Start hereDefine critical operations and impact tolerances
~105 days leftOpen step
Guide library
Regulation Asia–Pacific

APRA CPS 230

Prudential Standard CPS 230 Operational Risk Management

Identify critical operations, define acceptable disruption and test reliance on material service providers.

Who it applies to

Entities covered by the Australian prudential operational-risk standard, with application and any specific transitions governed by the applicable determination.

What you’ll produce

Operational risk controls, continuity tolerances and tests, and a maintained material service-provider program.

A tolerance has to work in a real disruption. Test how business teams meet it when key technology, people or suppliers are unavailable at the same time.

TEAM KICKOFF

Run the project

Go to the roadmap

Decisions for leadership

Which operations cannot tolerate a prolonged disruption?
Agree which operations are critical, who owns them and the tolerances for disruption, data loss and service levels under the applicable standard.
Which providers could stop those operations?
Map material service providers, dependency concentration and fallback arrangements. Fund contract remediation, testing and realistic exit arrangements.
How will the board know if the controls work?
Give leadership actual exercise results, unresolved tolerance failures and required decisions. Confirm accountability for notifying APRA and maintaining access to its reporting channel.

How to spend the first week

  1. Pick a regulated critical operation and trace the people, premises, data, technology and providers needed for minimum service.
  2. Ask its owner to show the recovery tolerance, actual restoration results and critical provider obligations. Identify what is based on an untested assumption.
  3. Schedule a recovery exercise and agree how the response team decides whether and when to notify APRA.

What to review each week

Track critical operations without verified recovery, material provider arrangements with unresolved gaps and reportable disruptions with approaching deadlines.

Kickoff brief to share with your team

Fill this in with your sponsor and send it to the people who own the work.

APRA CPS 230 kickoff brief
Business reason and required outcome  [Contract, obligation or approval]
Scope and exclusions  [Entities, systems or activities, and reasons]
Fixed deadlines and internal milestones  [Date, source and owner]
Executive sponsor and authorized signer  [Names]
Project lead and backup  [Names]
First workstream  Define critical operations and impact tolerances · Board + operational risk lead [Name and due date]
People, effort and external costs  [Estimate, owner and contingency]
Evidence folder and reviewer access  [Location, access owner and review date]
Decisions and blockers  [Item, options, owner and decision date]

First week
1. Pick a regulated critical operation and trace the people, premises, data, technology and providers needed for minimum service.
2. Ask its owner to show the recovery tolerance, actual restoration results and critical provider obligations. Identify what is based on an untested assumption.
3. Schedule a recovery exercise and agree how the response team decides whether and when to notify APRA.
How to hand off evidence
  1. Record the requirement or claim, the person responsible, the period and the result. Link to the original report, configuration, test or case record.
  2. Include the version, system or product, date, and how you gathered the record. When a report covers a group of items, retain its filters and reconcile the count to the source.
  3. Ask the reviewer to open the evidence and reproduce the key check. Record their decision and date. Keep missing items or failures visible with an owner, a due date and a retest when needed.
  4. Share through the agreed channel. Keep sensitive originals in a restricted location and link any redacted copies to the corresponding originals.
BEFORE YOU START

How this works in practice

Two questions to check scope

  • Which operations must continue through disruption, and what would customers or markets experience if they stop?
  • Which direct providers and downstream dependencies could prevent recovery within tolerance?

Worked scenario

An Australian bank tests a payments outage that also affects its usual support provider. The continuity team measures actual downtime and workarounds against documented tolerances, then sends missed thresholds to the owner of the remediation plan.

What to read next

  • APRA CPS 234

    Assign security obligations and test information asset controls.

  • ISO 22301

    Use a management system to structure business continuity work.

The whole journeyabout 105 days across 4 steps
  1. 01Define critical operations and impact tolerancesBoard + operational risk lead · ~21 days
  2. 02Build and test recovery optionsContinuity lead + operations · ~35 days
  3. 03Manage material provider arrangementsSupplier risk + procurement · ~35 days
  4. 04Report incidents and material disruptionsIncident lead + compliance · ~14 days
IMPLEMENTATION ROADMAP

Step by step

Checklist, instructions and 10 templates.

