionary
- Evidence it is done
- The catalogue record links to a current data dictionary.
- Measure
- Share of critical assets with a current dictionary.
- Who does what
- Accountable: Data steward
- Responsible (does the work): Catalogue manager, Data custodian
- Consulted: Data protection officer
- DAM Policy requirements it supports
- DAM-R02, DAM-R13
- Capabilities it builds
- Data cataloguing and metadata management, Interoperability, data standards and identifiers
- Standards and guidance
- CSV on the Web (CSVW) primerV: Verified against the primary source on 2026-11-05
- Data Catalog Vocabulary (DCAT) version 3V: Verified against the primary source on 2026-11-05
MOD-4 · Embed: business as usual
Control changes to models
Goal: A change to structure never breaks reports, sharing or quality rules without warning.O: This playbook's own synthesis of published guidance
Steps
- Version each model and dictionary. Record what changed and why.
- Before a change, list who uses the affected attributes, using the flow list and reuse records.
- Tell users in advance and give a date. Mark attributes as retired rather than deleting them.
- Where others depend on the asset, record the new structure in its data contract (see INT-4).
- You produce
- Versioned models
- Change notices
- Data contracts for shared assets
- Evidence it is done
- Structure changes have a version, a notice and a date.
- Measure
- Unplanned breaks to downstream users each year.
- Who does what
- Accountable: Data steward
- Responsible (does the work): Design authority or solution architect
- Consulted: Data quality lead or analyst
- Informed: Data users
- DAM Policy requirements it supports
- DAM-R04
- Capabilities it builds
- Interoperability, data standards and identifiers, Data risk management
- Goes with
- INT-4
- Standards and guidance
- Open Data Contract Standard (ODCS)V: Verified against the primary source on 2026-11-05
Knowledge area · 4 activities
Data storage and operations
Running the platforms that store data: backup, recovery, performance and retention.
Why it matters for the DAM Policy: A critical asset is only useful if it is available, recoverable and kept for the right time. The policy's purposes, such as public service delivery, depend on the data being there when needed.O: This playbook's own synthesis of published guidance
Usually led by: Data custodian. Content owner: Your technology operations or hosting team.
OPS-1 · Start: before your first return
Name the custodian and hosting for each critical asset
Goal: Someone is responsible for keeping each asset running, and you know where it is hosted.O: This playbook's own synthesis of published guidance
Steps
- Add a custodian to each ownership register entry, separate from the data owner.
- Record the hosting (on premises, cloud, supplier service) and the contract that covers it.
- Record where the data is physically stored, if your security or contracts require it.
- Tell the custodian which assets are critical and why.
- You produce
- Custodian and hosting in the ownership register
- Evidence it is done
- Every critical asset has a custodian distinct from its owner, or a recorded reason why not.
- Measure
- Share of critical assets with a named custodian.
- Who does what
- Accountable: Data owner
- Responsible (does the work): Data custodian
- Consulted: Data steward
- Informed: Lead policy coordinator
- DAM Policy requirements it supports
- DAM-R05, DAM-R02
- Capabilities it builds
- Data ownership and stewardship, Data architecture and data services
- Templates
- Data ownership register entry
- Standards and guidance
- Government Cloud First policyV: Verified against the primary source on 2026-11-05
- Cloud security guidanceV: Verified against the primary source on 2026-11-05
OPS-2 · Build: before your second return
Back up critical assets and test that you can restore them
Goal: A failure or attack does not lose critical data or the service that depends on it.O: This playbook's own synthesis of published guidance
Steps
- Agree with the owner how much data the service could lose and how long it could be without it.
- Check backups meet that, including copies held offline or out of reach of ransomware.
- Test a restore at least once a year and record the result.
- Fix gaps and record any accepted risk in the exception register.
- You produce
- Recovery targets
- Restore test record
- Evidence it is done
- A dated restore test for each critical asset.
- Measure
- Share of critical assets restored in a test in the last 12 months.
- Who does what
- Accountable: Data owner
- Responsible (does the work): Data custodian
- Consulted: Service or product owner, Information asset owner
- DAM Policy requirements it supports
- DAM-R09
- Capabilities it builds
- Cyber resilience and incident response for data assets, Data life cycle management
- Templates
- Exception register entry
- Standards and guidance
- Cyber Assessment Framework (CAF)V: Verified against the primary source on 2026-11-05
- Incident managementV: Verified against the primary source on 2026-11-05
OPS-3 · Build: before your second return
Monitor availability, freshness and capacity
Goal: Problems are found by the custodian before users find them.O: This playbook's own synthesis of published guidance
Steps
- Monitor whether each critical asset is available and when it was last updated.
- Set a freshness threshold with the data quality lead and alert when it is missed.
- Feed failures into the data quality issue log, not just the service desk.
- Review capacity before peaks, such as year end.
- You produce
- Availability and freshness monitoring
- Issues logged for failures
- Evidence it is done
- Freshness failures appear in the issue log with an owner.
- Measure
- Freshness failures for critical assets each quarter.
