A role, a person, and an organization are different concepts

One person may cover several roles in a small, bounded scope, while several people may share one role in a larger scope. What matters is whether one named role makes the final decision, the performer has competency and capacity, and a backup or escalation route preserves continuity.

A supplier can perform work without taking over business decisions, data and access approval, production risk, incident priority, spend, legal terms, or exit. Internal-led, hybrid, and supplier-led models can each be suitable depending on the project.

Product mechanics reveal work surfaces, not staffing numbers

A standalone app contains source, environment, infrastructure, migrations, tests, and deployment artifacts. Modules can add user interface, APIs, entities, migrations, events, workers, and overrides. This supports roles for architecture, development, data, testing, platform, and operation, but does not prove a person count, seniority mix, schedule, or supplier form.

Production launch does not end responsibility

Production workers run as separate processes with an asynchronous Redis queue. Structured logs create a signal, but the team still chooses collection, retention, access, alerting, and incident use. Upgrades can contain downstream actions. A service operator, recovery owner, upgrade owner, and incident route must be named before production handover.

A ready map means information is complete for review

Ready for decision review requires an owner, performer, competency evidence, confirmed capacity, backup or escalation, and handover path for every critical domain. It does not mean the team, system, security posture, supplier, or launch is approved.

Five tests for every assignment

A title, supplier contract, or tool license does not pass these tests by itself.

Accountable outcome owner

One named human role can make the final decision and accept the consequence.

Responsible performer

The internal or supplier role that performs the work is explicit.

Demonstrated competency

Relevant artifacts and observed outcomes support the assignment, not a title alone.

Available capacity

The role has agreed time, service coverage, tools, and decision access for the expected demand.

Continuity and backup

A backup or usable escalation route can act when the primary performer is unavailable.

A role is not a person or organization. One person may cover several archetypes in a bounded scope, several people may share one role in a larger scope, and a supplier may perform work while one internal human role remains accountable.

Product mechanics that trigger ownership

A standalone app contains source, environment, infrastructure, migrations, tests, and deployment artifacts. Modules can add pages, APIs, dependency injection, entities, migrations, validation, events, workers, translations, and overrides. Ejected modules become app-local maintenance. Production workers run separately with an asynchronous Redis queue. Structured logs exist, but collection, retention, access, alerts, and incident use remain operating decisions. Upgrade notes can require downstream action.

Product fact identifies a responsibility trigger. Editorial mapping proposes an owner pattern. Your project decides the real assignment. Supplier statements remain attributed commercial claims.

Revision: 01911d00e28f44cf484d0b1d04860dcfef5370bf · Reviewed: 2026-07-14

Lifecycle responsibility map

Lifecycle responsibility map
StageDecisionsOutputsAccountable role typeLikely performersEvidenceHandoffFailure signal
EvaluationFit, business outcome, edition and sourcing decisionDecision brief and unresolved assumptionsSponsorSponsor, process owner, technical advisorNamed outcome owner and evidence ledgerApproved scope hypothesisNo owner or supplier claim used as proof
DiscoveryProcesses, boundaries, data, users and acceptanceScope, process map and acceptance outcomesProduct or process ownerBusiness analysts, users, data owner, supplierSigned decisions with source ownersArchitecture and delivery inputsUnknown process or data authority
ArchitectureBoundaries, modules, tenancy, integrations and operationsDecision records and responsibility boundariesSolution architecture ownerArchitect, engineers, security and platform rolesReviewed decisions linked to product mechanicsBuild plan and control requirementsArchitecture exists only in supplier knowledge
BuildConfiguration, local modules, overrides and integrationsReviewed source, migrations and documentationProduct and technical ownersApplication and integration engineersReviewable source and change evidenceVersioned release candidateUnreviewed code or no internal access
Data and integrationMigration, interfaces, reconciliation and ownershipMapped data, trials and failure handlingData and integration ownersData, application and source-system performersTrial evidence and reconciliation resultsAccepted dataset and operating runbookUnowned mapping or silent failure path
VerificationTechnical tests, controls and business acceptanceTraceable test and acceptance evidenceQuality and business acceptance ownersEngineers, testers, process users and security reviewersObserved results against agreed outcomesRelease evidence packGenerated tests treated as approval
ReleaseDeployment, migration, rollback and final decisionApproved runbook, rollback and ownership handoffRelease owner and accountable sponsorPlatform, application, data and service rolesRehearsal and signed residual gapsOperable production serviceNo day-two operator or rollback owner
OperationAvailability, support, incidents, recovery and monitoringService records, exercises and reviewed signalsService ownerSupport, platform, database and application rolesIncident and recovery exercise evidenceKnown service health and action backlogSupplier-only access or no incident route
ChangeUpgrades, dependencies, features and technical debtImpact record, regression and rollback evidenceProduct, technical and release ownersEngineering, platform and acceptance rolesExact-version rehearsal and reviewUpdated runbooks and baselineNo upgrade owner or test capacity
ExitSource, data, credentials, knowledge and supplier transitionReproducible build, exports, revoked access and walkthroughSponsor and knowledge or exit ownerInternal owners and outgoing or incoming supplier rolesIndependent build and handover exerciseAccepted custody and transition backlogNo source, build, data or decision history