View plan as a table
Project roadmap with current target dates and completion status
StepOwnerWhen to actEstimated targetDeliverablesStatus
Define critical operations and impact tolerancesBoard + operational risk leadBefore continuity planningSet a start dateBusiness impact analysis; Critical-function inventoryOpen
Build and test recovery optionsContinuity lead + operationsBefore relying on the arrangements and on the required scheduleSet a start dateContinuity plan; Exercise report; Corrective-action registerOpen
Manage material provider arrangementsSupplier risk + procurementBefore contracting and on the applicable review cycleSet a start dateVendor review; Contract gap analysis; Exit plansOpen
Report incidents and material disruptionsIncident lead + complianceWhen applicable thresholds are metSet a start dateIncident classification; Notification recordOpen
0 of 4 completeOpen workspace

Targets use illustrative calendar-day estimates and assume the steps run in sequence. Check official dates and agree real durations with your team.

  1. 01

    Define critical operations and impact tolerances

    Who Board + operational risk lead When Before continuity planning

    Map operations, customers and dependencies. Set tolerances and record the assumptions, approval and measurement methods.

    How to do this step

    Before you start

    Service and entity inventory, dependency maps, board-approved risk settings and regulatory scope.

    1. Identify critical operations from the customer or market service delivered. Map systems, data, people, locations and providers that keep it operating.
    2. Propose maximum disruption, data loss and minimum service levels using realistic harms and legal obligations.
    3. Take the tolerances and dependencies through the required governance approvals. Record who can update them when a service changes.

    What a reviewer should see

    A payments operation has approved time, data-loss and minimum-service tolerances and a dependency map checked by Operations.

    When to escalate

    Raise an operation classified as critical with no owner or tolerance decision.

    What to produce

    Business impact analysisSpreadsheet

    Gather first

    • Process maps
    • Transaction volumes and legal or contractual obligations
    • Service owners and recovery records

    How to prepare it

    1. Interview each process owner about what fails when their service stops. Record impact over useful time intervals and distinguish customer, legal, safety and financial effects.
    2. List the people, sites, technology, information and providers needed to recover. Agree the maximum tolerable disruption, target recovery time, acceptable data loss and minimum service level.
    3. Check that upstream and downstream targets fit together. Ask the accountable executive to approve the priorities and record which dependencies need testing.

    Worked example

    Fictional details. Use the same fields with evidence from your work.

    Operation
    Fulfil urgent medicine orders
    Impact as disruption grows
    Manual work can cover the first two hours. Missed dispatch after six hours affects treatment schedules.
    Dependencies
    Stock data, label printers and courier booking
    Recovery target
    Resume priority shipments in four hours with no more than 30 minutes of lost order data.
    Owner and approval
    Warehouse director approves. Continuity team checks the targets during the next exercise.

    Check before handing it over

    Each target links to an observable impact and to the actual resources needed for recovery.

    Give the approved priorities to the people planning recovery and testing suppliers.

    Critical-function inventorySpreadsheet

    Gather first

    • Asset exports or system discovery
    • Service owner interviews
    • Approved scope and architecture

    How to prepare it

    1. Export a dated baseline from the source system. Give each asset, service or record class a stable ID and remove duplicates without losing links to the source.
    2. Add ownership, purpose, sensitivity and any vendor or system it depends on. Mark unknowns explicitly and assign them to someone to resolve.
    3. Reconcile the register against an independent source such as cloud billing or the production deployment list. Set a refresh cadence and an owner.

    Worked example

    Fictional details. Use the same fields with evidence from your work.

    ID
    SYS-014
    Item
    Production billing database
    Owner
    Platform operations
    Location and purpose
    Production EU region, stores invoice and account data
    Impact and dependencies
    High. Identity provider and managed backups.
    Verified on
    2026-11-10, cloud inventory export

    Check before handing it over

    Each item has one accountable owner. The source and as-of date let a reviewer trace it back.