- Who does what
- Accountable: Data owner
- Responsible (does the work): Data custodian
- Consulted: Data quality lead or analyst
- DAM Policy requirements it supports
- DAM-R06, DAM-R15
- Capabilities it builds
- Data quality management through action plans, Data life cycle management
- Templates
- Issue log entry
- Standards and guidance
- Data on the Web Best Practices: Data Quality VocabularyV: Verified against the primary source on 2026-11-05
- Government Data Quality HubV: Verified against the primary source on 2026-11-05
OPS-4 · Embed: business as usual
Plan retention, archiving and decommissioning
Goal: Data is kept as long as needed and no longer, and nothing critical is lost when a system retires.O: This playbook's own synthesis of published guidance
Steps
- Apply the retention set for each record (see DOC-2) to every store and copy that holds it.
- Archive or delete copies on schedule and record what was done.
- When a system retires, migrate or archive critical data first, and update the catalogue record.
- Remove decommissioned copies so they cannot be used by mistake.
- You produce
- Retention linked to each store
- Decommissioning checklist
- Evidence it is done
- Retired systems show where their critical data went.
- Measure
- Retired systems with critical data accounted for.
- Who does what
- Accountable: Information asset owner
- Responsible (does the work): Data custodian, Data migration lead
- Consulted: Catalogue manager
- DAM Policy requirements it supports
- DAM-R04
- Capabilities it builds
- Data life cycle management, Legacy migration and decommissioning, Records, retention and disposal
- Goes with
- DOC-2
- Standards and guidance
- Advice on retentionS: Found by web search; the site blocked automated checking, so open it before relying on it
- Retiring your serviceV: Verified against the primary source on 2026-11-05
- Managing legacy technologyV: Verified against the primary source on 2026-11-05
Knowledge area · 5 activities
Data security
Protecting data from unauthorised access, change and loss, in line with classification.
Why it matters for the DAM Policy: In-scope assets are classified OFFICIAL, including handling markings such as 'personal' (DAM-R01). The catalogue must say how each asset is protected and who may use it, or it cannot be shared safely.O: This playbook's own synthesis of published guidance
Usually led by: Information asset owner. Content owner: Your security team, data protection officer and information asset owners.
SEC-1 · Start: before your first return
Record classification and personal data for each asset
Goal: Every catalogue record says how the asset is classified and whether it holds personal data.O: This playbook's own synthesis of published guidance
Steps
- Record the classification and any handling marking from your information asset register.
- Mark whether the asset holds personal or special category data, with the data protection officer.
- Link to the data protection impact assessment where there is one.
- Check the record before it goes into the return. Never put the data itself in the catalogue.
- You produce
- Classification and personal data fields in each record
- Evidence it is done
- No critical asset record has a blank classification.
- Measure
- Share of records with classification and personal data recorded.
- Who does what
- Accountable: Information asset owner
- Responsible (does the work): Data protection officer, Catalogue manager
- Consulted: Data steward
- DAM Policy requirements it supports
- DAM-R01, DAM-R13
- Capabilities it builds
- Security classification and protective controls, Data protection and lawful sharing
- Templates
- Critical data asset catalogue record
- Standards and guidance
- Government Security ClassificationsV: Verified against the primary source on 2026-11-05
- Data Protection Act 2018V: Verified against the primary source on 2026-11-05
- Data sharing: a code of practice - data sharing agreementsS: Found by web search; the site blocked automated checking, so open it before relying on it
SEC-2 · Start: before your first return
Record who can access each asset, and on what basis
Goal: People asking for data can see the route and the conditions without starting from scratch.O: This playbook's own synthesis of published guidance
Steps
- Record who can access the asset now (roles, not names) and how access is granted.
- Record the legal basis for any sharing and link to the agreement.
- Say how to request access in the catalogue record.
- Review access at least once a year with the owner.
- You produce
- Access rules and request route in the catalogue record
- Evidence it is done
- Each record says how to request access.
- Measure
- Share of records with an access route.
- Who does what
- Accountable: Data owner
- Responsible (does the work): Information asset owner, Catalogue manager
- Consulted: Data protection officer
- DAM Policy requirements it supports
- DAM-R02, DAM-R13
- Capabilities it builds
- Security classification and protective controls, Data protection and lawful sharing
- Templates
- Data sharing agreement
- Standards and guidance
- ICO data sharing code of practiceS: Found by web search; the site blocked automated checking, so open it before relying on it
- Digital Economy Act 2017V: Verified against the primary source on 2026-11-05
- Data (Use and Access) Act 2025V: Verified against the primary source on 2026-11-05
SEC-3 · Build: before your second return
Link critical assets to cyber risk
Goal: Security effort goes first to the data that matters most.O: This playbook's own synthesis of published guidance
Steps
- Give your security team the list of critical assets and their purposes.
- Map each to the systems and Cyber Assessment Framework outcomes that protect it.
- Add any gap to the risk register with an owner and a date.
- Use the same mapping as evidence for GovAssure rather than a separate survey.
- You produce
- Critical assets mapped to systems and CAF outcomes
- Risks recorded
- Evidence it is done
- Each critical asset is covered by a system in your cyber assurance scope.
- Measure
- Share of critical assets covered by cyber assurance.
