The humidity complaint keeps getting booked as a service call
Somebody says the house feels damp, and the intake sheet turns it into a diagnostic visit with an hour allowed.
The complaint is common here and the cause is rarely a broken part. Equipment sized too large and short cycling. Return leaks pulling humid air from the space under a raised floor. Duct problems that no diagnostic slot has time to investigate. The technician arrives, finds nothing failed, charges a fee and leaves.
The customer was describing a system problem, and you sold them an inspection. Weeks later they call somebody else about a replacement.
Fix it at intake. Train whoever answers to recognize the language: sticky, muggy, mildew smell, running all day, water on a register. Route those to a longer assessment slot with a senior technician, and price it as an assessment rather than a repair call.
Then measure the difference. Count how many humidity described calls became proposals under the old routing and under the new one. It is usually the largest single conversion gain available in this market and it costs nothing but a change to the script.
An estimate on a Creole cottage stalls on where the equipment can go
The proposal never gets written because the placement question was left open on site.
On older raised houses, and anywhere a visible exterior may fall under historic district review, placement is not a detail. It decides the equipment, the line set, the labor and therefore the number. An estimator who leaves without resolving it cannot price the job, so the proposal drifts.
Give estimators a checklist that forces the question before they leave: where the outdoor unit goes, how the line set runs, whether the exterior is likely subject to review, what the customer will need to confirm with the relevant commission or the city, and who is doing that confirming.
Set the expectation on site rather than in a follow up email. Tell the customer plainly that placement has to be settled first, that you will not promise an approval, and what date they will have a written number. Ambiguity is what kills these jobs, not price.
Track proposals that were never issued as a separate category from proposals that were lost. Most shops lump them together, which conceals a failure of process behind an apparent failure of sales.
Ask for a picture of what is under the house before the truck rolls
Two photographs at intake decide whether the first visit finishes the job.
Under a pier and beam floor there could be reasonable clearance, or there could be barely enough to get a body through. The air handler could sit under the floor, in an attic, or in a closet. Nobody on the phone thinks to ask, and the technician discovers it on arrival.
Add it to the intake script and to the web form: where does the equipment sit, and can you send a photo of the outdoor unit and of the space under the house. Most customers will, and the ones who cannot are telling you something too.
The gain is not only in scheduling. Knowing the setup in advance lets you send the right technician with the right parts, which turns a return visit into a completed job and a completed job into an invoice.
Keep the request short and give a reason. People send photographs readily when they understand it means the right person shows up, and resist when it feels like a form.
The week the streets would not drain cost you the following month
The lost work is not the day of heavy rain, it is the rescheduling nobody managed afterward.
A stretch of heavy rain and slow drainage takes trucks off the road. The board fills with pushed jobs, the phone keeps ringing, and the calls that were bumped go to whoever answers next.
Decide in advance who owns the recovery. Somebody should be calling every displaced customer the same day with a new time, not sending a text at six in the evening and hoping. Booked customers who are moved without contact are the easiest work in your pipeline to lose.
Watch the estimates too, because they are the first thing cut when the schedule tightens. A replacement proposal delayed a week during a busy stretch is often a proposal delivered after somebody else has already been out.
Log those days. Over a season the pattern shows you whether the ceiling is demand, capacity or the recovery process, and only one of those three is fixed by more marketing.
Count the return trips caused by a question nobody asked
The most useful number in the business is usually one nobody records.
Every second visit for something intake could have captured is margin you paid for twice and a customer whose confidence dropped. Wrong truck, missing part, no access, tenant could not authorize, equipment in a place nobody mentioned.
Add one required field to the job close out: was a return visit needed, and why. Read the list monthly. Within a couple of months the top three causes will be obvious, and every one of them is fixable in the intake script.
Alongside it, keep a short funnel: inquiries received, answered, booked, attended, proposed, released. Most shops know the first and the last, which makes it impossible to say where anything is failing.
Split the funnel by parish. Orleans, Jefferson and St. Tammany will not perform alike, and the differences usually reveal a dispatch or coverage decision rather than a marketing one.
The offer and the answer time are the only variables big enough to move
One metro will not give you the volume to detect a small change, so stop trying.
Split testing a button color or a headline requires more traffic than a single market HVAC contractor generates. Running such a test anyway produces a number that looks like a result and is noise.
Change things large enough to show up in booked work. How quickly a call is answered outside business hours. Whether the diagnostic fee is applied to the repair. Whether a humidity assessment is offered as its own service. Whether an estimate arrives as a written proposal the same day.
Run one change at a time, hold it long enough to cover comparable weather, and compare against the same stretch of the previous year rather than against last month. Demand here does not sit still, so before and after comparisons across different conditions are worthless.
Support the numbers with call recordings. Listening to a batch of inquiries that did not book will tell you more about the real leak than any dashboard, and it costs an afternoon.
Questions we actually get
- Should we use an answering service for after hours calls?
- Only if you have listened to it. Most owners have never heard what their service says at one in the morning, and the script is frequently taking messages rather than booking work. Call it yourself on a weekend, listen to the recording, and decide whether that is the first impression you want. A well briefed service beats voicemail; an unbriefed one is not much better.
- Is charging a diagnostic fee costing us bookings?
- It filters, which is sometimes what you want and sometimes not. The question is testable at a scale that matters: try applying the fee to the repair, or waiving it on replacement assessments, and watch booked work rather than call volume. Whatever you choose, state it plainly on the site and on the phone, because surprise at the door is worse than the fee itself.
- How many calls should we review before changing anything?
- Enough to hear patterns rather than anecdotes, and specifically calls that did not book. Pull a batch across different days and times, listen with the script in front of you, and note every point where the customer had to repeat something or the answer was vague. The exercise usually pays for itself before the sample is finished.
- What do we actually need to track this properly?
- Call tracking so inquiries carry a source, required source and outcome fields in your field software, and a monthly reconciliation of invoiced jobs against those sources. Nothing exotic. The difficulty is discipline rather than software, and one person needs to own the reconciliation or it stops happening by the third month.
- Will A/B testing our website work for a company our size?
- Not for small changes. A single metro contractor rarely produces enough traffic to detect a modest lift, so the honest approach is to make larger changes to the offer, the response time and the intake process, hold them across comparable conditions, and judge them on booked and released work. Anyone promising statistically clean split tests at this volume is selling you a report, not a result.
