Illustrative Scenario — Revenue Growth
Predictive Maintenance Upselling for HVAC
How a $15M HVAC contractor used AI to identify and sell preventive maintenance agreements from reactive service data — adding $2.4M in recurring revenue.
Company Profile
Sector
Commercial HVAC services
Services
Installation, reactive repair, some maintenance
Revenue
$15M
Employees
78 (incl. 55 technicians)
Ownership
Founder-led, 12 years in operation
Exit Timeline
Planning sale within 18 months
A commercial HVAC business that had built its reputation on responsive reactive repair. The company was skilled, well-staffed, and profitable — but its revenue mix was heavily weighted toward one-off repair jobs. Only 18% of revenue came from recurring maintenance contracts. This made earnings volatile and the business less attractive to buyers, who strongly prefer predictable, recurring revenue streams.
The Revenue Problem
The business was leaving its most valuable asset — its existing customer relationships — almost entirely unmonetised beyond the initial reactive call.
Low recurring revenue
Only 18% of revenue ($2.7M) came from maintenance contracts. The remaining 82% was reactive, one-off work — highly variable quarter to quarter. Buyers typically apply a significant discount to non-recurring revenue when calculating valuation.
No upsell process after reactive work
Technicians completed the repair and moved on. There was no systematic process to convert a reactive customer into a maintenance agreement — despite the fact that every reactive call represented a customer with an aging system and no preventive plan.
No equipment lifecycle data
The business had years of service history — make, model, age, repair frequency, failure types — but it sat in disconnected job records. Nobody was analysing this data to identify which customers were most likely to need (and buy) a maintenance agreement.
Reactive pricing only
Emergency and after-hours work was priced the same as scheduled work, despite costing significantly more to deliver. There was no dynamic pricing based on urgency, time of day, or customer value.
Technicians not equipped to sell
Technicians were skilled tradespeople, not salespeople. They had no tools, prompts, or incentives to discuss maintenance agreements during service calls. The commercial opportunity at the point of service was completely missed.
What We Implemented
Over 12 months, we deployed AI systems designed to convert the business's existing customer base and service data into a recurring revenue engine.
Predictive maintenance opportunity engine
Analysed 4 years of service history — equipment age, failure frequency, repair costs, and system types — to identify customers with the highest propensity to purchase a maintenance agreement. The AI generated a ranked list of 340 target customers with predicted contract values, prioritised by likelihood of conversion.
Technician upsell prompts on mobile devices
Built an AI system that displayed tailored maintenance agreement recommendations on the technician's mobile device at the point of service. After completing a reactive repair, the technician could show the customer a personalised proposal based on their equipment, history, and risk profile. The prompt included a simple "interested / not now / not interested" response that fed back into the system.
Automated agreement generation and follow-up
When a customer expressed interest, the system automatically generated a tailored maintenance agreement and sent it within 2 hours. AI-driven follow-up sequences pursued warm leads with 3 touchpoints over 14 days. Previously, proposals took 3–5 days to produce and had no follow-up process.
Dynamic pricing for reactive work
Implemented AI-driven pricing that adjusted reactive repair rates based on urgency, time of day, customer history, and equipment complexity. After-hours and emergency premiums were applied transparently. Average reactive job value increased by 18% without any reduction in call volume.
Recurring revenue dashboard
Built automated reporting that tracked the maintenance contract book — new contracts signed, renewal rates, churn, average contract value, and projected annual recurring revenue. This data became a centrepiece of the buyer presentation, demonstrating a growing, predictable revenue base.
The Outcome
The shift toward recurring revenue fundamentally changed the business's profile — from a reactive repair shop to a managed services platform with predictable, growing earnings.
| Metric | Before | After | Change |
|---|---|---|---|
| Revenue | $15.0M | $18.1M | +20.7% |
| Recurring Revenue | $2.7M (18%) | $5.1M (28%) | +89% |
| EBITDA | $1.7M | $2.9M | +71% |
| EBITDA Margin | 11.3% | 16.0% | +4.7pp |
| Valuation Multiple | 3.4x (est.) | 5.3x (achieved) | +1.9x |
| Enterprise Value | $5.8M (est.) | $15.4M (achieved) | +166% |
| Avg. Reactive Job Value | $1,240 | $1,460 | +18% |
In this scenario, the growth in recurring revenue was the single most influential factor in the valuation. The business sold at a 5.3x EBITDA multiple — a significant premium over the pre-engagement estimate, driven by the quality and predictability of the revenue base.
This is an illustrative scenario based on a composite of common patterns in HVAC services. It does not represent a specific client engagement, and the figures shown are not guaranteed outcomes. Every business is different.
Is your business facing similar challenges?
We begin with a confidential discussion to assess where AI can have the greatest impact on your EBITDA and exit value.