Responsibility domain matrix

Responsibility domain matrix
DomainProduct or project triggerRequired outcomeLifecycle stagesAccountable role typePossible performersCompetency evidenceCapacity evidenceBackupEscalationCadenceHandover artifactNot applicable only when
Executive sponsorship and outcomesAny evaluation or funded scopeBusiness outcome, risk appetite and spend decisions have one ownerEvaluation, release, change, exitAccountable sponsorInternal sponsor office, shared advisorsDecision brief and accepted tradeoffsDecision access and scheduled gate attendanceNamed delegated decision roleExecutive governance routeAt every material gateDecision ledger and residual risksNever for an active project
Product and process ownershipAny business workflow in scopeScope, priorities and acceptance outcomes remain coherentDiscovery through changeProduct or process ownerInternal users, analysts, supplier facilitatorsProcess map and accepted outcome examplesBacklog and acceptance time reservedDeputy process decision roleSponsorEach planning and acceptance cyclePrioritized backlog and acceptance rulesOnly with a recorded no-business-process reason
User and adoption ownershipPeople will use or be affected by the applicationAccess, communication, training and feedback are ownedDiscovery, verification, release, operationAdoption or process ownerInternal champions, support, training supplierUsability evidence and support contentCoverage for launch and new startersSecondary support routeService or product ownerBefore release and recurring reviewRole-based guidance and issue themesOnly for a verified non-user-facing scope
Data governance and migrationExisting, imported, created or exported dataMeaning, quality, access, retention and reconciliation are decidedDiscovery, data, verification, operation, exitData ownerData engineers, source owners, supplierMapping, trial and reconciliation evidenceTime for cleansing, trials and exception reviewDeputy data stewardSponsor and privacy ownerEach migration and material data changeData dictionary, mapping and export proofOnly if no project data exists, with owner and review trigger
Solution architectureAny deployed application or integration boundarySystem boundaries and technical decisions are reviewableEvaluation through changeSolution architecture ownerInternal architect, senior engineers, supplier architectDecision records and reviewed system contextReview coverage for material changesSecond reviewer with repository accessTechnical governance and sponsorAt baseline and material changeArchitecture records and dependency mapNever for a deployed application
Module and customization designConfiguration, custom fields, local modules, overrides or ejectionThe smallest justified mechanism has owners and rollbackArchitecture, build, verification, changeProduct and technical ownerApplication engineers, supplier developersDesign record, reviewed code and testsMaintenance and regression timeReviewer able to maintain the mechanismArchitecture ownerEach customization and upgradeSource, tests, disable path and upgrade triggerWhen no customization is selected, record decision and review trigger
Application development and reviewAny app-local source or changed package usageChanges are reviewable, reproducible and maintainableBuild, verification, change, exitTechnical ownerApplication engineers and reviewersReviewed changes, build and defect evidenceDelivery, review and maintenance allocationIndependent code reviewerArchitecture and product ownersEvery change setRepository, build instructions and decision historyOnly if no app-local change exists, with exact baseline
API and integration ownershipAny external interface, event, webhook, file or synchronizationContracts, reconciliation and failure handling are ownedArchitecture, data, verification, operation, changeIntegration and process ownersIntegration engineers and external-system teamsContract tests and failure rehearsalCoverage for incidents and upstream changeSecond interface ownerService and data ownersAt change and recurring reconciliation reviewContract, credentials custody and runbookWhen no interface exists, record owner and review trigger
Security, privacy, access and tenancyAny users, roles, sensitive data, organizations or tenantsAccess, data isolation, control acceptance and response are ownedDiscovery through exitSecurity, privacy and business data ownersEngineers, operators, reviewers, supplierThreat and access review with test evidenceReview and incident coverageNamed security escalation roleAccountable risk ownerBefore access, release and material changeAccess model, decisions, findings and response routeNever when application data or users exist
