Service-board recovery OS for regional HVAC, plumbing, and electrical consolidators that keeps trucks full when schedules break.

1. A single shared context thread now spans intake, data cleaning, messaging, and scheduling, so exception recovery can be automated instead of stitched together manually.
2. Operators already report 100-to-1 technician-to-dispatcher leverage, which means the remaining operational pain has shifted from basic booking to handling the edge cases that still break the day.
3. Category adoption has moved beyond pilot shops into hundreds of independents and PE-backed platforms, creating a large installed base ready for a specialized recovery layer.
4. Investors and operators are signaling that operational execution, not just lead generation, is the most valuable white space in home-service AI.

Catalyst. AI dispatch platforms have already proven 100-to-1 dispatcher leverage and adoption across hundreds of shops, so the next urgent bottleneck for multi-location operators is day-of-service recovery when the board breaks.

The product sits on top of the operator's existing field-service stack and ingests open jobs, technician skill tags, travel times, customer message threads, and branch rules. When a job runs long, a technician no-shows, or an emergency call arrives, it re-optimizes the day in real time and ranks the best moves by recovered revenue, SLA risk, and customer impact. Dispatchers can approve a plan with one click, or let low-risk moves auto-execute with prewritten customer updates and internal handoffs. Over time, the system learns branch-specific duration patterns and technician fit, turning tribal dispatch knowledge into reusable operating playbooks.

What's different. Most home-service AI products chase top-of-funnel lead conversion or promise a full rip-and-replace dispatch suite. This company starts with the narrowest, highest-value workflow that still breaks manually: same-day exception recovery after the schedule is already built. Because it learns from branch-specific outcomes across callouts, overruns, and emergency inserts, it can become the system of action for daily dispatch without requiring operators to replace their incumbent field-service software on day one.

Jobs to be done

flowchart LR
  Ops[Central dispatch team] --> Breakage[Overruns, callouts, and urgent jobs]
  Breakage --> RecoveryAI[Exception recovery AI]
  RecoveryAI --> Rebook[Reassigned jobs and customer updates]
  Rebook --> Outcome[Higher utilization and fewer missed appointments]

Executive takeaways

• This market is real, but the strongest signal is not generic AI for the trades; it is AI that helps centralized teams recover the board when schedules break. Probook already shows 95% of dispatching decisions automated at Peterman Brothers and a 22-to-10 dispatcher consolidation at Del-Air, category coverage highlights 100:1 technician-to-dispatcher ratios in the market, and ServiceTitan is pushing the same control layer with Dispatch Pro and Contact Center Pro [1][6][9][32][35].
• The best initial buyer is the multi-branch operator, not the single-shop contractor. KPMG, West Monroe, and the active HVAC/plumbing roll-up trackers all point to continued consolidation and standardized operating playbooks, which means dispatcher efficiency and cross-branch service levels now have executive budget owners [17][18][21][22].
• Willingness to pay is credible only if the product proves recovered revenue or higher dispatcher leverage. ServiceTitan sells per-technician packages, Housecall Pro starts at $59/month, Workiz monetizes seat economics and AI add-ons, and FieldEdge sells quote-based tiers, so the question is incremental lift, not whether software budget exists [30][41][49][59].
• Competitive intensity is high. Probook is the closest emerging direct threat, ServiceTitan already markets AI-assisted dispatch, and Housecall Pro, Workiz, and FieldEdge own adjacent workflows. The wedge only works if it is faster to deploy, vendor-neutral, and materially better at day-of exception handling than a general suite [2][12][32][41][46][49][60].
• Adoption friction will come from trust and compliance more than raw model capability: dispatchers need explainability, customer communications need TCPA and opt-out controls, and overtime or certification constraints mean not every “optimal” assignment is actually executable [37][46][72][73][74][77].

Market definition

Defined market: an AI-assisted exception-recovery and dispatch-orchestration layer for residential and light-commercial HVAC, plumbing, and electrical operators. It sits between booking/contact-center tooling and field execution, using live technician, route, and customer-communication data to reassign work when the day breaks. It excludes full FSM replacement suites, generic answering services, and commercial-only project platforms [2][4][14][15][32][35][60][68].

Customer and buyer

Primary customer is the centralized dispatch lead or service-operations group inside 8-20 branch operators. The economic buyer is usually the COO, VP Service Operations, or platform operations executive because same-day recovery affects bookings, utilization, call-center performance, and post-acquisition standardization. Daily users are dispatchers and call-center managers; expansion sponsors are often portfolio-ops leaders inside regional roll-ups [6][9][17][18][21][22][37].

