An honest 'Thursday at the earliest' books more work in a cold snap than a maybe
During a deep freeze your constraint is trucks rather than leads, and the shops that keep the work are the ones that tell people the truth on the first call.
What usually happens instead: the phone says today if we possibly can, the customer waits until early afternoon, calls somebody else, and your technician arrives at five to an empty house or a cancellation. You paid for the click, the call and the windshield time.
Give the phone a script that names a real window, offers a firm later slot, and offers a place on a cancellation list. Add the safe things a homeowner can check in the meantime, and a clear instruction to leave and call the gas utility immediately if they smell gas.
Capture the ones you cannot take. A callback list from one freeze week is a replacement pipeline for March, and almost nobody keeps it.
Give the phone visibility of same-day capacity. Dispatchers should not be discovering the board is full at the same moment the customer does.
No-heat calls and boiler replacement quotes are two funnels, so count them apart
Blending emergency service and planned replacement into one conversion rate hides what is happening in both.
The emergency funnel runs answered, booked, ran, sold the repair, converted to an agreement. High volume, fast, modest ticket, and heavily weather driven.
The replacement funnel runs quote given, followed up, second conversation, financing if it is needed, signed, installed. Low volume, slow, high ticket, and measured in weeks rather than hours.
One conversion number across both moves with the temperature and tells you nothing you can act on. A cold week improves it. A mild February ruins it. Neither has anything to do with how the business is being run.
Tag the type at intake. It is one field and it is the difference between a report you can use and a monthly average that flatters everyone.
A two-flat owner cannot say yes until the tenant and the other unit have been asked
In a city full of small multi-unit buildings the person on the phone is often not the only person who has to agree, and follow-up has to survive that.
An owner-occupied two-flat with a tenant upstairs brings access, disruption, sometimes a shared chimney and often one boiler heating both units. The decision involves at least one person who is not in the conversation.
Shape the follow-up for that. Send a written scope the owner can forward without editing, with the options laid out, a note on tenant access and timing, and a stated next contact date.
Ask two questions at the quote and write down the answers: who else needs to see this, and when do you expect to decide. Then diary it rather than relying on somebody remembering.
Most replacement quotes are not lost to price. They are lost to silence, on both sides.
Diagnostic fee, trip charge and after-hours rate belong in the first sixty seconds
Saying the price of showing up filters out the callers who were never going to pay it and lifts the booking rate on everyone else.
A homeowner on a cold morning is ringing several companies. Vague answers about pricing send them to the shop that told them a number, because at least that one sounded certain.
Say what the fee covers and whether it is credited against the repair. Uncertainty about the second point is what makes people hesitate, not the first.
Put the same numbers on the site so the phone call is a confirmation rather than a surprise. Consistency between the two builds more trust than either does alone.
Watch two figures after you make the change: booking rate per answered call, and the share of run calls that sold. Call volume may fall. Revenue per truck day is the number that matters.
Between the web form and the dispatch board is where February inquiries disappear
Most leaks are handoffs, and a handoff is the one part of the process nobody owns.
Trace a single form submitted at 5:40 on a February morning. Who receives it, on what device, how quickly does anyone see it, and what happens if the notification lands in a shared inbox nobody opens until eight? Run that trace before you spend another dollar on traffic.
Do the same for the after-hours service. Call your own line at an awkward hour and listen to what is promised on your behalf, then decide whether it matches what your team can actually deliver the next morning.
Reconcile totals monthly. Count inquiries by source against jobs that reached the board. The gap between those two numbers is the leak, and in most shops it is larger than the owner expects.
Fix the handoff before touching the ad account. Extra volume flowing into a broken handoff just produces more people telling their neighbors you never called back.
The phone gives you enough volume to test something; the landing page does not
One metro will not produce the sessions needed for small-effect page tests, but a busy phone produces hundreds of conversations a month you can change and measure.
A test on a button or a headline needs volume a single Chicago HVAC shop rarely has, and the winter weather moves the baseline underneath it anyway. Results from those tests are usually noise presented as insight.
Where you do have sample size is the phone. Change the script, listen to fifty calls before and fifty after, and count booked jobs per answered call. That comparison is imperfect and it is still far more honest than a two week page test in January.
For the site, change big things and hold them for a full season, then compare against the same season a year earlier. Small sequential tweaks in a market with two hard peaks cannot be read.
Keep a one line note of why each lost quote was lost, written by whoever was there. Read the list once a month. Patterns show up in that list long before they show up in any dashboard.
Questions we actually get
- How much of our lost work is actually at the phone?
- We cannot tell you without looking, and the honest first step is measurement rather than a claim. Pull the call log by hour and day for a peak week, count rings that went unanswered or to voicemail, then reconcile inquiries by source against jobs that reached the dispatch board. Most shops find the gap is bigger than they assumed, but the size is specific to the business.
- Should we hire an answering service for winter nights?
- It depends on what happens next morning. A service that takes a name and a number and promises a call back can work if somebody genuinely calls back early. One that promises a same day visit you cannot deliver does more damage than voicemail. Before signing anything, call your current setup at an awkward hour and listen to what a customer actually hears.
- Is it worth running A/B tests on our website?
- For most single metro contractors, rarely on small elements. There is not enough traffic to separate a real effect from weather and seasonality. Change substantial things, hold them for a season and compare against the same period last year. Put testing energy into the phone script and the quote follow-up, where the sample size is real.
- What numbers should we be reporting every month?
- At minimum: inquiries by source, answered rate, booking rate per answered inquiry, run rate, sold rate, and average ticket, split between emergency service and planned replacement. Add booked jobs by county if you sell across Cook and the collar counties. Six numbers read monthly beat a thirty page dashboard nobody opens.
- Will fixing intake actually raise revenue, or just the conversion rate?
- Nobody can promise a revenue figure, and we will not put one in a proposal. What we can say is that the work targets jobs you have already paid to attract, so improvements show up in booked and sold work rather than in traffic charts. The effect is measured against your own prior season, not against an industry benchmark.