- Who does what
- Accountable: Information asset owner
- Responsible (does the work): Data custodian
- Consulted: Data owner
- Informed: Chief data officer
- DAM Policy requirements it supports
- DAM-R09, DAM-R01
- Capabilities it builds
- Data risk management, Cyber resilience and incident response for data assets, Security classification and protective controls
- Standards and guidance
- Cyber Assessment Framework (CAF)V: Verified against the primary source on 2026-11-05
- Government Cyber Security Strategy: 2022 to 2030V: Verified against the primary source on 2026-11-05
- Government Functional Standard GovS 007: SecurityV: Verified against the primary source on 2026-11-05
SEC-4 · Build: before your second return
Make data incidents reach owners and quality leads
Goal: A breach or corruption of critical data is handled as a data problem as well as a security one.O: This playbook's own synthesis of published guidance
Steps
- Add critical asset owners to the incident response contact list.
- After an incident, check data integrity and log any quality impact.
- Tell users of affected data what changed and when it is fixed.
- Record lessons in the decision log and update controls.
- You produce
- Incident plan that names data owners
- Quality impact logged
- Evidence it is done
- Incidents on critical assets have a quality impact entry.
- Measure
- Time from incident to notifying data users.
- Who does what
- Accountable: Information asset owner
- Responsible (does the work): Data custodian, Data owner
- Consulted: Data quality lead or analyst, Data protection officer
- DAM Policy requirements it supports
- DAM-R05, DAM-R15
- Capabilities it builds
- Cyber resilience and incident response for data assets, Data quality issue management
- Templates
- Issue log entry, Decision log entry
- Standards and guidance
- Incident managementV: Verified against the primary source on 2026-11-05
SEC-5 · Embed: business as usual
Record which critical assets depend on encryption, and where they are stored
Goal: When encryption or data residency rules change, you can find the affected assets by a search, not a survey.O: This playbook's own synthesis of published guidance
Steps
- Add two fields to the catalogue record or asset register: crypto-dependency (does it rely on encryption?) and data residency (where is it stored?).
- For crypto-dependency, record whether the asset relies on public-key cryptography in storage or transfer, and where.
- For residency, record the country where it is stored and processed, and any contract terms.
- Use the fields to plan migration to post-quantum cryptography on the NCSC timeline.
- You produce
- Crypto-dependency and residency fields
- Evidence it is done
- Critical assets carry both fields.
- Measure
- Share of critical assets with both fields recorded.
- Who does what
- Accountable: Information asset owner
- Responsible (does the work): Data custodian, Catalogue manager
- Consulted: Design authority or solution architect
- DAM Policy requirements it supports
- DAM-R13
- Capabilities it builds
- Cyber resilience and incident response for data assets, Avoiding future legacy: open standards, portability and exit
- Standards and guidance
- Timelines for migration to post-quantum cryptographyV: Verified against the primary source on 2026-11-05
- Cyber Assessment Framework (CAF)V: Verified against the primary source on 2026-11-05
Draws on: Sovereign Standards PlatformO: This playbook's own synthesis of published guidance
Knowledge area · 4 activities
Data integration and interoperability
Moving and joining data between systems and organisations using shared standards.
Why it matters for the DAM Policy: The policy's introduction promises minimum requirements for interoperability (DAM-R16). Until they are published, using the identifiers and standards government already expects makes your assets easier to join and reuse.O: This playbook's own synthesis of published guidance
Usually led by: Design authority or solution architect. Content owner: Your architecture and integration teams, with data stewards.
INT-1 · Start: before your first return
Use the identifiers and standards government expects
Goal: Your critical assets can be matched with others without guesswork.O: This playbook's own synthesis of published guidance
Steps
- List the identifiers in each critical asset, such as UPRN for addresses or company numbers for organisations.
- Check each against the Data Standards Authority's recommended standards and the open standards list.
- Record the identifiers and standards used in the catalogue record.
- Where you use a local identifier, record how it maps to the national one.
- You produce
- Identifiers and standards recorded per asset
- Evidence it is done
- Records name the identifiers and standards used.
- Measure
- Share of critical assets that use at least one recommended identifier.
- Who does what
- Accountable: Data owner
- Responsible (does the work): Data steward, Catalogue manager
- Consulted: Design authority or solution architect
- DAM Policy requirements it supports
- DAM-R16, DAM-R13
- Capabilities it builds
- Interoperability, data standards and identifiers
- Standards and guidance
- Identifying property and street information (UPRN and USRN)V: Verified against the primary source on 2026-11-05
- Data Standards AuthorityV: Verified against the primary source on 2026-11-05
- Open standards for governmentV: Verified against the primary source on 2026-11-05
- Data Standards Authority: Standards CatalogueV: Verified against the primary source on 2026-11-05
INT-2 · Build: before your second return
Classify and crosswalk the standards each asset meets
Goal: Anyone can see which standards an asset meets, and how your terms map to them, in a form machines can read.O: This playbook's own synthesis of published guidance
Steps
- For each standard you use, place it by category, topic and subject. The draft government classification has five building blocks: metadata, common reference data, service and platform design, data lifecycle, and domain-specific.
- Map your local terms to the matching concepts in the cross-government standards, with a confidence level for each match.
- Mark terms with no match and send them to the Data Standards Authority rather than inventing a standard.
- Publish the statement of standards met as part of the catalogue record.
- You produce
- Standards statement per asset
- Term mappings with confidence
- Gaps sent to the DSA
- Evidence it is done
- Critical asset records carry a standards statement.