Buying triggers

• Branch centralization or recent tuck-ins create more exceptions than local boards can absorb cleanly. [6][9][17][18]
• Dispatchers become swamped by ETA updates, reschedules, and emergency inserts once call-center and dispatch workflows are consolidated. [37][38][46]
• Teams already piloting AI voice or overflow coverage want to capture same-day revenue after the call is answered, not just avoid voicemail. [35][48][54][58]

Willingness to pay

Budget exists, but it is benchmarked against adjacent software rather than a pure-play exception-recovery line item. ServiceTitan prices per technician, Housecall Pro starts at $59/month, Workiz monetizes seats plus AI/communications add-ons, and FieldEdge sells tiered quote-based packages. That means a new layer must prove recovered revenue, higher ticket quality, or more technicians per dispatcher to earn spend [30][41][49][59]. [30][41][49][59]

Category dynamics

Growth signal 15.6% CAGR (2025-2032)

Validation signals

• Peterman Brothers already centralized dispatch for 200 technicians across 11 markets with about 95% of dispatching decisions made by Probook.
• Del-Air centralized dispatch from 22 dispatchers to 10 across eight locations, showing that large operators will reorganize around software once trust is established.
• ServiceTitan and Workiz both publish customer evidence that AI-supported booking and dispatch tools move bookings or revenue, confirming buyer appetite for automation on the board edge.

Regulatory & technical constraints

• HVAC dispatching cannot treat all capacity as interchangeable because refrigerant-related work requires EPA Section 608-compliant technicians.
• Automated calls and texts require consent-aware workflow design under U.S. telephone restrictions.
• Aggressive same-day recovery can change labor economics because overtime rules still matter once tech hours stretch beyond normal shifts.
• Integration and outbound orchestration need opt-out and systems-governance controls, not just better ranking models.

The crowded edge of this market already includes full suites (ServiceTitan and FieldEdge), SMB and mid-market generalists (Housecall Pro and Workiz), and AI-first full-stack challengers (Probook). BuildOps shows that AI dispatch can be productized in adjacent field-service settings. The white space is not “dispatch software” in general; it is a cross-stack, same-day recovery layer that improves dispatcher leverage without demanding a full front-office replacement [2][12][30][32][35][41][46][49][53][59][60][68].

Why incumbents do not win by default

• Enterprise trade suites. ServiceTitan and FieldEdge already own the dispatch board and adjacent customer workflows, but their advantage weakens when buyers run mixed stacks or want a vendor-neutral recovery overlay rather than a deeper suite commitment.
• SMB and mid-market generalist suites. Housecall Pro and Workiz simplify dispatch, messaging, and AI answering, yet their positioning is about running the whole shop rather than governing same-day exceptions across 8-20 branches.
• AI-first vertical OS vendors. Probook proves that operators will trust AI at the board, but it bundles broader front-office automation, leaving room for a narrower overlay where buyers want recovery gains without changing more of the stack than necessary.
• Commercial-adjacent dispatch platforms. BuildOps shows AI dispatch can be operationalized, but it targets commercial contractors rather than the residential multi-branch service board that defines this wedge.
• Manual dispatch plus branch tribal knowledge. The default substitute is still a skilled dispatcher juggling calls, ETA texts, and reassignment decisions manually, which scales poorly once operators centralize across brands or markets.

Service-board recovery OS targets regional HVAC, plumbing, and electrical consolidators whose centralized dispatch teams lose same-day capacity when installs overrun, technicians call out, or urgent jobs land midday. The beachhead is 8-20 branch operators in the Southeast and Sun Belt with 80-300 technicians and both service and install calendars, because they have enough density and exception volume to prove ROI quickly. The initial product is an overlay on incumbent field-service software, not a rip-and-replace suite: it watches the live board, ranks recovery moves, and sends compliant customer updates after dispatcher approval. The GTM system starts with founder-led sales to a VP Service Operations or COO during peak season, branch centralization, or post-roll-up standardization, using a paid pilot that converts to per-technician annual software once recovered appointments and dispatcher leverage are proven. Research supports the pain, buyer readiness, and budget benchmarks, but it does not yet prove the key assumption that large operators will buy a vendor-neutral recovery layer instead of waiting for ServiceTitan, Probook, or other incumbents to extend native modules. The first proof point is straightforward: show on 60-90 days of board data that the product recovers same-day revenue and earns dispatcher trust faster than manual reshuffling. The plan deliberately avoids full dispatch replacement, single-shop SMBs, and adjacent trades until one centralized dispatch hub can deploy in weeks, convert to production, and expand across brands. If the company cannot show reliable write-back, more than 50% pilot-to-production conversion, and measurable recovered jobs without custom-services sprawl, the venture case weakens materially.

