How AC Hotel South San Francisco Put AI to Work on Every Shift

AC Hotel San Francisco Airport & Oyster Point Waterfront is a busy airport-market property: late arrivals, early departures, and a front desk that has to be on for all of it. The phone rang all day and all night with the same questions: what time is checkout, is there a shuttle, is the pool heated, how much is parking, what's the Wi-Fi. And every call pulled an agent away from the guest standing in front of them. At check-in peaks the team was paying for bodies at the desk just to keep up with the ringer. Add it up and the phone alone was eating roughly four hours of front desk time a day. The bigger cost was less visible. Like most branded hotels, the back office still ran on walkie-talkies, clipboards, and spreadsheets. Every morning a supervisor spent close to two hours hand-building the housekeeping board before the first room attendant could start. Room status lived in someone's head until it was radioed across. The labor schedule was a spreadsheet rebuilt each week from last week's. Requests got passed along in hallways and lost at shift change. Good people were spending their day coordinating work instead of doing it.

Lance went live first as the hotel's AI front desk. It answers every call in a natural voice, around the clock, in multiple languages, pulling answers from the property's own knowledge base, turning guest requests into tracked tasks that route straight to housekeeping or engineering, capturing group and event inquiries as leads, and handing the call to a person the moment it matters. No hold queue, no voicemail, no "let me find someone." Once the team was spending time inside the hotel, it was obvious the real bottleneck was the back office. So Lance rebuilt it. Tablet-based operations software replaced the radios and clipboards, with AI agents embedded directly in the staff workflow: the housekeeping board is generated in about two minutes and balanced against the day's arrivals, stayovers, and checkouts; labor is scheduled automatically from forecasted occupancy; tasks are routed, timed, and closed across departments; and automated checks run across the PMS and operations systems so room status is visible to everyone at once: the front desk, housekeeping, engineering, and the GM, without anyone keying a radio.

Today Lance answers more than 3,000 guest calls a month at the property, taking roughly four hours of phone work off the desk every day. Front desk agents spend that time with the guests in the lobby. The housekeeping board that took two hours now takes two minutes, and housekeeping starts the day on time. As operations manager, supervisor, and inspector roles turned over, the hotel did not backfill them; the coordination work those roles carried is now handled by Lance agents, and the rest of the back-of-house team runs about 10% more efficiently with dramatically less manual coordination. What started as an AI receptionist became AI agents running the operation, and it's the template OTO Development is now rolling out across its portfolio.