What if a General Manager could start Monday morning with a thoughtful conversation instead of a stack of reports?

It is Monday morning.

The General Manager is on the way from home to the hotel. There are already dozens of things waiting: last week’s performance, this month’s forecast, staffing questions, guest issues, revenue decisions, maintenance concerns, department handovers and the meetings that somehow always multiply.

Instead of opening ten different systems, the GM simply talks to Zoya.

“Zoya, how did we perform over the last 14 days, what is changing, and what should I be concerned about for the next two weeks?”

Zoya gets to work.

She looks across the hotel’s reality, PMS, reservations, room inventory, rate plans, cancellations, OTA production, forecast data and recent guest feedback. She compares the last 14 days with the same period last year and with the current month’s forecast. Read more about hotel data architecture.

But the important part is not simply collecting the numbers.

Zoya tries to understand what is actually happening.

Is the problem occupancy? Pricing? Channel mix? Cancellations? Market demand?

She identifies the three most significant causes, explains the evidence behind them, and then builds a 14-day recovery plan. See the AI readiness checklist for hotels.

Perhaps one option is to adjust rates and restrictions.

Another may focus on changing channel mix and promotions.

A third may protect inventory differently because the hotel’s current demand pattern does not justify aggressive discounting.

Each option comes with evidence and practical actions for Revenue, Reservations, Sales and Front Office.

And the GM does not have to accept Zoya’s recommendation.

The GM can say:

“I don’t think option one fits our market. We have a large corporate group arriving next week. Rework the plan around that.”

Zoya continues the conversation, incorporates that hotel-specific context and refines the options.

That is the part I find particularly important.

Hotel data is messy. That’s not a flaw.

Hotels live with data silos and several versions of reality.

The PMS says one thing. Reservations knows another. Revenue sees a different pattern. Sales knows about a group that has not fully appeared in the numbers. Front Office knows what guests are actually saying. Housekeeping knows what is happening on the floors.

This is not necessarily a technology failure.

It is hotel life.

A long-horizon task exists precisely because someone, whether a GM, Revenue Manager, Housekeeping Manager or F&B Manager, has to go several places, reconcile what is available, understand the context and decide what to do next.

Zoya is being designed to work in that reality, rather than pretending every hotel already has perfectly integrated data.

If Zoya cannot confidently determine the right direction because the data is ambiguous, it should say so.

It can present two or three possible paths, perhaps with one recommended path based on the best expected outcome or the lowest risk.

And if Zoya still cannot decide?

The GM decides.

The human remains accountable.

Don’t wait for perfect systems.

This is also why I don’t believe a hotel needs to clean up every underlying process before it can benefit from AI.

Let some of the processes and data remain messy.

Start with a focused pilot.

Give Zoya a clearly defined set of long-horizon tasks. Establish a fixed budget. Calculate expected ROI and total cost of ownership. Measure what actually happens. Explore hotel AI assessment and pricing.

Then evolve.

If Zoya can eventually take 40% of a GM’s recurring analytical and coordination workload, that does not mean replacing the GM.

It means giving the GM more time to walk the hotel, talk to guests, support the team, solve the difficult problems and simply be present.

And the same principle can apply across the hotel.

Imagine a concierge talking to Zoya about a difficult guest request.

A Front Office Manager discussing a service recovery.

A Housekeeping Manager working through room readiness and staffing.

An F&B Manager reviewing today’s operational situation.

A department head asking about an SOP, handover or shift change.

A team member dealing with an AC repair, room cleaning issue or guest complaint.

Instead of searching through documents, calling three people and piecing together information, they can simply talk to Zoya.

Zoya analyses the hotel context and comes back with two or three practical recommendations.

The staff member chooses a path.

Or defines their own.

Or simply talks to Zoya a little more until the situation becomes clearer.

The goal isn’t another hotel software tool.

The more I work with hotels, the more convinced I become that hotels already have great people.

The problem is often the coordination burden placed on those people.

SOPs, handovers, service recovery, staffing, sales, revenue, guest context and operational information become much more valuable when people can work from a shared understanding of what is happening.

That, to me, is the right operating lens for hospitality AI.

AI should not replace the human part of hospitality that guests remember.

It should remove some of the work that prevents hotel people from being present with those guests.

That is what we are trying to build with Zoya.

A tireless buddy for hotel teams.

Not another dashboard.

Not another system that expects the hotel to change how it operates.

A conversational operating agent that can take a complicated task, go through the hotel’s context, reconcile what it finds, think through the possible paths and help the person responsible decide what to do next.

We believe Zoya can eventually handle around 40% of recurring GM tasks, giving GMs more time for the hotel, their teams and their guests.

And if we can do that for a GM, why stop there?

What do you think?

If you work in a hotel, as a GM, department head, supervisor or frontline team member, I’d genuinely like to hear your opinion.

Would an AI buddy like Zoya actually make your Monday morning easier? What would you trust it to handle, and what would you always want to decide yourself?

Please write your thoughts and tell me what you think about this approach to AI in hospitality. Write your opinion to me