- Measure
- Share of critical assets with a standards statement.
- Who does what
- Accountable: Data owner
- Responsible (does the work): Data steward, Catalogue manager
- Consulted: Design authority or solution architect
- DAM Policy requirements it supports
- DAM-R16, DAM-R13
- Capabilities it builds
- Interoperability, data standards and identifiers
- Standards and guidance
- Data Standards AuthorityV: Verified against the primary source on 2026-11-05
- SKOS: Simple Knowledge Organization System referenceV: Verified against the primary source on 2026-11-05
- Data Catalog Vocabulary (DCAT) version 3V: Verified against the primary source on 2026-11-05
Draws on: Sovereign Standards PlatformO: This playbook's own synthesis of published guidance
INT-3 · Build: before your second return
Share critical data through documented interfaces
Goal: Other teams reuse the authoritative source, not a copy made for one project.O: This playbook's own synthesis of published guidance
Steps
- For each regular consumer, offer an API or a published distribution from the authoritative source.
- Document it with an open description format and link it from the catalogue record.
- Put a data sharing agreement in place for each external consumer.
- Retire one-off extracts once the interface is in use.
- You produce
- Documented interfaces
- Sharing agreements
- Evidence it is done
- Regular consumers use a documented interface.
- Measure
- One-off extracts of critical assets each year.
- Who does what
- Accountable: Data owner
- Responsible (does the work): Design authority or solution architect, Data custodian
- Consulted: Data protection officer
- Informed: Data users
- DAM Policy requirements it supports
- DAM-R12, DAM-R11
- Capabilities it builds
- Data architecture and data services, Data protection and lawful sharing, Cross-boundary collaboration
- Templates
- Data sharing agreement
- Standards and guidance
- API technical and data standardsV: Verified against the primary source on 2026-11-05
- OpenAPI SpecificationV: Verified against the primary source on 2026-11-05
- ICO data sharing code of practiceS: Found by web search; the site blocked automated checking, so open it before relying on it
INT-4 · Embed: business as usual
Agree data contracts and record lineage
Goal: Producers and consumers know what to expect, and anyone can trace a figure back to its source.O: This playbook's own synthesis of published guidance
Steps
- For shared assets, write a data contract covering structure, quality rules, freshness and support.
- Record lineage for critical flows in a standard format, from source through each transformation.
- Check contracts automatically where you can, and log breaches as quality issues.
- Review contracts with suppliers and partners each year.
- You produce
- Data contracts
- Lineage records
- Evidence it is done
- Shared critical assets have a contract and lineage.
- Measure
- Contract breaches each quarter.
- Who does what
- Accountable: Data owner
- Responsible (does the work): Design authority or solution architect, Data quality lead or analyst
- Consulted: Commercial lead
- DAM Policy requirements it supports
- DAM-R07, DAM-R06
- Capabilities it builds
- Interoperability, data standards and identifiers, Third-party and supplier data
- Standards and guidance
- Open Data Contract Standard (ODCS)V: Verified against the primary source on 2026-11-05
- OpenLineageV: Verified against the primary source on 2026-11-05
- PROV-O: the PROV ontologyV: Verified against the primary source on 2026-11-05
Knowledge area · 3 activities
Document and content management
Managing unstructured information: documents, content and records.
Why it matters for the DAM Policy: Many critical assets are also public records. The catalogue and the records schedule should agree on what the asset is, who is responsible for it and how long it is kept.O: This playbook's own synthesis of published guidance
Usually led by: Information asset owner. Content owner: Your records and information management team.
DOC-1 · Start: before your first return
Check which critical assets are also public records
Goal: Records managers and data owners work from the same list.O: This playbook's own synthesis of published guidance
Steps
- Share the critical asset list with your records team.
- For each asset, record whether it is a public record and which schedule covers it.
- Agree who decides on retention and disposal, the data owner or the information asset owner.
- Note any asset whose records status is unclear and resolve it before the return.
- You produce
- Records status in the ownership register
- Evidence it is done
- Each critical asset says whether it is a public record.
- Measure
- Share of critical assets with a records status.
- Who does what
- Accountable: Information asset owner
- Responsible (does the work): Lead policy coordinator
- Consulted: Data owner
- Informed: Catalogue manager
- DAM Policy requirements it supports
- DAM-R01, DAM-R11
- Capabilities it builds
- Records, retention and disposal
- Templates
- Data ownership register entry
- Standards and guidance
- Public Records Act 1958V: Verified against the primary source on 2026-11-05
- Manage information: guidance for governmentV: Verified against the primary source on 2026-11-05
DOC-2 · Build: before your second return
Attach retention and disposal to each catalogue record
Goal: Users can see how long data is kept and what happens to it next.O: This playbook's own synthesis of published guidance
Steps
- Copy the retention period and disposal action from the schedule into the catalogue record.
- Record the review date and who approved any longer retention.
- Check the record's own metadata is kept as long as the data.
- Flag assets past their retention date to the owner.
- You produce
- Retention and disposal fields filled in
- Evidence it is done
- No critical asset is past its retention date without an approved reason.
- Measure
- Assets past retention without a reason.
