Not all inbound HVAC leads are worth the same effort. For an HVAC contractor running 15 to 20 trucks and fielding more than 20 calls a day, that difference is measured in real money. A homeowner whose air conditioning has been out for six hours in July, who found your number through a neighbour’s referral, and who is calling at 10am on a weekday is an extremely high-likelihood lead. An anonymous web form submission with no system details, submitted at 11pm, about “getting a quote on HVAC at some point” is a low-likelihood lead. Without a system that tells your team which is which, the emergency caller waits in the same queue as the quote-seeker, and every minute that passes increases the chance they book with whoever answers first.
Lead scoring is the practice of assigning each inbound lead a priority level based on the signals available at intake: urgency, system details, contact source, timing, and other factors that correlate with booking conversion rate. The job may already be logged in Jobber or Housecall Pro, but without a scoring layer, the insight that separates a hot lead from a cold one stays locked in a text field nobody reads in time. Leads with high scores get immediate, high-touch follow-up. Leads with low scores get automated nurture. Leads in the middle get a follow-up sequence that escalates to a human if they don’t self-convert.
The cost of skipping this step is concrete: an emergency lead that scores “high” after a CSR reviews it 45 minutes later is a lead your competitor has already booked. For a $2M+ HVAC operation, that is not a minor inefficiency. It is a direct drain on marketing spend and a compounding leak in booking conversion rate across every busy week of the season.
This article covers the automated lead scoring workflow that runs at the moment of contact.created, routes each lead to the correct response path within seconds, and ensures no high-likelihood lead waits more than 15 minutes for a callback.
TL;DR
- The problem: Treating all inbound HVAC leads with the same response time and effort misallocates CSR time and loses hot leads to competitors who respond faster.
- Lead score factors: Issue urgency, system age, contact source, time of contact, service area match, and any urgency language in the intake notes.
- The scoring workflow: AI Analyze node runs immediately on contact.created, produces a score of 1-10, and routes the lead to one of three response paths, Hot (8-10), Warm (5-7), or Cold (1-4).
- Hot lead path: Immediate CSR callback task (15-minute target) plus an instant acknowledgement SMS to the customer.
- Warm lead path: Immediate booking link SMS, then automated follow-up if no booking within 2 hours.
- Cold lead path: Email with information and booking link, light nurture if no response after 48 hours.
- Setup time: One workflow, approximately 60 minutes.
- Right fit: HVAC contractors handling 20 or more inbound calls per day get the clearest return. Smaller operations can run it with fewer nodes.
How Does Automated HVAC Lead Scoring Work?
Automated HVAC lead scoring evaluates each inbound contact against weighted intake signals, including urgency, system age, source, and timing, assigns a score, and routes the contact to the right response path within seconds. ServiceAgent runs this as a single workflow: an AI Analyze node scores the lead 1-10 at contact.created, and an AI Decision node routes it to the Hot, Warm, or Cold path.
Why Does Every HVAC Lead Need a Score?
The answer is response time. Speed-to-contact is the single strongest predictor of lead conversion in local service businesses. A lead contacted within 5 minutes is 10 times more likely to convert than one contacted within 60 minutes. Beyond 2 hours, the conversion likelihood drops significantly regardless of the quality of the follow-up.
Most HVAC contractors don’t have a system that knows which leads need the 5-minute response. Every inbound contact goes into the same queue. The CSR works through the queue in order. A hot lead that came in at 9am may not get a callback until 9:45am because three cold leads were ahead of it. By 9:45, that customer has called two more companies and booked with whichever one picked up fastest.
Lead scoring solves this by creating the queue prioritization automatically. Hot leads jump to the top. Cold leads can wait for automated follow-up. The CSR’s time is allocated to the leads where human speed makes the biggest difference.
What Signals Does the Scoring Use?
| Signal Category | Factor | Score Contribution |
|---|---|---|
| Issue Urgency | System down: no heat in winter, no cooling in summer, or any safety issue (gas odour, burning smell, water leak) | +3 to +4 |
| Issue Urgency | Performance issue: system running but underperforming, unusual noise, intermittent operation | +2 |
| Issue Urgency | Routine service: maintenance, tune-up, estimate with no urgency indication | 0 to +1 |
| System Age | Older system: 12+ years old, multiple recent failures, or end-of-life symptoms mentioned | +2 |
| System Age | Newer system: recently installed or under warranty | 0 |
| System Age | Unknown age | +1 |
| Contact Source | Referral from existing customer | +2 |
| Contact Source | Emergency search (Google “emergency HVAC near me”) or after-hours call | +2 |
| Contact Source | Standard Google search or website form | +1 |
| Contact Source | Third-party lead marketplace or shared lead service | 0 |
| Time of Contact | Peak season (Jun-Aug cooling, Dec-Feb heating) business hours | +1 |
| Time of Contact | Peak season after hours | +2 |
| Time of Contact | Off-season weekday business hours | 0 |
| Urgency Language | High-urgency phrases detected in intake notes (“not working,” “no heat,” “emergency,” “children,” “elderly”) | +2 |
| Urgency Language | Neutral language | 0 |
| Service Area Match | Address within primary service area | +1 |
Introducing the Workflow Builder
The Workflow Builder is a visual drag-and-drop canvas inside ServiceAgent where you build automated sequences that fire the moment a trigger event occurs. For lead scoring, the trigger is contact.created, the earliest possible moment in the customer relationship. The AI Analyze node evaluates the lead within seconds. The AI Decision node routes it to the correct response path. The entire scoring and routing workflow completes before a CSR would even finish reading the intake notes. You configure the workflow once. It runs for every new contact from that point forward.
