A well-run SAMA BCM tabletop exercise ends with numbers: how long each critical service took to recover against the objective the institution has committed to. Many exercises end instead with a transcript of who said what. They rehearse incident response, which matters, but they leave the recovery-time commitments untested. This article names four recovery objectives worth testing before the annual resilience assessment and describes a scenario-design method that makes the results measurable.
The objectives below reflect the institution's own recovery-time commitments under the SAMA Business Continuity Management framework. Your approved business impact analysis and BCM policy define the exact values. Use those documents, not the figures in this article or in anyone else's template.
Why discussion-based exercises fail to measure recovery
A typical tabletop presents an incident, asks participants what they would do, and records the conversation. Three weaknesses follow.
- Participants state intent, not duration. "We would fail over to the secondary site" carries no time.
- Dependencies stay implicit. Nobody checks whether the person, credential, vendor or approval needed for failover is available.
- Nothing is compared to an objective. Without a stated recovery time objective (RTO) at the start, there is nothing to measure against at the end.
Incident response readiness and recovery readiness are related but distinct. Response covers roles, escalation and notification in the first hours. Recovery covers restoring the service within its objective. Both need testing, and this article concentrates on the second.
The four recovery-time objectives to test
1. Payment-system failover
Test the time from the decision to fail over until payments are processed again through the alternate path. Include connectivity to the relevant national payment infrastructure, certificates and credentials at the secondary site, and the reconciliation of in-flight transactions. Failover is rarely a single event, so the objective should be measured at each stage.
2. Core-banking resumption
Test the time to bring core banking back to a usable, consistent state. This is where the recovery point objective (RPO) sits next to the RTO. Ask how much data would be lost and how long validation of the restored data would take. Restoration without validation is not resumption.
3. Customer-channel restoration
Test mobile, online, branch, ATM and contact-centre channels, and the order in which they return. A channel can be technically up while its upstream dependency is not. The measure that matters is when customers can complete a transaction, not when a server responds.
4. Regulatory-reporting continuity
Test whether the institution can keep meeting its obligations to notify and report to SAMA during a disruption. Reporting depends on people, data and systems that may themselves be affected. Measure how long it takes to produce an accurate first notification and a follow-up update with the primary reporting tools unavailable.
The scenario-design method
The method turns each objective into a timed, observable exercise. It has six steps.
- Start from the BIA. Take the critical services, dependencies and RTO/RPO values from the approved impact analysis. If a service has no stated objective, record that as a finding before the exercise begins.
- Write one scenario per objective, with a single trigger. Keep the cause simple, such as loss of a data centre, a ransomware event that makes core systems unavailable, or a failure at a critical third party. Complexity belongs in the dependencies, not the story.
- Define start and stop events. For each objective, agree in writing what starts the clock and what proves recovery. For example, "first successful customer payment through the alternate path" is a stop event; "team confirms failover complete" is not.
- Build timed injects. Release information on a schedule, including unwelcome facts: the on-call engineer is unreachable, the backup is older than expected, the vendor needs an approval. Each inject should test a dependency identified in the BIA.
- Record timestamps, not opinions. Assign an observer to each objective to log decisions, hand-offs and delays with times. Ask participants for realistic estimates from runbooks and past tests, and have the observer note when an estimate is a guess.
- Compare and classify. Set measured or estimated duration against the objective. Classify each result as met, at risk or not met, and record the cause of any gap.
Making failover duration realistic
Realistic duration comes from adding up the steps that people tend to leave out: detection, assessment, decision and approval, technical failover, validation, and communication. Participants often quote only the technical step. Ask each owner to list every step, assign a duration to each, and identify which steps depend on a person or an approval. Where an estimate cannot be supported, schedule a technical test to replace the estimate with a measurement. A tabletop identifies where to look; a technical test confirms the time.
Who should be in the room
Include the service owners for each of the four areas, IT recovery and infrastructure leads, the information security team, third-party relationship owners, communications, and a senior executive with authority to approve failover. Add a compliance or regulatory-affairs representative for the fourth objective. Observers should come from outside the teams being measured where possible.
What to produce afterwards
The output should be an exercise report that a reviewer can read without having attended. It should contain:
- the scenario, start and stop events and objectives for each of the four areas
- measured and estimated durations against each objective
- gaps classified by cause, such as people, process, technology or third party
- an improvement plan with owners and dates
- a note on which results need a technical test to confirm
Keep the evidence together with the BIA and plans so that the link between objective, test and fix is visible during the annual resilience assessment.
How we support this
Our Business Continuity & Cyber Resilience service builds and tests your business continuity and cyber resilience capability in line with SAMA BCM and SAMA cyber resilience requirements. The work follows four stages: Analyze, Strategize, Plan and Exercise. In the Exercise stage, we run tabletop and technical tests and deliver lessons and fixes. The deliverables include a BIA report, a BCM policy and framework, business continuity and DR plans, cyber recovery playbooks, and exercise reports with an improvement plan. The project runs 8–12 weeks.
Where your priority is the response side, our Incident Response Readiness & Tabletop Exercises service prepares technical teams and executives on roles, escalation and regulatory notification. If you would like to design your next SAMA BCM tabletop exercise around measurable recovery objectives, speak with us about the Business Continuity & Cyber Resilience service.
