An AI-first redesign of facility management operations.

The challenge

The client supports a major UK hotel group with housekeeping, daily turnaround, and room readiness. The business is operationally complex: 900+ frontline staff, ~80 sites, and a quality bar set by the hotel brand it supports. The operating model was paper-based.

• 15–20 hotels routinely failing to close housekeeping by 7pm before the engagement.
• 45 minutes of typical daily untracked time per worker between sign-in and productive work.
• Up to 20% of daily revenue impact at some sites from untracked Do Not Disturbs and unclear room status.
• 7 to 9 sites covered by each operations manager. Visibility was reactive at best.
• No data foundation. No baseline to improve against. No path to AI-enabled scheduling or forecasting.

Our approach

We took the engagement through the AI Factory's five-stage flow. Ingestion agents pulled every piece of operational data the business already had: cleaning roster spreadsheets, payroll records, hotel booking signals, exception logs. Analysis agents built the baseline.

From there, senior consultants ran the design. The right answer was not just a digital cleaner-roster app. It was a real-time mobile platform that captured exception data at source, fed it back into AI-enhanced operations dashboards, and surfaced the operational signals that managers had been working without.

The build ran on the AI Factory operating model. Agentic engineering on the heavy lift, senior engineers on architecture, every release reviewed and signed off by senior humans before it touched production sites.

What we delivered

A real-time, AI-enabled workforce management platform deployed across the in-scope sites, with a structured data foundation underneath it.

• Mobile workforce platform deployed to 900+ housekeeping staff across ~80 hotels.
• Live operations dashboard giving each operations manager visibility across all 7–9 of their sites.
• Exception capture at source: untracked Do Not Disturbs, green stays, and room status changes feed back into the same data pipeline.
• Structured data foundation for the next stage: AI-enhanced scheduling, predictive resourcing, dynamic exception routing.
• Benefits realisation tracked monthly against operational KPIs and the original case.