Problem

• Centralized dispatch teams at 8-20 branch HVAC, plumbing, and electrical operators lose same-day revenue when installs overrun, technicians call out, or emergency jobs land and incumbent boards do not recover the schedule automatically.
• Operators still solve exceptions with spreadsheets, phone trees, and branch tribal knowledge, which drives missed appointments, inconsistent customer communication, and pressure to add dispatcher headcount.

Solution

• Overlay the incumbent field-service stack, ingest live jobs, technician skills, routes, customer threads, and branch rules, and rank recovery moves by recovered revenue, SLA risk, and customer impact.
• Start with dispatcher-approved recommendations and templated customer updates, then auto-execute low-risk reassignments only after each branch's overrides and compliance rules are learned.

Why we win

• We attack a narrow, budget-linked trigger—same-day board recovery—so buyers can pilot without ripping out their system of record or funding a full dispatch replacement.
• Branch-specific override data, technician-fit learning, communication-response history, and cross-stack write-back reliability create a workflow moat that generic routing or contact-center AI tools do not.

Milestones

flowchart LR
  Wedge[Centralized dispatch recovery wedge] --> MVP[Assist-mode recovery overlay]
  MVP --> Proof[Recovered appointments and dispatcher trust]
  Proof --> Expansion[Multi-branch rollout plus broader orchestration]

Founding team

Experiment roadmap

Risk assessment

What must be true

• Multi-branch home-service operators experience enough daily board breakage that recovered same-day capacity is a funded KPI, not just a dispatcher annoyance.
• Buyers running ServiceTitan, FieldEdge, or mixed stacks will still trial a vendor-neutral overlay instead of waiting for native modules.
• One dispatch hub can deploy in under 30 days with limited data cleanup and show measurable recovered appointments within a single peak season.
• Dispatcher acceptance and override learning improve quickly enough to support low-risk automation after assist mode.
• Early accounts expand from one hub to multi-branch rollouts before suites or AI-first incumbents compress pricing.

Open diligence questions

• What percentage of daily overruns, callouts, and emergency inserts create recoverable same-day revenue rather than unavoidable deferrals?
• How often do large ServiceTitan or Probook customers reject third-party write-back into the live board?
• Which KPI actually releases budget first: recovered appointments, technician-to-dispatcher ratio, or ticket quality?
• What minimum data cleanliness for technician skills, status codes, and durations is needed to beat senior dispatchers?
• Can the first pilot deploy during peak season in under 30 days without turning into a custom integration project?

Model sanity

• Revenue engine. Base-case revenue comes from turning 3 Y1 paid pilots into 12 paying operators by Q4Y2 and 24 by Q4Y3 at roughly $120K steady-state annual ARPU.
• Must go right. Deployments must stay productized enough that solutions work does not drag gross margin below 70% while pilot-to-production conversion stays above 60%.
• Model breaks if. If pricing compresses toward $55 per tech per month or expansion slips by a quarter, the downside case runs out of cash before Y3 ends.
• Next-round proof. The seed case is strongest once two incumbent FSM environments support 10-12 production operators and first-hub-to-second-hub rollouts are repeatable.

Scenarios

Sensitivity

flowchart LR
  Leads --> PaidPilots
  PaidPilots --> ProductionOperators
  ProductionOperators --> TechniciansCovered
  TechniciansCovered --> Revenue
  Revenue --> GrossProfit
  GrossProfit --> Cash

Flags: The base case still depends on operators choosing a vendor-neutral overlay before Probook, ServiceTitan, or other incumbents close the feature gap. · Gross margin does not fully clear the plan target until late Y2, so a services-heavy onboarding motion would increase the funding need. · Reaching 24 paying operators by Q4Y3 with only two GTM hires requires partner channels and land-and-expand execution to work on schedule. · Customer concentration remains meaningful through Y2 because a delayed enterprise rollout can move quarterly revenue materially when only 10-12 operators are live.

• Dispatcher trust gap. Dispatch leaders may resist letting software move jobs or message customers during high-stress moments. Mitigation: Start human-in-the-loop with explainable recommendations, approval thresholds, and proof on recovered appointments before automating low-risk moves.
• Incumbent suite response. Existing field-service platforms or AI dispatch vendors could add exception workflows once the category proves valuable. Mitigation: Launch as an overlay that works across incumbent stacks, go deep on cross-branch recovery logic, and build a proprietary dataset on schedule-break outcomes.
• Messy branch data. Multi-location operators often have inconsistent status codes, technician tags, and job-duration data, which can degrade recommendations. Mitigation: Use onboarding templates, AI data cleaning, and a narrow first workflow that relies on a small set of high-signal inputs before expanding automation.