| Trigger | What fires | What it does |
|---|---|---|
| contact.created (Hot, score 8-10) | AI Analyze, AI Decision, Create Task, Send SMS | Creates a 15-minute CSR callback task and sends the customer an immediate acknowledgement SMS. |
| contact.created (Warm, score 5-7) | AI Analyze, AI Decision, Send SMS, Wait 2h, AI Decision, Create Task | Sends a booking link SMS, checks for a booking after 2 hours, and creates a follow-up task if no booking is made. |
| contact.created (Cold, score 1-4) | AI Analyze, AI Decision, Send Email, Wait 48h, AI Decision, Nurture or Archive | Sends an informational email, checks for a response after 48 hours, and moves unresponsive contacts to light nurture or archive. |
What Does the Automated Lead Scoring Workflow Look Like?
Right now, most HVAC contractors handle lead prioritization through CSR judgment, Jobber notes, or verbal handoffs between front-desk staff. When a new contact comes in, someone reads the intake note, decides how urgent it feels, and adds a callback to the task list. There is no consistent score, no automatic routing, and no guarantee that the emergency call from a homeowner with no AC in July surfaces before the routine maintenance agreement inquiry from two hours earlier. Front-desk turnover makes this worse: a new CSR has no institutional memory of what a high-value lead looks like, and by the time they develop that instinct, they may already be gone.
Trigger: contact.created
What it does: Fires the moment a new contact is created in ServiceAgent via any channel, AI voice agent inbound call, web form, manual CSR entry, or referral form.
What you do: Ensure all intake channels create a contact.created event in ServiceAgent. If your AI voice agent runs on a separate system, configure it to push new contacts to ServiceAgent immediately after the intake conversation ends.
What to check: Test-create a contact via each channel and confirm the trigger fires within 10 seconds of contact creation.
Node 1: AI Analyze (lead score 1-10)
What it does: Reads all available fields from the contact.created event, intake notes, source, timestamp, address, system details mentioned, and produces a lead score from 1 to 10 using the factor weights above. Outputs the score and a brief rationale (e.g., “Score: 9, system down, peak season, emergency search, urgency language detected, in service area”) to the contact record.
What you do: Configure the AI Analyze node with your scoring factor weights. Review the first 20 scored leads manually to calibrate the weights to your market. Adjust the score thresholds (Hot/Warm/Cold cutoffs) based on your CSR team’s capacity, if you can handle more Hot-path leads, lower the Hot threshold from 8 to 7.
What to check: After the first batch of real inbound contacts, review the scoring rationale for five High and five Low scored leads. Confirm the AI Analyze output matches your intuition about those leads’ likelihood.
Node 2: AI Decision (Hot / Warm / Cold)
What it does: Routes scores 8-10 to Path A (Hot), 5-7 to Path B (Warm), and 1-4 to Path C (Cold).
Path A: Hot Leads (Score 8-10)
Node A1: Create Task (immediate CSR callback, 15-minute target)
What it does: Creates a High-priority task for the first available CSR with the lead details, score rationale, and a 15-minute response time target. Fires push notification if the CSR has mobile notifications enabled.
Why it matters: Hot leads need a human within 15 minutes. The task creation puts the lead at the top of the CSR queue immediately, ahead of any other open tasks.
What to check: After a test hot lead, confirm the task appears in the CSR queue within 60 seconds with the correct priority level and score rationale.
Node A2: Send SMS (immediate acknowledgement)
What it does: Sends an immediate SMS to the customer: “Hi [Name], we received your call about [issue]. Our team will reach you within 15 minutes., [Business Name].” This holds the customer while the CSR is preparing to call.
Why it matters: A customer with an urgent problem who gets an immediate acknowledgement SMS is far less likely to call the next company on the list during the wait. The 15-minute commitment gives them a specific expectation to hold rather than an open-ended wait.
Path B: Warm Leads (Score 5-7)
Node B1: Send SMS (booking link)
What it does: Sends an SMS with a direct booking link: “Hi [Name], thanks for reaching out about [issue]. Book your appointment here: [link]. We’ll also call you within 2 hours to confirm., [Business Name].”
Node B2: Wait/Delay (2 hours)
Node B3: AI Decision (booked in 2 hours?)
