The city question comes before the name question, because it decides the state
In a metro that crosses the Ohio River, the address determines whether the job is servable at all, so it should be the first thing you capture.
Train the phone to open with the property location. Not as an interrogation, as routing. A caller in New Albany and a caller in Fern Creek are on different tracks from that moment forward.
If you serve only Kentucky, you find out in ten seconds instead of after an estimator has driven over a bridge. If you serve both, you get to say so immediately, which is worth more from a person than from a web page.
Requirements and practice differ between the two states and between jurisdictions, so keep whatever your team says about it accurate and confirm the specifics with the appropriate state or local authorities and your own counsel. A confident wrong answer on the phone is the expensive kind.
Record the county, not just the city, in the CRM. Jefferson, Oldham, Bullitt, Clark and Floyd should each be a value you can filter on later, because that is how you will find out where you are actually winning.
A warehouse schedule does not bend around a two o'clock measure
With a large logistics and air-freight workforce in the market, a meaningful share of your customers are asleep during your default appointment window.
Ask when they are normally available before offering a slot. One question, asked early, prevents the reschedule that costs you a week and often costs you the job to whoever could come sooner.
Look at your cancellation and no-show data by time of day. If your afternoon slots hold worse than your morning slots, you are probably scheduling around your convenience rather than theirs.
Early morning and evening measures are a real differentiator in this market rather than a marketing line, because the demand for them is genuine. Offer them if you can staff them and say nothing about them if you cannot.
Confirm twice, with the day before and the morning of, and make the confirmation cancellable in one tap. A clean cancellation you can refill beats a no-show that burns the whole slot.
Photos of the openings before the drive save a morning in Oldham County
Drive times across this metro are long enough that an unqualified measure costs half a day, and half of the qualification can be done from a phone.
Ask for two or three photos with the inquiry: the outside of the house, a typical window from inside, and anything unusual. On a shotgun house in Germantown or a Victorian in Old Louisville, one photo tells you more about the job than five minutes of description.
Establish the opening count and the material honestly. A homeowner who says twelve windows often means eight windows and four somethings, and the difference decides whether this is a full measure or a phone quote.
Stack appointments by corridor. A morning that runs Prospect to Shepherdsville has already lost an hour to the Gene Snyder and the Watterson before anybody measures anything.
Give the small job a cheaper path. Two windows does not need a senior estimator in a truck if a photo and a phone call can produce a number the customer trusts.
The financing step is where a signed Louisville job goes quiet
Plenty of window and door contracts die between the verbal yes and the funded approval, and almost nobody measures that gap.
Track it as a stage of its own. Quote sent, verbal acceptance, application started, application approved, contract signed, job scheduled. If the drop between application started and approved is invisible in your reporting, you are losing jobs to a step you never see.
Decide who owns the follow-up during that window and give them a script. The customer is not ignoring you, they are usually waiting on paperwork and quietly reconsidering.
Publish the process plainly wherever the customer sees it: what documents are needed, roughly how long a decision takes, what happens if it does not go through. Uncertainty at this stage is what sends people back to the other two quotes.
Never promise an approval or an outcome. Describe the process, keep the language general, and let the lender speak for the lender.
Count contracts and openings by county, because one average hides Southern Indiana
A single metro-wide close rate is three or four different businesses averaged into one number that tells you nothing you can act on.
Report by county and by job type. Whole-house replacement in Jefferson County, whole-house in Clark and Floyd, doors, small jobs. The averages between them will not resemble each other.
Count signed openings and contract value, not just contracts. Twenty two-window jobs and four whole-house jobs can look identical in a lead report and represent completely different quarters.
Watch the path all the way through: inquiry, contact made, measure booked, measure held, quote delivered, contract signed. The biggest single leak in most shops is between measure booked and measure held, and almost nobody instruments it.
Read the numbers monthly, not weekly. One metro does not generate the volume for weekly conclusions, and reacting to noise is how good processes get thrown out.
You have enough volume to judge a rule, not enough to judge a color
A single metro will never produce the traffic for small split tests, so change things that are big enough to see without statistics.
Test rules, not variants. Answering every inquiry within five minutes during business hours. Requiring photos before booking a measure. Offering a two hour window instead of a four hour one. Following up for ninety days instead of two weeks.
Run one change at a time, hold it for a full season, and compare booked and held measures against the same period before it. Note the weather and anything unusual, because winter distorts the numbers here.
Where you do have volume, it is usually on the phone, so start there. Thirty call recordings will show you more about where inquiries leak than any dashboard, and the fixes are usually free.
Write down what you changed and when. Six months later, the only thing that will let you interpret a shift in close rate is a record of what you did to cause it.
Questions we actually get
- What is the single biggest leak in a window and door shop?
- In most shops it is the gap between an inquiry arriving and a human making contact, followed closely by the gap between a booked measure and a held one. Neither requires new software to fix. Both require somebody to own the number and look at it weekly until it moves.
- How fast do we need to respond to a web inquiry?
- Fast enough that you are the first company to reach them, which in practice means minutes during business hours and a clear early morning process for overnight inquiries. Set a standard you can actually staff and measure compliance against it, rather than adopting a number from an article and quietly missing it every day.
- How long should we follow up on an unsigned quote?
- Longer than most dealers do. Window and door replacement is often deferred for months, especially when it competes with other household spending, and a quote sent in November may be decided in March. Build a cadence that keeps going after the first two weeks and give the customer something useful each time rather than another request for a decision.
- Should we ask for photos before booking a measure?
- It is one of the few changes with a clear payoff in a metro with long drives and a wide range of housing. Some customers will not send them, so make it a request rather than a barrier. For older homes on narrow lots, the photos also let you send the right person with the right equipment the first time.
- Can we run A/B tests on our site?
- Realistically, not on small changes. One metro will not generate enough traffic to resolve a small difference, and a test that runs long enough to try will be contaminated by seasonality. Change substantial things instead, one at a time, hold each for a season, and judge them on booked and held measures rather than clicks.