    Share the reviewed inventory with the people building the control, privacy or testing scope.

  2. 02

    Build and test recovery options

    Who Continuity lead + operations When Before relying on the arrangements and on the required schedule

    Prepare continuity plans. Test a severe but plausible scenario, measure actual performance and have owners address weaknesses.

    How to do this step

    Before you start

    Approved tolerances, continuity plans, exercises and actual failover and restore evidence.

    1. Write who declares a disruption, where to get emergency access and how teams continue minimum service.
    2. Simulate loss of an essential dependency. Measure time and data recovered and compare service volume and integrity to your tolerances.
    3. Assign failed limits to a remediation owner with interim measures and a test date. Escalate unproven workarounds to management.

    What a reviewer should see

    An exercise records start time, first restored service, recovered records, minimum service output and what needs correction.

    When to escalate

    Escalate a test that breaches tolerance or depends on an unverified provider promise.

    What to produce

    Continuity planMemo

    Gather first

    • Approved business impact analysis
    • Staff and supplier contact lists
    • Recovery procedures and technology dependencies

    How to prepare it

    1. Write when the plan is triggered, who declares it and how staff learn which version is active. Put first-hour actions near the top.
    2. Describe the workarounds and recovery steps by service. Specify dependencies, checks before restoration, fallback if a supplier is unavailable and criteria for returning to normal.
    3. Run a timed exercise without relying on the usual owner. Record achieved service and data recovery against targets and assign any fixes.

    Worked example

    Fictional details. Use the same fields with evidence from your work.

    Trigger
    Order system down for 30 minutes with no confirmed recovery time
    First actions
    Incident lead opens a log. Dispatch lead starts the emergency order queue.
    Recovery sequence
    Recover database, reconcile recent orders, verify labels, resume priority dispatch.
    Fallback
    Use backup printer and call the alternate courier contact.
    Return to normal
    Service owner signs off after reconciliation and successful shipment test.

    Check before handing it over

    Someone covering the primary owner can activate the plan and verify recovery with the written instructions.

    Give controlled copies to responders and suppliers and update the record after each exercise.

    Exercise reportMemo

    Gather first

    • Approved scenario and participant roles
    • Goals and scoring criteria
    • Event timeline and expected records

    How to prepare it

    1. Prepare a scenario and timed new information that tests decisions, handoffs and evidence. Assign a facilitator and separate observer.
    2. During the exercise, note what participants actually did, the timestamps and where expected responses failed. Distinguish simulated actions from real ones.
    3. Write a short after-action summary with owners, deadlines and a retest plan. Attach the scenario, attendance and evidence references.

    Worked example

    Fictional details. Use the same fields with evidence from your work.

    Scenario
    Simulated export leak found at 09:20 UTC
    Objective
    Reach operational and legal leads and classify the incident within the exercise target.
    Observed behavior
    Engineering preserved logs. Initial escalation went to an inactive mailbox.
    Action and owner
    Privacy operations updates the routing list before 2026-12-01.
    Retest
    Repeat the handoff in the next on-call drill.

    Check before handing it over

    An absent participant can see what was tested, what happened and why an action was assigned.

    Send approved findings to action owners and file the record for governance or assessment.

    Corrective-action registerSpreadsheet

    Gather first

    • Finding, risk or test exception
    • Root-cause analysis
    • Named owner, due date and expected evidence

    How to prepare it

    1. Record the observation and link the exact test or requirement. Describe the business effect and investigate why the issue was able to happen.
    2. Separate immediate containment from the permanent fix. Give each action an owner, date, verification method and any required approval for deferral.
    3. Retest against the original failure once the fix is live. Record who accepted the outcome and keep unresolved issues visible.

    Worked example

    Fictional details. Use the same fields with evidence from your work.

    Finding or risk
    F-008, a departed contractor retained production access
    Cause and effect
    HR event never reached IAM. Admin access remained active.
    Action and owner
    Revoke access and alert on failed HR sync. IAM lead.
    Due and status
    2026-11-28, in progress
    Closure test
    Simulate departure and verify access disappears within policy
    Approver and evidence
    Risk owner, ticket IAM-114 and retest results

    Check before handing it over

    The closure test addresses the original problem and has a dated result from an independent reviewer where required.

