Hashlogics
Answers

How should a hotel handle the 4pm check-in wave?

The answer isn't a faster agent or a better hold message. It's deciding which of the three jobs at 4pm a person has to do, and moving the other two.

Answered in short

4 things that decide this

  1. 01Take the phone and the guest messages off the desk during the wave. One agent can serve the guest in front of them or the line ringing beside them, never both. So route overflow calls and requests to a system that quotes from the PMS, books and texts confirmations, and keep your person for the people standing there.
  2. 02July runs 16.1 calls per room, and answered lead calls convert at 44 to 46% in summer (Revinate 2026). Operators describe the same failure in plainer terms: the hold button, then a caller who waits about thirty seconds, hangs up, and rings the next hotel on the list.
  3. 03A towel request should become a task with an owner and an ETA, not a note your agent carries in their head until the queue clears. That's the second job leaving the desk, and it's the one that quietly turns into a bad review.
  4. 04Safety never moves. Fire, medical, security and any call about a guest's wellbeing goes to a person before the caller finishes the sentence, whatever else is happening at the desk.
Why the obvious answer doesn't work

Staffing the wave is the answer everyone tries first

Every hotel that has this problem has already tried the sensible fix: put a second person on the desk between three and six. It works, right up until somebody calls in sick, a group arrives early, or the wave lands on a Tuesday nobody rostered for. And front-desk shortages are common enough that the second person often isn't there to roster in the first place, with 26% of hotels reporting them (AHLA/Hireology 2025).

A deeper problem sits underneath it: three jobs collide at one moment, and only one of them needs the person standing at the desk. Checking a guest in is face to face work. Quoting availability for next weekend isn't. Logging a towel request isn't either. When all three land on one agent, the two that could go elsewhere are the ones dropped. The guest in front of them can see them.

So the useful question isn't how to move faster through the queue. It's which of these jobs has to be done by the person at the desk at 4:10pm, and what happens to the rest of them.

What actually happens to the call

The hold button is where the booking goes

Ask operators how they handle the phone during the wave and you get the same answer, in almost the same words: greet the caller, say please hold, press the button. Some have stopped picking up at all during the busiest hour. Their reasoning: a caller who waits thirty seconds and hangs up will ring again, and a queue of held lines costs more attention than a ringing phone.

Both of those are rational responses to an impossible minute. Neither is free. A caller checking availability for a weekend is a booking with real money attached, and the summer numbers say those calls convert well when somebody answers them (Revinate 2026). Your competitor two streets away is the next number in the search result.

What makes this hard to see is that nothing records it. A missed call at 4:10 leaves no trace in the PMS, no line in the revenue report and no complaint. You find out about the guest who booked elsewhere never, which is why the wave keeps getting solved with a hold button rather than with a decision.

The 4pm wave, split three waysLive
  1. The guest at the deskStays with your agent, face to face, undisturbed.
  2. The reservations callAnswered, quoted from live availability, booked, confirmed by text.
  3. The towel requestBecomes a housekeeping task with an owner and an ETA to the room.
  4. The safety callReaches a person before the caller finishes the sentence. Always.

One agent can carry the first of these well. The system carries the middle two, and the last one is never negotiable.

The mechanism

What a working setup does between three and six

Your overflow call gets answered on the second ring rather than held. It reads live availability out of your PMS, quotes by your own rate rules, takes the booking, and texts the confirmation while the caller is still on the line. It handles parking, late arrival and the "is it cheaper on Booking.com?" question the way you tell it to. Those three are most of the wave's call volume.

A guest message becomes a task rather than a memory. A text asking for towels lands in your housekeeping module or task app with a room number, an owner and a due time. Your guest gets an ETA straight back. If that ETA passes with the task still open, your duty manager hears about it before the guest does.

And room-ready fires from room status instead of from a walk. When housekeeping marks 214 clean, the guest waiting in your lobby gets told, without your agent breaking off a check-in to go and look. That's the third thing leaving the desk, and it's usually the one that surprises operators most, because it never felt like a phone problem.

  • 01Answer the overflow, don't hold it: quote from the PMS, book, and text the confirmation.
  • 02Turn requests into tasks with an owner and an ETA, so nothing depends on the desk being free.
  • 03Fire room-ready from room status, not from somebody walking the floor.
Where a person stays

The four things that never leave a human, whatever the queue looks like

Automating the wave only works if the boundary is written down before anything is built, and tested first rather than last. That list is short, and it doesn't bend for how busy the lobby is.

We write it as a one-page map at the start of every build: what the system answers, what it only proposes, and where a person gets woken. If a vendor can't show you that map, the boundary doesn't exist yet.

  • 01Fire, medical, security and any call about a guest's safety reach a person immediately.
  • 02Rate parity, overbooking walks and chargebacks stay with your duty manager. The system prepares the numbers and never sends a rate change on its own.
  • 03Review replies are drafted for approval and never posted automatically, because your name goes on them.
  • 04Card details never enter a call transcript or a message log. Payment stays inside your PMS or gateway.
Questions, answered
01Isn't hiring a second desk agent for the wave simpler?+

It's simpler, and it's often the right first move. It just doesn't survive a normal week. Somebody calls in sick, a group arrives early, or the wave lands on a day you didn't roster for, and you're back to one agent and three jobs. Front-desk shortages are common enough that the second person is frequently not available to hire at all, with 26% of hotels reporting them (AHLA/Hireology 2025). A layer around the desk keeps working on the bad days, which are the days it matters.

02Will guests know they're not talking to my staff?+

On the overflow call, most people ask about availability, parking or a late arrival. They want a fast answer more than a conversation. Meanwhile the people standing in your lobby get your agent's full attention, which is the point of moving the phone off the desk. You decide the voice, the rules and what it will and won't say, and anything with heat behind it goes to a person.

03What about the guest who's been on hold and is already annoyed?+

That call should reach your duty manager, not a system. Complaint calls, guests who sound upset and anything about a previous stay belong on a person's list from the start. You write that trigger list before anything gets built. Wave automation exists to stop that call happening, not to handle it once it has.

04Does this need us to change PMS?+

No, and we'd usually argue against it. What matters is what your current account exposes, not which platform you're on. So the first thing to establish is whether your PMS lets a system read availability and write a reservation, and to what depth. Where the connection stops, a reviewed queue your team accepts with one click still beats re-keying at night audit.

05How long does the phone log reading take?+

It's a reading exercise rather than a build, so it happens well before anything is scoped. You bring a month of calls by hour, plus your arrivals for the same period. Out of it comes a one-page map of where the wave leaks, per shift. That note is yours whether or not you build anything, which is what makes it worth doing first.

By Abdul Basit, CEO, HashlogicsUpdated
Start

Let’s deploy working AI into your business.

We build AI agents and automation, ship them into the tools you already run, then stay on under an agreed service level. A senior engineer reads every brief, and your call gets scheduled within 24 hours.

What happens next

  1. 01

    You send a brief or book a call

    Two minutes, whichever you prefer.

  2. 02

    A senior engineer replies within 24 hours

    Not a sales rep.

  3. 03

    Honest scoping, in writing

    And if we’re not the right fit, we say so.

Abdul Basit, CEO of Hashlogics

“I started Hashlogics because too many teams ship a demo, get paid, and disappear. We build to a standard we’d run ourselves — and we stay to keep it running.”

Abdul Basit · CEO · a direct line

Not ready to talk? Take the checklist.

12 questions to ask any AI agency before you sign. They separate a demo shop from a team that ships to production.

Get the checklist

Free · no newsletter