Platform, environment, secrets and configurationAny running environmentDeployment, configuration and secret custody remain controlledArchitecture, release, operation, change, exitPlatform or service ownerInternal operator, cloud team, hosting supplierDeployment rehearsal and configuration evidenceCoverage for deploys and service eventsSecondary operator with approved accessService owner and security ownerEach release and configuration reviewInfrastructure definition, inventory and access transferNever for a running environment
Database, migration, backup and recoveryAny persistent application dataSchema change, backup custody and recovery proof are ownedBuild, release, operation, change, exitData recovery or platform ownerDatabase and platform operatorsMigration rehearsal and recovery proofCoverage for release and recovery eventsSecond operator with restore accessService, data and sponsor rolesEach migration and scheduled recovery exerciseSchema history, backup inventory and restore runbookNever when persistent data exists
Queues, Redis, workers and scheduled processingAsynchronous queues, workers, events or schedules are enabledSeparate processes, retries, idempotency and failure handling are operatedArchitecture, build, release, operation, changeService or platform ownerApplication and platform operatorsQueue failure and retry exerciseWorker monitoring and incident coverageSecondary operatorApplication and service ownersContinuous signals and recurring reviewQueue inventory, concurrency choices and runbookWhen disabled, retain decision owner and enablement review trigger
Logging, monitoring and observabilityAny operated environmentSignals, collection, retention, access, alerts and incident use are decidedVerification, release, operation, changeService and security ownersApplication and platform operatorsMonitoring review and incident exerciseAlert and investigation coverageSecondary responderIncident routeContinuous signals and scheduled reviewSignal catalog, access, alerts and runbookNever for a production service
Test strategy and business acceptanceAny changed or accepted behaviorTechnical and business evidence matches agreed outcomesDiscovery, build, verification, release, changeQuality and business acceptance ownersEngineers, testers, users, supplierTraceable test evidence and observed acceptanceRegression and user review timeDeputy acceptance roleProduct owner and sponsorEvery release and material changeTest inventory, results and accepted gapsNever for changed behavior
Release, deployment, rollback and change controlAny environment promotion or production changeOne controlled change can be deployed, observed and reversedVerification, release, operation, changeRelease ownerApplication, platform, data and supplier rolesDeployment and rollback rehearsalRelease window and response coverageSecondary release operatorSponsor and service ownerEvery releaseRelease record, artifacts and rollback decisionNever for production change
Service desk and incident responseUsers or business processes depend on the serviceRequests, incidents, priority and communication have routesRelease, operation, change, exitService ownerInternal support, supplier support, technical respondersTicket and incident exercise evidenceAgreed service coverage and surge routeSecondary responderNamed incident command and business routeContinuous intake and recurring reviewService catalog, contacts, known errors and runbooksOnly for a verified non-operated experiment
Upgrade, dependency and technical-debt ownershipVersioned packages, dependencies or ejected code existRelease review, impact, regression and backlog remain ownedOperation, change, exitTechnical and product ownersApplication engineers, supplier, platform rolesExact-version review and rehearsalRecurring maintenance and regression allocationSecond maintainerSponsor for deferred riskEach release window and scheduled debt reviewDependency baseline, upgrade log and debt registerNever while versioned dependencies are operated
Licensing, supplier, knowledge, handover and exitAny external code, supplier, package, service or future transitionRights, source, data, build, access and knowledge can transferEvaluation through exitProcurement, technical and knowledge or exit ownersInternal custodians, legal reviewer, outgoing and incoming supplierDocument review, independent build and walkthroughTime and access for recurring handover upkeepInternal repository and credential custodianSponsor and qualified legal routeAt procurement, acceptance and recurring transition reviewRights register, source, build, data export and access inventoryNever when an external dependency or transition exists