    Send open items and decisions to governance, audit or the assessor on the applicable cadence.

  3. 03

    Manage material provider arrangements

    Who Supplier risk + procurement When Before contracting and on the applicable review cycle

    Identify and maintain the required register, check contract and monitoring requirements, and plan how to leave or replace a provider.

    How to do this step

    Before you start

    Contract register, dependency maps, service descriptions, provider evidence and renewal dates.

    1. Identify material service providers using the applicable categories and actual business dependencies. Confirm accountability for services and relevant subcontracting.
    2. Check contracts for relevant provisions, including access, continuity, notifications and termination. Record each gap and its commercial owner.
    3. Test the exit or substitute path, including who can export data, transfer service and validate it after migration.

    What a reviewer should see

    The provider record links a critical operation to the contract, continuity terms, current due diligence and tested or scheduled exit work.

    When to escalate

    Raise material services without workable continuity or cooperation terms before signing or renewal.

    What to produce

    Vendor reviewSpreadsheet

    Gather first

    • Signed contracts and supplier inventory
    • Critical functions, service relationships and subprocessors
    • Reporting instructions and data dictionary when applicable

    How to prepare it

    1. Identify the contracting entity, supplier identifier, service, dates and location. Map dependencies to critical functions and distinguish direct providers from subcontractors.
    2. Gather evidence on security, continuity, location, access and exit. In a required reporting register, use the current authority workbook and allowed identifiers.
    3. Reconcile records against contracts and procurement data, and validate reference keys. Send missing or inconsistent elements to the supplier and record the final approval.

    Worked example

    Fictional details. Use the same fields with evidence from your work.

    Arrangement ID
    ICT-018, managed database hosting
    Contract and provider
    Enter signed agreement reference and verified legal entity identifier.
    Service and function
    Managed database supporting invoicing. Record entity and function reference.
    Location and subcontractors
    Check documented data locations and supplier chain.
    Evidence and status
    Contract verified, subcontractor fields under review.

    Check before handing it over

    Each provider and contract record links to an actual function and all required identifiers are present.

    Use the verified register in supplier reviews and put required submissions in the authority’s prescribed format.

    Contract gap analysisMemo

    Gather first

    • Current agreement and service description
    • Applicable flow-down terms or engagement requirements
    • Contract owner and legal review

    How to prepare it

    1. Make a clause-by-clause worksheet. Quote only what you need to identify the current clause, and record each required subject and unresolved question.
    2. Mark where scope, instructions, evidence, security, subprocessors, incident notice, audit rights or termination affect the work. Route proposed wording to authorized reviewers.
    3. Record the signed or agreed version, responsible parties and the dates you need to revisit it. Save the source agreement in its restricted store.

    Worked example

    Fictional details. Use the same fields with evidence from your work.

    Party and service
    Northstar hosting provider, production backup
    Clause or requirement
    Incident notification and contacts
    Current position
    General notice terms. No operational escalation contacts listed.
    Change to review
    Add named response channel and request confirmation of the timing.
    Owner, status and record
    Vendor lead, legal review pending, contract register V-018

    Check before handing it over

    The executed version covers the agreed roles and evidence paths. Review any jurisdiction-specific clauses with counsel.

    Give named obligations to operations and procurement. Use this worksheet to track review and keep the executed agreement as the authority.

    Exit plansMemo

    Gather first

    • Approved scope
    • Governing requirements and risks
    • Teams, service dependencies and review gates

    How to prepare it

    1. State the objective, scope, accountable owner and conditions that trigger action. Translate commitments into tasks with periods, thresholds and evidence.
    2. For each task set the input, decision rule, output and handoff. Include escalation for failures and approvals for exceptions.
    3. Walk through one real case with owners and reviewers. Approve the version, schedule review, and check that promised tasks have operational records.

    Worked example

    Fictional details. Use the same fields with evidence from your work.

