Skip to main content

Templates and registers

These are optional starting points for projects that need a written record. Adapt, shorten or ignore them to suit the kaupapa. They are not a substitute for the relationships and decisions around the work.

Decision context register

This can be useful where somebody returning to a dataset later will need to know why it exists and how the current version has been used.

FieldRecord
Dataset / knowledge group
Kaupapa
People or decision process
Knowledge holder or source relationship
Intended uses
Intended audiences
Spatial detail
Derived analysis in the current kaupapa
Public version, if any
AI use in the current kaupapa, if any
Review trigger or date
Contact / role
Notes / contested interests

Only record personal or cultural detail here when it is actually useful to the purpose of the register.

Sensitive-layer review

For a layer where location or attributes could matter, these questions can help shape a shared version:

  • What is the current kaupapa?
  • Who is the intended audience?
  • Is exact geometry useful for this output?
  • Would a generalised version do the same job?
  • Could the map imply more certainty than the source supports?
  • What can be inferred using parcels, imagery, elevation, roads, historic maps and environmental data?
  • What does the metadata reveal?
  • Are attachments part of the intended output?
  • Are all fields useful to this audience?
  • Can users export or cache the data?
  • Does access need to be temporary for this project?
  • What change would prompt another review?

Copy register

For datasets with several working or external copies, a simple register can show where they have gone:

CopyLocation / providerPurposeCustodianAudienceUsers / groupCreatedReviewStatus
Main working copy
Backup
Disaster recovery
Mobile/offline
Test/development
Consultant
Archive
Public derivative

Add rows for material exports and research copies where useful.

Cloud assessment worksheet

Organisation and use

  • Who owns the subscription or tenant?
  • Who are the organisational administrators?
  • What kaupapa will the service support?
  • What kinds of data are expected to be held there?
  • What material, if any, will stay elsewhere?

Location and jurisdiction

  • Primary storage region/country:
  • Processing locations:
  • Backup/DR locations:
  • Support-access locations:
  • Provider contracting entity:
  • Provider domicile:
  • Relevant foreign jurisdiction considerations:
  • Subprocessors:

Security

  • MFA available/enforced:
  • Role model:
  • Encryption in transit:
  • Encryption at rest:
  • Customer-managed keys if relevant:
  • Logging available:
  • Incident notification:
  • Backup restore tested:

Data lifecycle

  • Complete export format:
  • Attachments exportable:
  • Metadata exportable:
  • Deleted-data retention:
  • Backup expiry:
  • Exit process:
  • Egress/exit cost:
  • Independent archive location:

AI

  • Is content used for shared model training?
  • Is content used for evaluation or product improvement?
  • Can AI features be disabled?
  • Are prompts/outputs retained?
  • How would use of AI change the original kaupapa or audience?

Supplier questionnaire

Questions worth getting clear answers to before relying heavily on a GIS or data supplier include:

  1. Who owns the tenant, account and billing relationship?
  2. Who owns the domain and certificates?
  3. Who holds administrator rights?
  4. Where is each data class stored?
  5. Where is it processed?
  6. Where are backups and disaster-recovery copies?
  7. Which subprocessors are used and where?
  8. What legal jurisdictions may apply to the provider?
  9. What security certifications or assurance reports are available?
  10. Is customer material used for AI training, evaluation or unrelated model improvement?
  11. What activity and access logs are available to the client?
  12. What happens if payment stops?
  13. How is a complete export performed?
  14. Which spatial formats are available?
  15. Are attachments included?
  16. Is metadata included?
  17. Are relationships, schemas and coded values included?
  18. Is application configuration included?
  19. How long does deleted information remain in backups?
  20. What is the termination process?
  21. What is the exit or egress cost?
  22. Who owns custom source code?
  23. Where is that code stored?
  24. What documentation is delivered?
  25. What internal capability will exist after handover?
  26. How are supplier and subprocessor copies returned or deleted?
  27. How will a migration be tested before contract end?

Consultant handover checklist

At project close it can be useful to confirm:

  • production accounts sit with the organisation continuing the work
  • more than one organisational administrator can access the system
  • domain and certificates are under organisational control
  • source code is in the agreed repository
  • main data have been exported
  • attachments are included
  • metadata and schema are included
  • public derivatives are identified
  • passwords/secrets are transferred securely
  • supplier access is documented and can be removed
  • backup and restore procedure has been demonstrated
  • full export has been tested
  • runbook is current
  • licences and renewal dates are recorded
  • known risks and unresolved issues are listed
  • working copies and backup-retention arrangements are understood
  • internal staff have had practical administration handover

Data-sharing agreement outline

For projects that need a formal data-sharing agreement, the headings below can help structure the discussion. Consequential agreements may also need legal advice. Te Kāhui Raraunga's 2026 Iwi Data Sharing Guidance is a useful reference for individual-level iwi data.

Parties and kaupapa

  • parties
  • kaupapa and expected iwi or collective benefit
  • people or process involved in the sharing decision
  • whether the purpose can be met without individual-level data
  • effective period

Information

  • datasets covered
  • minimum information required
  • whether information can be de-identified before transfer
  • source
  • spatial precision
  • sensitive attributes
  • metadata
  • attachments

Use

  • named purpose
  • users/roles
  • analyses expected
  • whether the data will be linked or integrated with other datasets
  • whether derived data will be created
  • secondary uses that are outside the agreement
  • public outputs
  • mapping and precision arrangements

Decision context

  • contacts
  • consent or other requirements that apply to the particular data
  • correction process
  • review process
  • handling of contested interests
  • agreed benefit or capability contribution where relevant

Security and copies

  • access controls across the full data pipeline
  • storage/processing locations
  • subcontractors
  • exports
  • mobile copies
  • backups
  • breach notification

AI

  • whether AI forms part of the work
  • purposes for which it will be used
  • model training or product-improvement settings
  • treatment of AI-derived spatial information

Return, archive and deletion

  • data return format
  • derivatives to return
  • repatriation arrangements
  • archive responsibilities
  • retention period
  • live deletion
  • backup expiry
  • evidence of destruction where the agreement calls for it

Termination

  • what happens when the kaupapa ends
  • continuing obligations
  • transfer to another custodian
  • dispute or escalation process

Annual recovery check

For systems expected to last, an occasional recovery check can confirm:

  • another person in the organisation can find the data
  • another administrator can recover the platform
  • one backup has been restored
  • a complete export still works
  • external access still reflects current projects
  • supplier details and costs are current
  • important technology claims have been rechecked

Source

Last reviewed: 25 August 2026