- Who does what
- Accountable: Information asset owner
- Responsible (does the work): Catalogue manager
- Consulted: Data owner
- DAM Policy requirements it supports
- DAM-R13, DAM-R04
- Capabilities it builds
- Records, retention and disposal, Data life cycle management
- Templates
- Critical data asset catalogue record, Data ownership register entry
- Standards and guidance
- Advice on retentionS: Found by web search; the site blocked automated checking, so open it before relying on it
- ISO 15489-1:2016 Records management: concepts and principlesS: Found by web search; the site blocked automated checking, so open it before relying on it
DOC-3 · Embed: business as usual
Transfer, archive or destroy with evidence
Goal: Disposal is lawful, recorded and reflected in the catalogue.O: This playbook's own synthesis of published guidance
Steps
- When disposal is due, carry out the action in the schedule and record it.
- For records selected for permanent preservation, follow the transfer process.
- Keep the catalogue record, marked as retired, with where the data went.
- Report disposals to the governance board each year.
- You produce
- Disposal records
- Retired catalogue records
- Evidence it is done
- Retired records say what happened to the data.
- Measure
- Disposals recorded each year.
- Who does what
- Accountable: Information asset owner
- Responsible (does the work): Data custodian, Catalogue manager
- Informed: Data governance board
- DAM Policy requirements it supports
- DAM-R04
- Capabilities it builds
- Records, retention and disposal, Legacy migration and decommissioning
- Standards and guidance
- Digital records transferV: Verified against the primary source on 2026-11-05
- ISO 23081-1:2017 Metadata for records: principlesS: Found by web search; the site blocked automated checking, so open it before relying on it
Knowledge area · 7 activities
Reference and master data
Agreeing and maintaining the shared, authoritative versions of key entities and code lists.
Why it matters for the DAM Policy: The policy asks for dates in ISO 8601 and plain values instead of blanks (DAM-R13). Reference data goes further: shared code lists for countries, places, organisations and categories are what let critical assets be joined and compared.O: This playbook's own synthesis of published guidance
Usually led by: Data steward. Content owner: Your data stewards, with the owners of national registers.
Read with the Reference data and identifiers supplement.
REF-1 · Start: before your first return
Find the reference data each critical asset depends on
Goal: You know which code lists each asset uses and where they come from.O: This playbook's own synthesis of published guidance
Steps
- List the coded fields in each critical asset, such as country, local authority, sector or status.
- For each, record the code list used and its source, for example ISO 3166 for countries or ONS codes for geography.
- Record dates in ISO 8601 and currencies in ISO 4217.
- Mark local code lists that duplicate a national one.
- You produce
- Code lists recorded per asset
- Evidence it is done
- Coded fields name their code list.
- Measure
- Share of coded fields with a named national code list.
- Who does what
- Accountable: Data owner
- Responsible (does the work): Data steward
- Consulted: Catalogue manager, Data quality lead or analyst
- DAM Policy requirements it supports
- DAM-R13
- Capabilities it builds
- Reference data, master data and authoritative sources, Interoperability, data standards and identifiers
- Standards and guidance
- ISO 8601-1:2019 Date and time: representations for information interchangeV: Verified against the primary source on 2026-11-05
- ISO 3166-1:2020 Country codesV: Verified against the primary source on 2026-11-05
- ISO 4217 Currency codesV: Verified against the primary source on 2026-11-05
- Names and codes listings (GSS geography codes)V: Verified against the primary source on 2026-11-05
REF-2 · Build: before your second return
Use shared controlled vocabularies, not local copies
Goal: Fewer local lists, so data joins across services without mapping by hand.O: This playbook's own synthesis of published guidance
Steps
- Replace local lists with the national one where it exists, for example ONS geography codes, UPRN and USRN, company numbers, school identifiers and the GSS harmonised standards for age, sex, ethnicity and language.
- Use SIC, SOC and CPV codes for sector, occupation and spend.
- Where you must keep a local list, map it to the national one and publish the mapping.
- Use shared values for status and lifecycle (draft, active, superseded, retired) across catalogues.
- You produce
- Local lists retired or mapped
- Evidence it is done
- Local lists that duplicate a national one have a mapping or a retirement date.
- Measure
- Local lists retired or mapped.
- Who does what
- Accountable: Data owner
- Responsible (does the work): Data steward
- Consulted: Design authority or solution architect, Data quality lead or analyst
- DAM Policy requirements it supports
- DAM-R16
- Capabilities it builds
- Reference data, master data and authoritative sources, Interoperability, data standards and identifiers
- Standards and guidance
- Harmonised standards and guidanceV: Verified against the primary source on 2026-11-05
- Names and codes listings (GSS geography codes)V: Verified against the primary source on 2026-11-05
- Identifying property and street information (UPRN and USRN)V: Verified against the primary source on 2026-11-05
- SKOS: Simple Knowledge Organization System referenceV: Verified against the primary source on 2026-11-05
Draws on: Sovereign Standards PlatformO: This playbook's own synthesis of published guidance
REF-3 · Build: before your second return
Name a master record for shared entities
Goal: People, organisations and places are recorded once and matched consistently.O: This playbook's own synthesis of published guidance
Steps
- Pick the shared entities your critical assets hold in more than one place.
- Name the master source for each and the identifier used to match records.
- Write the matching and duplicate rules, and who resolves doubtful matches.
- Measure duplicates and fix them at source.