    Purpose
    Identify failed production backup jobs and restore service data.
    Scope
    Production billing databases and backup vault
    Activity and timing
    On-call reviews daily job alerts and runs a quarterly restore test.
    Evidence and exception
    Keep job records and restore test output. Page the service owner if a job fails.
    Owner and approval
    Operations lead owns. Service owner approves.

    Check before handing it over

    A person who receives the plan can tell when to act, what to check and where to save the result.

    Give the approved plan to the people responsible for operation. Add dated checkpoints to the working calendar.

  4. 04

    Report incidents and material disruptions

    Who Incident lead + compliance When When applicable thresholds are met

    Evaluate incident materiality and disruption outside tolerance as separate triggers. Retain awareness, duration, impact and the submitted notices.

    How to do this step

    Before you start

    Alert timeline, approved tolerances, impact measurements, reporting criteria and APRA contact procedure.

    1. Record awareness, known impact and actual tolerance breaches. Apply the separate triggers for operational risk incidents and critical operation disruptions.
    2. Confirm which reporting timeline applies under the current standard and assign a filer and backup. Draft facts and unknowns even while recovery continues.
    3. Preserve the submitted material, receipt, follow-up and retest. Review changes required in tolerances, provider response or communication.

    What a reviewer should see

    The record shows when the team identified the trigger, how it calculated the deadline and when the notice was sent.

    When to escalate

    Escalate a possible reportable event promptly and assign someone to track the regulatory clock during recovery.

    What to produce

    Incident classificationSpreadsheet

    Gather first

    • Alert or breach report
    • Investigation facts and relevant thresholds
    • Escalation contacts and official reporting clocks

    How to prepare it

    1. Start a case with UTC timestamps for detection, awareness, assessment and containment. Keep known facts separate from estimates and update both as evidence changes.
    2. Identify affected systems, people, data and likely harms. Link your classification, notice decision and the rule that starts each clock.
    3. Record containment, preservation, escalation and notification actions. Capture who approved the decision and set a time to confirm outstanding facts.

    Worked example

    Fictional details. Use the same fields with evidence from your work.

    Incident ID
    INC-041
    Awareness and facts
    2026-11-13 09:20 UTC. Export link exposed to an unauthorized account.
    Scope and uncertainty
    One confirmed access. Log review for further recipients in progress.
    Decision and reason
    Privacy lead to assess reporting threshold with counsel by 10:30 UTC.
    Actions and status
    Access revoked, logs preserved, next review at 12:00 UTC.

    Check before handing it over

    Timelines distinguish different events and a reviewer can reconstruct the reporting decision from the available facts.

    Give the incident lead and privacy decision maker a time-stamped update. Preserve working evidence separately.

    Notification recordSpreadsheet

    Gather first

    • Case, change or communication records
    • Required fields and timing
    • Named case owner and authorized recipients

    How to prepare it

    1. Give each event a record ID. Record the action or request, receipt or send time, owner and the deadline or next date.
    2. Write the decision and link the response or supporting record. Track delivery, escalation or verification separately so a sent message is distinguishable from an accepted one.
    3. Review open rows on a regular cadence. Restrict personal details to the case system and keep the register to IDs and necessary dates.

    Worked example

    Fictional details. Use the same fields with evidence from your work.

    Record ID
    REQ-022
    Event and date
    Verified access request received 2026-11-07 UTC
    Owner and target
    Privacy operations, internal review by 2026-11-23
    Decision and status
    Approved response, sent securely
    Completion evidence
    Case delivery record, 2026-11-24
    Follow-up
    None, approved by the reviewer

    Check before handing it over

    Every open item has a next action and closed items have a verifiable record of closure or delivery.

    Use the log at case reviews and provide a restricted copy only where the engagement requires it.

PRIMARY REFERENCES

Official sources

How we build these guides

Keyboard shortcuts

⌘K or /
Search everything
?
This list
g h
Home
g l
Guide library
g f
Find my path
g t
Templates
g p
Playbooks
g w
My workspace
g d
Deadline calendar
g c
Regulatory changes