Ask what the frame and the opening are made of before you book a measure
Two questions on the first call decide whether the appointment is worth a morning, and in this market one of them is about the wall.
Brick construction is the default here, and older brick openings are rarely a standard size. An estimator who arrives without knowing whether the wall is brick or frame is guessing at appointment length and often at scope.
Get it on the intake script. What is the exterior wall made of, how many openings are you replacing, and what is failing right now. Three questions, under a minute, and they sort a whole house job from a single stuck door before anyone drives anywhere.
Add a photo request while the caller is still on the phone. People are almost always standing near the opening when they call, and a picture texted in during the conversation is easier to get than one requested afterward.
Then act on the answers. A two opening inquiry deserves a faster and cheaper path than a whole house measure, and running both through the same appointment process is how dealers end up with estimators driving all day for jobs that will never pay for the fuel.
Your schedule has to absorb a permit desk that answers on Tuesday mornings only
Administrative variation across St. Louis County municipalities is a scheduling cost, and pretending it is not shows up as missed promise dates.
A service area here can cross a dozen separate municipalities in a short drive, each with its own permitting and inspection practice. Two identical jobs can carry different amounts of office time before the truck moves.
Build that into what you tell customers rather than absorbing it silently. A promise date that assumes the fastest jurisdiction will be wrong often enough to damage the relationship on jobs you already won.
Track it. Log the municipality on every job and record the days from contract to permit in hand. After a season you will know which places to quote longer and which office tasks to start earlier.
Do not publish specifics about any jurisdiction's requirements. Requirements generally differ and they change, so the safe framing for customers is that the process varies by municipality and is worth confirming with the local building department.
Custom sizes stretch turnaround, so measure the days from measure to firm number
The gap between the measure appointment and a real quote is the least watched number in this business and one of the most expensive.
When openings are not standard, pricing takes longer. Sizes have to be checked, materials confirmed, sometimes a supplier consulted. Meanwhile the homeowner has two other quotes arriving.
Start counting. Days from measure to quote delivered, tracked by job type and by estimator. Most shops discover a wider spread than they expected, and the slow tail is where jobs are lost to whoever answered first.
Fix the process before you touch the copy. A same day ballpark with a firm number to follow beats a perfect quote that lands four days later, provided you are clear about which is which.
Consider what the quote document itself does. In a market where the buyer takes weeks to decide, the estimate is the artifact that sits in their inbox arguing your case. Priced opening by opening, with what is included and what is not, it does more work than any follow up email.
If you change one thing this quarter, this is a strong candidate, because the effect is large enough to see without a statistics package.
Deposit to install is a lag that unsells jobs you already won
A signed contract is not a completed conversion, and the weeks between deposit and install are where cancellations come from.
Custom windows get built. The wait is real and mostly outside your control. What is inside your control is whether the customer hears from you during it.
Silence in that window is what turns a signed job into a cancellation request, particularly when a competitor is still marketing to them. A short scheduled contact at ordering, at a manufacturing update, and at scheduling costs almost nothing and removes most of the anxiety.
Measure the cancellation rate as its own number, separate from close rate. A shop with a strong close rate and a quiet backlog can be losing more revenue after the sale than before it.
The same period is your best moment to set up a referral or a review request later, because the customer's experience of you is being formed by whether you stayed in touch, not only by the install.
Count signed openings, split city from county and Missouri from Illinois
A single blended conversion rate in this metro averages together markets that behave nothing alike.
Openings, not leads, is the right unit. A forty opening whole house job and a single patio door look identical in a lead count and are not remotely the same business.
Then split the report. City of St. Louis, St. Louis County, St. Charles County and the Metro East, at minimum. The city and the county are separate governments and separate markets, and the Illinois side runs under another state's rules entirely.
Instrument the full path: inquiry, contacted, qualified, measure booked, measure held, quote delivered, contract signed, job installed. Booked and held are different numbers, and the gap between them is often the largest single leak in a window shop.
Held measures deserve special attention in winter. Weather cancels appointments here, and a rebooking process that runs the same day beats one that waits for the customer to call back.
Read the whole thing monthly. Weekly reads on one metro's volume will have you chasing noise.
Before you test a headline, sit with the calls where the homeowner said no
One metro does not generate enough traffic to detect small effects, so your evidence is qualitative and your changes have to be large.
A split test that needs thousands of conversions to resolve will never resolve for a regional window dealer. Running it anyway produces a number that feels like proof and is not.
Use what a single metro does give you: recorded calls, lost quote reasons written down in the CRM, and the estimator's read on why the job went elsewhere. Thirty honest lost job notes will tell you more than a month of button color testing.
When you do change something, change something big. The offer, the intake script, the response time commitment, the structure of the quote document. Sequential changes with a clear before and after are the right method at this scale.
One change at a time, and give it long enough that the pipeline has turned over. A window sale takes weeks to close, so a change read after ten days is being judged on the previous month's work.
Write down what you expected before you change it. The discipline is what separates a real read from a story told afterward.
Questions we actually get
- What is the single most common leak for a window and door dealer here?
- Usually the gap between a booked measure and a held measure, closely followed by the days between the measure and a firm quote. Both are invisible in a report that stops at leads. Counting them separately for a month is often enough to show a shop where its revenue is going without changing anything else first.
- Can we run A/B tests on our site with one metro's traffic?
- Not usefully for small changes. The volume simply is not there to detect a few percentage points, and a test that ends early gives you a number that looks like evidence and is not. Make large sequential changes instead, to the offer, the script, the response time or the quote document, and read them in booked measures and signed contracts.
- How fast do we need to respond to an inquiry?
- Faster than the second dealer the homeowner contacted, which in practice means minutes rather than hours during business time and a same day path in the evening. Rather than adopting a number from an article, measure your own response times against your own close rates for a quarter. The relationship will be clear enough to act on.
- Should we track conversions differently for the Illinois side?
- Yes. The Metro East operates under a different state's rules, licensing generally differs by state, and the buyers there do not consider themselves shopping in Missouri. Reporting it inside a metro total will let a weak segment hide behind a strong one. The same logic applies to keeping the City of St. Louis separate from St. Louis County.
- Our close rate looks fine but revenue is flat. Where should we look?
- At the unit you are counting and at what happens after the signature. A close rate on leads hides the mix between whole house jobs and single openings, so switch to signed openings and average ticket. Then check cancellations between deposit and install, because a quiet backlog during a long build time is a common and largely fixable loss.