- You produce
- Master source and matching rules per entity
- Evidence it is done
- Shared entities have a named master source.
- Measure
- Duplicate rate for shared entities.
- Who does what
- Accountable: Data owner
- Responsible (does the work): Data steward, Design authority or solution architect
- Consulted: Data quality lead or analyst
- DAM Policy requirements it supports
- DAM-R12
- Capabilities it builds
- Reference data, master data and authoritative sources, Data architecture and data services
- Standards and guidance
- Identifying property and street information (UPRN and USRN)V: Verified against the primary source on 2026-11-05
- Develop your data and APIs using a reference architectureV: Verified against the primary source on 2026-11-05
REF-5 · Build: before your second return
Find peers before you buy or build
Goal: You learn from public bodies that already run the tool or keep the list, before you commit.O: This playbook's own synthesis of published guidance
Steps
- Look up the tool in the repos asset catalogue's register of expertise, which lists the public bodies that buy or build with it, from tender and contract notices and their public code.
- Check whether the list you need is already published, for example the GOV.UK organisations, the planning data registers, the NHS Organisation Data Service, Get Information about Schools or Companies House.
- Ask in the cross-government data management community, and note who has done it before.
- Read any public code peers share, such as deployment and pipeline code, before you design your own.
- You produce
- A list of peers and existing sources
- Questions to ask them
- Evidence it is done
- The business case names peers asked and existing sources considered.
- Measure
- Proposals that name peers and existing sources.
- Who does what
- Accountable: Data owner
- Responsible (does the work): Data steward, Design authority or solution architect
- DAM Policy requirements it supports
- DAM-R16
- Capabilities it builds
- Reference data, master data and authoritative sources
- Standards and guidance
- Be open and use open sourceV: Verified against the primary source on 2026-11-05
- Choosing technology: an introductionV: Verified against the primary source on 2026-11-05
REF-6 · Build: before your second return
Choose a master data tool with exit in mind
Goal: The tool fits how you run services, and your master data can leave it.O: This playbook's own synthesis of published guidance
Steps
- Ask how it is hosted and supported, and whether your cloud and support teams can run it without the supplier.
- Ask how releases and upgrades work, and how you test them before they reach live data.
- Ask how security vulnerabilities are found, reported and fixed, and how fast.
- Check the screens your stewards and developers will use meet the accessibility regulations, not only the public ones.
- Make export a contract term. Master and reference data, with match and survivorship rules and history, must come out in an open format through a documented API.
- You produce
- Evaluation against hosting
- releases
- support
- vulnerabilities
- accessibility and exit
- Exit terms in the contract
- Evidence it is done
- The contract names the export format and the API.
- Measure
- Master data tools with tested export.
- Who does what
- Accountable: Data owner
- Responsible (does the work): Design authority or solution architect
- Consulted: Data steward
- DAM Policy requirements it supports
- DAM-R12
- Capabilities it builds
- Reference data, master data and authoritative sources, Data architecture and data services
- Standards and guidance
- Technology Code of Practice point 11: define your purchasing strategyV: Verified against the primary source on 2026-11-05
- Technology Code of Practice point 4: make use of open standardsV: Verified against the primary source on 2026-11-05
- Cloud security guidanceV: Verified against the primary source on 2026-11-05
- Understanding accessibility requirements for public sector bodiesV: Verified against the primary source on 2026-11-05
- API technical and data standardsV: Verified against the primary source on 2026-11-05
REF-7 · Embed: business as usual
Publish shared lists as open, versioned data
Goal: Others reuse your list instead of copying it, and every copy can tell when it is out of date.O: This playbook's own synthesis of published guidance
Steps
- Name an owner and a steward for each list you publish.
- Give every entry a persistent identifier that does not change when its name or web address does.
- Publish JSON first, with CSV and its metadata for spreadsheets, and a catalogue record so others can find it.
- Take changes as proposals a steward accepts or rejects in the open, and version every release.
- Join other bodies' lists by their identifiers rather than copying their values.
- You produce
- Published list with owner
- identifiers and versions
- Catalogue record
- Evidence it is done
- The list has a named owner, persistent identifiers and a version history.
- Measure
- Published lists reused by another team.
- Who does what
- Accountable: Data owner
- Responsible (does the work): Data steward, Catalogue manager
- Informed: Data users
- DAM Policy requirements it supports
- DAM-R04
- Capabilities it builds
- Reference data, master data and authoritative sources, Interoperability, data standards and identifiers
- Standards and guidance
- Publish reference data for use across governmentV: Verified against the primary source on 2026-11-05
- SKOS: Simple Knowledge Organization System referenceV: Verified against the primary source on 2026-11-05
- API technical and data standardsV: Verified against the primary source on 2026-11-05
REF-4 · Embed: business as usual
Govern changes to code lists
Goal: A change to a code list never silently breaks a report or a share.O: This playbook's own synthesis of published guidance
Steps
- Version each code list you own. Record what changed and from when.
- Never delete a code. Mark it as retired and point to its replacement.
- Tell users before a change takes effect.
- Publish your code lists in an open format so others can reuse them.
- You produce
- Versioned code lists
- Change notices
- Evidence it is done
- Code lists have versions and deprecation markers.
- Measure
- Code list changes with notice given.
