East African Bank Group Brings Service Assurance to Its ATM Estate With Ollandi

Summary

A bank group operating ATMs in several East African markets deployed Ollandi for ATM to replace country-by-country monitoring with a single view of estate service. Cash shortfalls are now forecast before the next cash-in-transit visit, vendor performance is measured from terminal evidence, and incidents close only when the terminal is serving customers again.

Before Ollandi

• Each country team monitored its own terminals. No one had a view of the full estate.
• Terminals regularly ran out of cash before the scheduled cash-in-transit (CIT) visit.
• Vendors closed tickets while terminals were still failing customers.
• A single carrier outage produced dozens of separate tickets.
• SLA disputes relied on the vendor's own ticket history.

After Ollandi

• One estate view across all markets, shared by every country team.
• Per-cassette cash forecasts against each terminal's next CIT visit, early enough to reschedule.
• Vendor response and restoration clocks tracked from terminal evidence, with breach and penalty exposure recorded per contract.
• Cases marked "provider declared fixed" remain open until Ollandi verifies the terminal is dispensing.
• Carrier and upstream failures grouped into one case and routed to the owning team.
• Declared maintenance windows suppress alarms for planned work.

Deployment

• Product: Ollandi for ATM, powered by the ATM Domain Intelligence Pack.
• Data sources: ATM monitoring, cash systems, vendor and SLA records, network status. All access is read-only.
• Operating control: Observe & Notify. Ollandi does not act on terminals or schedule CIT visits.
• Repairs and cash operations remain with the bank's teams and contracted vendors.

By the numbers

Measured across Glemad's production deployments and published in the ADT research paper:

• Mean time to detect: 0.8 minutes, against an industry average of 287 minutes.
• False positive rate: 1.2%, against an industry average of 23.5%.

These figures come from environments with mature telemetry and defined policy baselines. Results depend on the quality of the telemetry available.