Agents in the Trades
Dispatch

Marketing holds the board before it launches the campaign

It is 9:12 a.m. on a Monday and the marketing manager at a three-truck HVAC shop is staring at two things at once: a $79 tune-up campaign ready to send to 1,400 past customers, and a dispatch board already full through Thursday.

Source: ServiceTitan, "Pantheon 2026: Live coverage from ServiceTitan" — https://www.servicetitan.com/blog/pantheon-2026-live-coverage — the coverage describes agents working across demand generation, field work and the back office, including a demo of agents holding a slot for a more valuable job and refilling a cancellation gap in minutes. The workflow below is built from that idea, not presented as a shipped product claim.

The problem

Most campaigns are launched at the board, not with it. Marketing sends the offer, the phones light up a day or two later, and dispatch discovers there was never room for the work just sold. The shop then chooses between turning callers away, pushing regular customers out, or working the crew into unplanned overtime.

The failure is timing, not effort. By the time the response arrives, the slots it needed are gone — taken by lower-value work that booked first simply because it booked first.

Get one workflow like this, every day.

The free Agents in the Trades newsletter — each workflow broken down: what it does, what it costs, and where it breaks.

How the agent workflow works

  1. The campaign triggers a hold, not just a send. The moment a campaign is approved to launch, the agent places a temporary capacity hold on the dispatch board for the window when responses are expected — before a single email or text goes out.
  2. The hold is sized from that channel’s own history. Email, text and postcard offers do not respond at the same speed or rate. The agent uses the shop’s past response lag and booking rate for that channel and offer type, not a generic guess.
  3. Dispatch sees the board as it will look. Held slots are shown as held — who they are for, when they expire, and what happens if they go unfilled — so dispatch can object before launch, not after the phones ring.
  4. Campaign bookings fill the hold first. As responses come in, they book into the reserved capacity. Ordinary booking continues around it, and emergency work can still override the hold under rules the shop sets in advance.
  5. An unfilled hold releases itself. If bookings fall short by the deadline, the remaining held slots return to the open board automatically, while there is still time in the day to sell them to someone else.

What you need to set it up

Where it breaks

The forecast is wrong in both directions. Hold too little and the campaign still oversells the board. Hold too much and you have turned away real, paying work to protect slots for callers who never came. A new offer, a new list, or a heat wave in the middle of the window makes last time’s lag a poor guide.

A hold can quietly starve the customers you value most. If maintenance members, repeat customers or urgent calls find the board artificially full, the campaign has bought new work by taxing loyalty. The override rules are not a detail — they are the difference between a smart hold and a blocked door.

Marketing and dispatch want different things, and the agent will not settle that. Marketing is judged on response; dispatch is judged on a board that runs. If nobody owns the size of the hold, it will drift toward whoever complains loudest. A person still has to approve the first holds and review what each campaign actually used.

The bottom line

Run this if your campaigns regularly sell work the board cannot fit. Reserve the capacity before you create the demand, show dispatch the cost up front, and let the hold expire on its own when the response does not come.

Related use cases

← All use cases

One working agent workflow, every day.

Real AI agent use cases from solar, construction, and field operations — what they do, what they cost, and where they break. No hype, no theory.