- Who does what
- Accountable: Data steward
- Responsible (does the work): Catalogue manager
- Informed: Data users
- DAM Policy requirements it supports
- DAM-R04
- Capabilities it builds
- Reference data, master data and authoritative sources, Interoperability, data standards and identifiers
- Standards and guidance
- SKOS: Simple Knowledge Organization System referenceV: Verified against the primary source on 2026-11-05
Knowledge area · 4 activities
Data warehousing and business intelligence
Bringing data together for analysis, reporting and decisions.
Why it matters for the DAM Policy: Each critical asset must support at least one of the policy's seven purposes (DAM-R11). Reports, dashboards and statistics are where much of that value shows, and where quality problems reach decision makers.O: This playbook's own synthesis of published guidance
Usually led by: Data quality lead or analyst. Content owner: Your analysis and reporting teams.
BI-1 · Start: before your first return
List the reports and statistics that use each critical asset
Goal: You can show who relies on the asset, which supports its designation and its priority.O: This playbook's own synthesis of published guidance
Steps
- Ask analysis and reporting teams which reports, dashboards and statistics use each critical asset.
- Record each use against the asset, with the decision or publication it supports.
- Use the list as evidence of the purpose in the return.
- Share it with the data quality lead, who uses it to set the importance of purpose in the issues framework.
- You produce
- Reuse list per asset
- Evidence it is done
- Each critical asset has at least one recorded use.
- Measure
- Recorded uses per critical asset.
- Who does what
- Accountable: Data owner
- Responsible (does the work): Data users, Data steward
- Consulted: Data quality lead or analyst
- DAM Policy requirements it supports
- DAM-R11, DAM-R15
- Capabilities it builds
- Data strategy and prioritisation by reuse value, Performance measurement and annual review
- Standards and guidance
- Code of Practice for StatisticsV: Verified against the primary source on 2026-11-05
BI-2 · Build: before your second return
Show quality and provenance next to published figures
Goal: People reading a figure know how good the data behind it is.O: This playbook's own synthesis of published guidance
Steps
- Publish the data quality statement with each regular report that uses a critical asset.
- Say where the data came from and when it was extracted.
- Flag known issues from the issue log that affect the figures.
- Follow the Code of Practice for Statistics where the output is official statistics.
- You produce
- Quality statements with reports
- Evidence it is done
- Regular reports link to a quality statement.
- Measure
- Share of regular reports with a quality statement.
- Who does what
- Accountable: Data owner
- Responsible (does the work): Data quality lead or analyst, Data users
- DAM Policy requirements it supports
- DAM-R06
- Capabilities it builds
- Data quality management through action plans, Performance measurement and annual review
- Templates
- Data quality statement
- Standards and guidance
- Quality assurance of administrative dataS: Found by web search; the site blocked automated checking, so open it before relying on it
- Code of Practice for StatisticsV: Verified against the primary source on 2026-11-05
BI-3 · Build: before your second return
Make analytical copies traceable to their source
Goal: Every warehouse table or extract can be traced back to its critical asset.O: This playbook's own synthesis of published guidance
Steps
- Record which critical asset each analytical table comes from.
- Record the transformations applied, in the lineage format agreed for shared flows (see INT-4).
- Remove unmanaged copies, or bring them under the same quality rules.
- Link analytical copies from the catalogue record as distributions.
- You produce
- Lineage from source to report
- Evidence it is done
- Analytical copies link back to their source.
- Measure
- Share of analytical copies with recorded lineage.
- Who does what
- Accountable: Data owner
- Responsible (does the work): Data quality lead or analyst, Catalogue manager
- Consulted: Design authority or solution architect
- DAM Policy requirements it supports
- DAM-R12
- Capabilities it builds
- Data architecture and data services, Data risk management
- Goes with
- INT-4
- Standards and guidance
- PROV-O: the PROV ontologyV: Verified against the primary source on 2026-11-05
- OpenLineageV: Verified against the primary source on 2026-11-05
BI-4 · Embed: business as usual
Measure the value and reuse of critical assets
Goal: The annual review knows which assets earn their keep and which need investment.O: This playbook's own synthesis of published guidance
Steps
- Count uses, users and decisions supported for each critical asset each year.
- Set this beside the cost of keeping the asset and its quality score.
- Report to the governance board as part of the annual review.
- Use the results to prioritise improvement and investment.
- You produce
- Annual value and reuse summary
- Evidence it is done
- The annual review includes value and reuse.
- Measure
- Critical assets with a value and reuse figure.
- Who does what
- Accountable: Chief data officer
- Responsible (does the work): Data quality lead or analyst, Lead policy coordinator
- Consulted: Data governance board
- DAM Policy requirements it supports
- DAM-R04
- Capabilities it builds
- Benefits realisation and evaluation, Performance measurement and annual review, Investment, business cases and value
- Standards and guidance
- The Green Book and accompanying guidanceV: Verified against the primary source on 2026-11-05
Knowledge area · 2 activities
Metadata management
Describing data so people and machines can find, understand and trust it.
Why it matters for the DAM Policy: The cataloguing playbook covers what the policy asks for. These activities go further, toward governed vocabularies and machine-readable metadata that AI tools and the National Data Library can use.O: This playbook's own synthesis of published guidance
Usually led by: Catalogue manager. Content owner: Your catalogue manager, with data stewards.