Three neutral sourcing patterns

Three neutral sourcing patterns
PatternStrengthsDependenciesEvidence to demandTransition conditionFailure mode
Internal-ledDirect context and custodyInternal capacity and specialist gapsDecision records, reviewed work and backup coverageUse specialist support when evidence or capacity is missingTitles hide missing competence or overloaded owners
HybridInternal decisions with flexible specialist executionClear boundaries, shared tools and coordinated handoffsOne final owner per decision, shared evidence and review accessRebalance when internal competence or supplier dependency changesShared responsibility has no final decision owner
Supplier-led with retained internal accountabilityBroad external execution and specialist capacityContract scope, evidence access, continuity and exitInternal decision owners, repository and credential custody, independent acceptanceTransfer knowledge and access before service or supplier changeSupplier-only credentials, knowledge, build or production decisions

Supplier coverage is not ownership transfer

A supplier may perform most technical work. Named internal roles still decide scope, data meaning, access approval, business acceptance, production risk, incident priority, spend, legal terms, and exit. Supplier execution is evidence to inspect, not transferred accountability.

Competency evidence comes from observed work

Use architecture decisions, reviewed code, migration rehearsal, threat and access review, test results, deployment rehearsal, recovery proof, monitoring and incident exercises, exact-version upgrade rehearsal, documentation, and handover walkthroughs. Credentials may add context but do not certify coverage.

AI tools assist work, not accountability

AI assistants, coding agents, generated tests, and repository skills may support analysis, implementation, and documentation. They cannot be accountable owners, approve business acceptance, hold legal duties, accept security risk, command incidents, or replace competent human review and operation.

Evidence gates

Evidence gates
GateRequired owner and evidence
Evaluation approvalOutcome owner, scope assumptions, evidence sources, missing competence and escalation
Implementation startProcess, data, architecture, delivery, security and acceptance owners with capacity
Production handoverDay-two operator, incident route, recovery proof, release evidence, backup and accepted gaps
Ongoing serviceService signals, support capacity, recurring reviews, upgrade owner and current handover artifacts
Material changeImpact owner, technical review, regression, security and data decisions, rollback and operating update
Supplier exitInternal decision ownership, source and build, data export, revoked access, knowledge walkthrough and receiving capacity

Stop or escalate when

  • No business outcome owner
  • Production credentials are held only by a supplier
  • Data meaning, migration or retention decisions are unowned
  • No security or access approver
  • No test strategy or business acceptance owner
  • No day-two application or platform operator
  • No incident route or decision priority owner
  • No owner for backup and recovery proof
  • No upgrade and dependency owner
  • No source, reproducible build or handover path
  • A critical responsibility depends on one person
  • Assigned roles lack confirmed capacity
  • Responsibility is shared but no final decision owner exists

Sanitized examples

Smaller hybrid implementation

An internal sponsor and process owner keep scope and acceptance. A supplier performs application and integration work. An internal platform role controls environments and secrets. Gaps remain where recovery has not been rehearsed and the supplier engineer has no named backup. Decision: do not hand over to production until recovery evidence and continuity are recorded.

Supplier-led implementation approaching handover

A supplier performs delivery, tests, and operation setup. Internal data, security, service, spend, legal, and acceptance roles retain decisions. The repository is accessible internally, but the build walkthrough and production credential transfer are incomplete. Decision: keep the exit and production-handover gates open.

Evidence classification and cutoff

Reviewed on 2026-07-14 at the pinned product revision. Product facts come from repository and documentation files. Responsibility mappings are editorial. Actual assignments, capacity, accepted gaps, and service choices are project decisions. Supplier promises are time-sensitive claims requiring contract evidence.

No headcount recommendation, maturity score, certification, or aggregate readiness percentage is produced.

Browser-local planning tool

Responsibility and continuity register

The 18 starter rows mirror the domain matrix. The bounded state can say responsibility map ready for decision review only after every row has the required assignment, evidence, capacity, continuity, and handover fields. It cannot approve a team, system, security posture, or launch.

Privacy: Entries stay in this browser and local exports. Use role labels, not names. Do not enter personal, confidential, security-sensitive, production, or credential data.

Register state

NOT READY FOR DECISION REVIEW

Open information gates: 180

The 18 starter rows mirror the domain matrix. The bounded state can say responsibility map ready for decision review only after every row has the required assignment, evidence, capacity, continuity, and handover fields. It cannot approve a team, system, security posture, or launch.
DomainLifecycleProduct or project triggerRequired outcomeAccountable ownerPerformerSourcing modeCompetency evidenceCapacity evidenceBackupEscalationEvidence link or noteHandover artifactCadenceCoverage stateReview dateNext action or not-applicable reason
Executive sponsorship and outcomes
Product and process ownership
User and adoption ownership
Data governance and migration
Solution architecture
Module and customization design
Application development and review
API and integration ownership
Security, privacy, access and tenancy
Platform, environment, secrets and configuration
Database, migration, backup and recovery
Queues, Redis, workers and scheduled processing
Logging, monitoring and observability
Test strategy and business acceptance
Release, deployment, rollback and change control
Service desk and incident response
Upgrade, dependency and technical-debt ownership
Licensing, supplier, knowledge, handover and exit

Method, assumptions and limitations

Reviewed 14 July 2026. Product facts were checked against both the public code at the cited repository revision and official documentation. Where documentation and code differ, this guide describes behavior supported by code. Interpretations and recommendations concern implementation work, not product guarantees.

This material is not a quote, audit, certification, or legal, tax, or accounting advice. Edition, enabled modules, configuration, custom code, infrastructure, data, third-party providers, and operating practices affect the outcome.