If booked: workflow closes, contact tagged “Lead Converted.” If not: Create Task (standard CSR follow-up within business day).
Path C: Cold Leads (Score 1-4)
Node C1: Send Email (information and booking link)
What it does: Sends a low-pressure email: what types of service you offer, a link to your booking page, a phone number for questions, and a brief note about your response time for non-urgent requests.
Node C2: Wait/Delay (48 hours)
Node C3: AI Decision (responded or booked?)
If yes: contact moves to Warm path. If no: apply “Light Nurture” tag, contact enters a low-cadence drip sequence (one message per month for up to 3 months, then archived if no response).
Full workflow summary:
contact.created → AI Analyze (score 1-10) → AI Decision (Hot/Warm/Cold) → [Hot: Create Task (15-min callback) + Send SMS (acknowledgement)] | [Warm: Send SMS (booking link) → Wait 2h → AI Decision (booked?) → Create Task (if no)] | [Cold: Send Email (info + link) → Wait 48h → AI Decision (responded?) → Nurture or Archive]
What Changes After Running Lead Scoring for 90 Days?
Hot lead response time drops to under 15 minutes, consistently. When the CSR queue is automatically prioritized by score, hot leads always surface to the top. The 45-minute response time that was losing hot leads becomes a 10-minute response time that wins them. HVAC operators running this see a 75% booking conversion rate on the calls their AI handles, the direct result of routing hot leads to a human before the customer has time to dial the next number on their list.
CSR time reallocates toward higher-value leads. Cold leads handled by automated email sequences free up CSR capacity for warm and hot follow-up. The same team converts more business without additional headcount. HVAC operations also report saving 10 or more hours per week in front-office admin once the repetitive routing and follow-up work runs automatically.
Lead source quality becomes measurable. After 90 days, you can compare conversion rates by source, score distribution by source, and average score by season. Referral leads scoring an average of 8.2 versus web form leads averaging 5.1 is actionable data, it tells you exactly what your best lead channel is worth compared to any paid source.
Why ServiceAgent Handles This for HVAC
In HVAC, the difference between a hot lead and a cold one is visible in the intake data: the urgency language, the system age mentioned, the search term that brought the caller to your number, the time of the call relative to the season. ServiceAgent’s AI Analyze node reads all of those signals simultaneously at the moment of contact.created and produces a score before a CSR has finished the intake conversation.
The 15-minute callback target for hot leads is not achievable through a manual queue. A CSR working through inbound contacts in order will not always reach the emergency call first. The automated scoring and task creation at intake ensures the hot lead is always at the top of the queue, regardless of what came in before it and regardless of how busy the office is.
For HVAC contractors in markets where multiple companies are competing for the same emergency calls, speed-to-contact is frequently the deciding factor. A customer calling about no AC in July will book with the first professional company that calls them back. Lead scoring is the infrastructure that makes consistent 15-minute response times possible at scale. Visit serviceagent.ai to see how the intake scoring workflow is built.
Frequently Asked Questions
What if the AI Analyze node doesn’t have enough intake data to score accurately?
Configure a minimum data threshold for scoring: the AI Analyze node requires at least the contact source, timestamp, and any notes from the intake. If the contact.created event has no notes (manual entry with no intake data), the node defaults to a score of 5 (Warm) and routes to the Warm path. Add a note in the Create Task output that the score is data-limited and may require manual review. As more data is captured at intake, system type, urgency signals, issue description, the scoring accuracy improves.
How do we handle leads that come in outside business hours?
Hot-path leads that arrive after hours should still receive the immediate acknowledgement SMS, that message holds the customer. The Create Task should be flagged as “After Hours Hot Lead” with a target callback time of first thing the next business day (or, if you have after-hours dispatch capability, immediately). Warm and Cold path automation runs regardless of hours, since SMS and email delivery doesn’t require staffing. Configure the after-hours variant of the Hot path task to include the customer’s preferred morning callback window if the AI voice agent captured it.
Can the scoring model be updated as we learn more about what converts?
Yes. After the first 90-day period, review the conversion rate by score band. If score-7 leads are converting at the same rate as score-9 leads, the Hot threshold may need to be lowered. If score-4 leads are converting more often than expected, the Warm threshold may need to be lowered. The factor weights in the AI Analyze node can be adjusted at any time, reconfigure the weights, run a calibration batch on a sample of historical leads, and compare the new scores to actual outcomes before deploying the updated model live.
Is this workflow right for my size of HVAC operation?
HVAC contractors handling 20 or more inbound calls per day and running 10 or more trucks get the clearest return from this workflow. At that volume, manual queue management almost always lets hot leads wait behind cold ones, and a single missed emergency booking can cost more than a full month of automation. Smaller operations can run it with fewer nodes: the trigger logic stays the same, the output volume is lower, and the Warm and Cold paths can be simplified without losing the core benefit of automatic hot-lead prioritization.