The core activities for this area are the cataloguing playbook. The ones below go further.
META-1 · Build: before your second return
Govern a business glossary
Goal: Each important term has one agreed meaning, an owner and a history.O: This playbook's own synthesis of published guidance
Steps
- For each term, record a preferred label, other labels (synonyms and abbreviations), a one-sentence plain-English definition and a scope note saying when to use it.
- Name the owner (a steward), the source standard or law it comes from, and the body that endorsed it.
- Give it a status (draft, in review, approved, deprecated), a version and valid from and to dates.
- Map it to the equivalent term in other standards. When a term retires, point to its replacement; terms are deprecated, never deleted.
- Review each term every year, or when its source changes, and log every change.
- You produce
- A business glossary with owners
- status and history
- Evidence it is done
- Approved terms have an owner, a source and a review date.
- Measure
- Share of governed terms with a named steward.
- Who does what
- Accountable: Data steward
- Responsible (does the work): Catalogue manager
- Consulted: Data governance board
- Informed: Data users
- DAM Policy requirements it supports
- DAM-R13
- Capabilities it builds
- Data cataloguing and metadata management, Data ownership and stewardship, Interoperability, data standards and identifiers
- Standards and guidance
- SKOS: Simple Knowledge Organization System referenceV: Verified against the primary source on 2026-11-05
- Metadata standards for sharing and publishing dataV: Verified against the primary source on 2026-11-05
Draws on: Sovereign Standards PlatformO: This playbook's own synthesis of published guidance
META-2 · Embed: business as usual
Climb from glossary to knowledge graph
Goal: Your terms, standards, assets and owners become linked data that people, catalogues and AI tools can query.O: This playbook's own synthesis of published guidance
Steps
- Level 1, glossary: agreed terms with owners (META-1).
- Level 2, thesaurus: relate terms as broader, narrower, related, and "use instead", so search finds the preferred term.
- Level 3, vocabularies and taxonomies: closed code lists and a classification tree (category, topic, subject) applied the same way everywhere.
- Level 4, ontologies: describe concepts with SKOS, data sets with DCAT, lineage with PROV-O and quality with DQV, and validate them with SHACL shapes before publishing.
- Level 5, knowledge graph: give each term, standard, asset and owner a persistent, resolvable identifier and publish the links, so questions such as "who owns this term?" and "what breaks if it changes?" can be answered by query.
- You produce
- Machine-readable vocabularies with persistent identifiers
- Validation in your publishing pipeline
- Evidence it is done
- Published vocabularies resolve and pass validation.
- Measure
- Share of governed terms with a resolvable identifier.
- Who does what
- Accountable: Catalogue manager
- Responsible (does the work): Design authority or solution architect
- Consulted: Data steward
- DAM Policy requirements it supports
- DAM-R02
- Capabilities it builds
- Data cataloguing and metadata management, Interoperability, data standards and identifiers, AI readiness and data and AI ethics
- Standards and guidance
- SKOS: Simple Knowledge Organization System referenceV: Verified against the primary source on 2026-11-05
- Data Catalog Vocabulary (DCAT) version 3V: Verified against the primary source on 2026-11-05
- PROV-O: the PROV ontologyV: Verified against the primary source on 2026-11-05
- Data on the Web Best Practices: Data Quality VocabularyV: Verified against the primary source on 2026-11-05
- Shapes Constraint Language (SHACL)V: Verified against the primary source on 2026-11-05
Draws on: Sovereign Standards PlatformO: This playbook's own synthesis of published guidance
Knowledge area · 1 activities
Data quality
Defining, measuring and improving how fit data is for its purposes.
Why it matters for the DAM Policy: The quality playbook covers what the policy asks for. This activity prepares critical assets for use with AI, which the capability model treats as an inferred need.O: This playbook's own synthesis of published guidance
Usually led by: Data quality lead or analyst. Content owner: Your data quality lead.
The core activities for this area are the data quality playbook. The ones below go further.
DQ-AI · Embed: business as usual
Make critical assets ready for AI
Goal: Critical assets can be used by AI tools safely, with known quality, meaning and limits.O: This playbook's own synthesis of published guidance
Steps
- Check the asset has a current catalogue record, glossary-linked definitions and a quality statement.
- Record known biases, gaps in coverage and groups under-represented in the data.
- Publish machine-readable metadata with persistent identifiers so retrieval tools cite the right source.
- Assess any AI use against the data ethics framework and record algorithmic tools where required.
- You produce
- AI readiness check per asset
- Evidence it is done
- Assets used by AI have a readiness check.
- Measure
- Critical assets used by AI with a readiness check.
- Who does what
- Accountable: Data owner
- Responsible (does the work): Data quality lead or analyst, Data steward
- Consulted: Data protection officer
- DAM Policy requirements it supports
- DAM-R06, DAM-R13
- Capabilities it builds
- AI readiness and data and AI ethics, Equality and inclusion in data
- Standards and guidance
- Making government datasets ready for AIV: Verified against the primary source on 2026-11-05
- Data and AI Ethics FrameworkV: Verified against the primary source on 2026-11-05
- Algorithmic Transparency Recording Standard hubV: Verified against the primary source on 2026-11-05
Was this page useful? Tell us Tell us privately, without GitHub