A lot of contractors get an AI receptionist, watch it answer a few calls, and figure they're good. Problem solved. Back to work.
Then six months later, they're not sure it's actually moving the needle. Leads still feel inconsistent. They're still stressed about the phone. They can't really say whether the tool is earning its keep.
That's not an AI problem. That's a measurement problem.
If you don't track the right numbers, you can't know what's working. And if you don't know what's working, you can't improve it. This post is about the three numbers every contractor should be watching if they're using an AI receptionist — and what it means if yours are off.
Number One: Answered Call Rate
The most basic metric. Out of every call that comes in, how many are actually being answered — by your AI, by you, or by someone on your team?
Before AI, most solo operators and small crews answer somewhere between 50% and 65% of inbound calls. The rest go to voicemail or ring out. That's just the reality of being in the field.
With an AI receptionist running 24/7, that number should be close to 100%. Not 80%. Not 90%. Close to 100%.
If your answered call rate isn't near the top, something is wrong. Either the AI isn't triggering on missed calls, the routing isn't set up right, or there's a gap somewhere in after-hours coverage.
How to track it: Pull your call log for the last 30 days. Count total inbound calls. Count how many were answered (by AI or human). Divide. If you're not near 100%, that's your first fix.
Number Two: Lead-to-Estimate Conversion Rate
This one matters more than most contractors realize. It's not just about answering calls — it's about what happens after the call is answered.
Lead-to-estimate conversion is the percentage of inbound leads who actually book an estimate or site visit. For most contractors, this rate sits somewhere between 25% and 45% depending on trade, market, and how fast you follow up.
Here's what changes with an AI receptionist: speed and consistency.
The industry average for callback time on a missed call is about 47 hours. That's not a typo. Nearly two full days. Studies in home services show that calling back within 5 minutes makes you 100x more likely to reach the prospect than calling back after 30 minutes. After an hour, most people have already made a decision.
An AI that responds instantly — even at 10pm on a Saturday — captures leads that would have disappeared by Monday morning. That should show up directly in your lead-to-estimate rate going up.
If your conversion rate hasn't moved since adding an AI receptionist, look at two things: (1) Is the AI collecting contact info and job details, or just saying "someone will call you back"? (2) Is there a follow-up sequence in place, or does the lead just sit in an inbox?
A good AI receptionist doesn't just answer. It qualifies, collects, and either books or triggers a follow-up immediately. If yours isn't doing that, it's leaving money on the table.
Number Three: After-Hours Leads Captured
This is the one most contractors forget to track — and it's often the most eye-opening number.
Look at your call logs from the past 30 days. Filter for calls that came in outside your normal hours — evenings, weekends, early mornings. How many came in? How many got answered? How many turned into booked estimates?
For most trades, 30–40% of inbound calls come in outside of 8am–5pm Monday through Friday. That's not a small slice. It's a third of your lead volume happening when you and your team aren't available.
Before AI, those leads mostly disappeared. Now they can be captured at the same rate as calls during business hours — IF your AI is actually set up to handle them properly.
Contractors who track this number consistently report that after-hours leads have a slightly higher close rate than daytime leads. The theory makes sense: someone calling at 9pm on a Tuesday about their HVAC going out isn't shopping around. They need someone. If you answer and book the estimate, you're in pole position.
What Good Numbers Actually Look Like
Here's a simple benchmark table for contractors who are running their AI receptionist well:
- Answered call rate: 95%+
- Lead-to-estimate conversion: 35–50% (up from industry average of 25–35%)
- After-hours leads captured: Same rate as business-hours leads
- Response time for after-hours leads: Under 60 seconds
If you're hitting these numbers, your AI receptionist is doing its job. If you're not, it's worth diagnosing which number is off and why — rather than assuming the whole thing isn't working.
The One Thing That Breaks Most AI Setups
After watching a lot of contractors set up and use AI receptionists, the most common failure point isn't the technology. It's the handoff.
The AI answers. The AI collects the lead info. And then... nothing happens fast enough. The lead sits in a queue. You get to it when you get to it. By then, the prospect has moved on.
The fix is simple: set up an immediate follow-up trigger. The moment the AI ends a call, a text should go out to the prospect confirming you received their inquiry and you'll reach out to schedule within the hour. That one step — a 30-second automated text — dramatically increases the likelihood that the lead is still warm when you call.
It's not glamorous. But it's the difference between capturing a lead and losing one.
Running the ROI Numbers
Let's make this concrete. Say you're a contractor doing $400K per year. You get 50 inbound calls per month. Before AI, you're answering 60% — 30 calls. After AI, you're answering 95% — 47 or 48 calls.
That's 17–18 extra calls per month that actually get answered. At a 35% lead-to-estimate conversion and a 40% close rate on estimates, that's roughly 2–3 extra jobs per month. At an average job value of $2,500, that's $5,000–$7,500 in additional monthly revenue.
Against a $197/month AI receptionist subscription, that's a return of 25x to 38x. Even if your numbers are half as good, it still pays for itself many times over.
The math works. But only if you're measuring it — and only if your setup is actually running the way it's supposed to.
What to Do Right Now
If you have an AI receptionist, spend 20 minutes this week pulling those three numbers. Answered call rate. Lead-to-estimate conversion. After-hours captures. Write them down. Compare against the benchmarks above.
If you don't have an AI receptionist yet, those same numbers show you exactly what you're leaving on the table every month. The gap between what you're capturing now and what you could be capturing is real revenue — not hypothetical.
Either way, the information is worth having.
Track the numbers. Fix what's broken. Let the AI do the work it was built to do.
That's how you know it's actually working.