Cannabis POS Platform Failover: A Practical Downtime Runbook

A cannabis point-of-sale outage can interrupt more than checkout. It may affect age verification, purchase limits, package selection, inventory deductions, payments, receipts, online orders, and compliance reporting. Without a prepared response, employees may create duplicate transactions or inventory changes that are hard to explain later.

A dependable cannabis POS platform should provide clear system alerts, protected transaction queues, and detailed audit logs. Technology alone, however, is not a failover plan. The store needs a written runbook explaining what must stop, what may continue, and who can restore normal operations.

What POS Failover Means in Cannabis Retail

Failover is the controlled move from a failed system to an approved backup workflow. Staff may transfer to another register, activate secondary internet, use an authorized regulatory portal, or temporarily stop sales.

A point-of-sale built for cannabis retail often connects with payments, e-commerce, inventory, and a state track-and-trace platform. Metrc explains that integrated retail and ERP systems exchange compliance data through its API.

A payment approval does not prove that inventory or compliance data synchronized successfully. Treat register availability, payment availability, and reporting connectivity as separate conditions.

Classify the Incident First

Identify the Failure Type

Train employees to distinguish among common scenarios:

• One terminal, scanner, or printer has failed;
• The internet or entire POS is unavailable;
• Payment processing is offline;
• The POS works, but Metrc synchronization is delayed;
• A cyber incident may have affected data integrity;
• The facility has lost power.

Moving to another register may solve a hardware problem, while continuing sales during an unresolved sync failure may create a larger compliance issue.

Set Severity Levels

Use simple internal levels: minor, major, and critical. Critical incidents include data loss, unauthorized access, or mandatory stop-sale conditions.

Employees should never invent a workaround because the customer line is growing.

Assign Roles and Escalation Paths

Name one incident lead per shift and backup decision-makers. Assign responsibility for:

• Diagnosing affected systems;
• Pausing checkout or online ordering;
• Contacting POS, payment, network, or compliance support;
• Protecting temporary records;
• Approving reconciliation and reopening;
• Communicating with customers and staff.

Keep contacts and procedures outside the all-in-one dispensary platform.

Define Stop-Sale Conditions

Protect Compliance Controls

Each retailer must adapt the runbook to its licensing jurisdiction. Sales should stop when employees cannot confirm product eligibility, package quantity, customer qualifications, purchase limits, or required reporting.

Other stop conditions may include:

• Uncertainty about duplicate transactions;
• Inability to identify the correct package;
• A suspected security compromise;
• Loss of required verification tools;
• Conflicting inventory data.

Management should define stop conditions before an outage, not debate them during one.

Control Online Orders

Pause online menus when accurate availability cannot be confirmed. Retail POS for cannabis stores should show whether an order is accepted, reserved, paid, or reported. The runbook should address open carts, canceled pickups, and duplicate attempts.

Prepare an Approved Temporary Record

When local rules and the retailer’s SOP permit limited activity during a specific outage, use a controlled temporary record. Do not rely on loose notes, personal phones, or memory.

Capture:

• Date, time, store, and temporary number;
• Employee and approving manager;
• Customer type and verification steps;
• Product, quantity, price, and package identifier;
• Discounts, taxes, payment, and receipt details;
• Reason the system was unavailable;
• Final reconciliation status.

Use pre-numbered forms or a protected offline template. Every temporary record needs one owner and one final result: entered, canceled, or escalated.

Protect Queues and Prevent Duplicates

Do Not Repeatedly Resubmit

A timeout may occur after the receiving system accepts a request but before the POS receives confirmation. Repeated submissions can create duplicates.

A Metrc-integrated dispensary POS should show whether an event is saved, queued, accepted, rejected, or automatically retried. Metrc’s API documentation describes modification timestamps that integrations can use to identify changes.

Before retrying, check the receipt, package balance, submission time, error response, and automatic-retry history.

Restrict Manual Entry

Manual entries can create duplicates when queued transactions later synchronize. Limit direct portal entry to authorized employees and require an incident reference in the notes.

Do not delete evidence while correcting the outage. Preserve error messages, screenshots, queue history, and ticket numbers.

Restore Systems in a Controlled Sequence

Verify Before Reopening

When service returns, confirm the network, POS, payments, online ordering, inventory, and compliance integration separately.

Use this recovery sequence:

1. Record the restoration time;
2. Review pending, rejected, and retried transactions;
3. Match temporary records to completed records;
4. Identify missing or duplicate events;
5. Verify package quantities and payment totals;
6. Test one controlled transaction;
7. Obtain manager approval before reopening.

Recovery is complete only when the systems agree, not when the login screen reappears.

Reconcile Every Exception

Compare the dispensary inventory and POS system with physical stock, payments, online orders, and the applicable seed-to-sale cannabis software. Investigate differences before making adjustments.

For each correction, record what failed, which records were affected, how the result was verified, who approved the change, and what will prevent recurrence.

Build Resilience Before the Outage

The NIST Contingency Planning Guide recommends identifying critical processes, recovery priorities, preventive controls, contingency strategies, testing, training, and plan maintenance.

Useful safeguards include:

• Secondary internet connectivity;
• Backup power for network equipment;
• Spare scanners, printers, and terminals;
• Exportable inventory and transaction reports;
• Role-based access and multifactor authentication;
• Documented vendor escalation routes;
• Regular recovery tests.

Test and Maintain the Runbook

Run exercises for a terminal failure, internet outage, payment interruption, integration error, and suspected cyber incident.

After each test, review:

• How quickly the incident was classified;
• Whether stop-sale rules were followed;
• Whether temporary records were complete;
• Whether duplicates were prevented;
• Which instructions were unclear.

Update the runbook after migrations, regulatory changes, new integrations, or staffing changes. A plan that has never been tested is only a theory.

Final Takeaway

A useful downtime runbook gives cannabis retailers a repeatable path from detection to recovery. It classifies failures, assigns authority, defines stop conditions, protects temporary records, controls retries, and requires reconciliation before normal operations resume.

A compliant cannabis retail platform can support failover with alerts, queues, permissions, and audit logs. Resilience comes from combining those tools with trained employees and decisions made before the outage begins.