We delivered an AI revenue and operations manager for a woman running three boutique properties in downtown Charleston — one that dynamically prices rooms against event calendars, occupancy curves, and comp-set data, manages the seasonal staffing math across all three buildings, and gives her one P&L view across properties that her accountant and her GM can both read

She bought the first property — a twelve-room inn on Church Street — because she fell in love with the building and thought she understood hospitality from fifteen years on the restaurant side. She was right about the hospitality and wrong about everything else. The second property came two years later, a sixteen-room house on Wentworth, and the third — twenty rooms in a converted warehouse on upper King — opened last spring. Now she runs forty-eight rooms across three buildings with three different guest profiles, three different operating rhythms, and a pricing strategy that until recently was her looking at the calendar and her gut, adjusted by whatever she remembered the Belmond and the Spectator were charging when she checked last Tuesday. The problem was not that she was bad at pricing. She has excellent instincts. The problem was that pricing forty-eight rooms across three properties requires a kind of constant, granular attention that no human can sustain alongside everything else an owner does. Spoleto fills all three properties, but the Church Street inn sells out two weeks before King Street does because the location premium is real but not reflected in the rate differential. Restaurant Week moves her King Street occupancy twelve points but barely touches Wentworth. Cruise ship days create same-day demand from walk-ups that she was never capturing because her rates were set weekly, not daily. Meanwhile the seasonal staffing calculus ran on a spreadsheet her GM kept in a personal Google Drive folder. In peak season she needed thirty-one housekeepers across the three buildings. In January she needed fourteen. The ramp-up and ramp-down happened differently at each property because each one has a different seasonal curve, and the cross-training that lets a housekeeper work at any of the three buildings required tracking certifications, building-specific procedures, and individual schedule preferences that the spreadsheet could not hold. We built her a revenue manager that sees all forty-eight rooms as one inventory and prices them as a portfolio. It pulls the event calendar for Charleston — not just the headline festivals but the smaller drivers: garden tours, gallery walks, college move-in weekends, the military events that fill specific neighborhoods — and layers that against her historical occupancy curves, the rates her comp set is posting on OTAs right now, and her own booking pace for the next ninety days. Rates adjust nightly and by room type, with floors and ceilings she sets so nothing prices below her cost or above what the brand can credibly charge. When Spoleto announces dates, the system reprices the entire portfolio within an hour, not after she happens to notice. The staffing module knows what occupancy means in labor hours by property. A sold-out night at Church Street is twelve rooms and takes four housekeepers the next morning. A sold-out night at King Street is twenty rooms with larger suites and takes seven. It builds the weekly schedule across all three buildings, respects the cross-training matrix so it knows who can work where, accounts for the overtime rules that South Carolina does not require but that she enforces anyway because she wants to keep her people, and flags the weeks where projected occupancy means she needs to bring back seasonal staff — giving her three weeks of notice instead of the three days she used to get. Guest communication follows a sequence she designed once and now runs automatically: confirmation with neighborhood guide, pre-arrival message with parking instructions and restaurant recommendations tailored to the property, a mid-stay check-in that routes to the front desk if anything needs attention, and a post-stay message that asks for a review and offers a return-stay rate. The tone matches her brand — warm, specific, never corporate — and each property's messages reference that building's character because a guest at the Church Street inn and a guest at the King Street warehouse are having two different experiences. The P&L rolls up weekly across all three properties with the breakdowns her accountant needs — revenue by room type, ADR, RevPAR, labor cost per occupied room, amenity cost per guest — and the summary her GM needs, which is simpler: are we making money, where are we soft, and what do we do about it. "I bought three buildings," she told us. "I didn't realize I was also supposed to become a data analyst, a scheduling coordinator, and a pricing strategist. Now I get to just be the person who makes sure the guest has a beautiful stay."