The first screen has to say old house, not new construction
A stock hero of a waterfront new build tells a Hyde Park homeowner they are on the wrong site, and they leave before reading a word.
Lead with work that looks like the visitor's house. A bungalow porch in South Tampa, a block wall in St. Petersburg, a casement in an older Ybor City building. Recognition does the persuading faster than any headline.
One line of text should say what you install and where you install it. A second should say what happens next, which is a measure, and roughly how it is booked. Both need to be visible before any scrolling on a phone.
Put a call action within thumb reach and keep it there as the visitor scrolls. Most of your traffic is on a phone in the room with the problem, and asking them to hunt for a number loses a share of them.
Say the unglamorous thing early: that you replace openings in older homes where sizes are frequently not standard. Every competitor claims quality. Almost none of them claim that.
Explain the retrofit and the full frame path before the visitor asks
Homeowners in older stock have no idea whether their opening takes an insert or needs the frame taken out, and the site that explains it plainly earns the call.
Give each path a page. What an insert retrofit is, what conditions favor it, and the trade off that the visible glass gets smaller. What full frame replacement involves, why an out of square or deteriorated frame usually forces it, and what happens to stucco, trim or siding afterward.
Show the messy middle. A photograph of a frame coming out of a block wall does more for trust than a rendering of a finished room, because it tells the reader you have done this before and are not hiding what it looks like on day two.
Explain why the measure is not a formality. In a house where no two openings match, the measure is the moment the real scope appears, and framing it that way makes booking it feel like progress rather than a sales visit.
Avoid publishing a number you cannot defend. If you give ranges, tie each to a defined scope and say what is excluded, particularly wall repair and finish work.
The form should ask for the wall, the count and a photo
Six or seven fields, and the ones that earn their place are the address, the opening count, the wall type and a picture of the worst opening.
The address decides dispatch, and in this metro it decides the day, because which side of the bay a house sits on determines when a crew can realistically be there. Capture it properly, not as a free text city field.
Opening count separates a two-window job from a whole house before anyone drives anywhere. Wall type, block or wood frame, changes the scope and the crew. Both can be radio buttons.
A photo upload is worth three questions. It must accept a camera roll image from a phone without an account, a plugin or a resize step, and the form must not fail silently when the file is large.
Ask who signs, and skip a required budget field. Owners, renters and family members standing in for a relative all fill in these forms, and a mandatory budget question loses more good inquiries than it filters bad ones.
Prove Pinellas work on a Pinellas page or the Clearwater visitor leaves
Someone in St. Petersburg is checking whether you cross the water regularly, and a single map pin over Tampa says you do not.
Build a real page per side of the bay. Jobs you actually completed there, the housing types common in that area, and review quotes that name the neighborhood. Generic city pages with the name swapped are transparent and they read as such.
State the service area honestly, including which days you are usually working on which side. Telling a Clearwater homeowner that measures over there happen on set days is more convincing than implying you are everywhere at once.
Never claim an address you do not staff. Beyond the ranking risk, a homeowner who drives past an empty suite has told everyone they know.
Where you have worked in an older district such as Ybor City or Hyde Park, show it. Old-house work is the hardest thing to fake and the easiest thing to prove.
The chat widget and the finance calculator cost more speed than your photos
Galleries of real openings are the strongest asset on the site, and they are almost never the main reason it is slow.
Third party scripts are the usual culprit. Chat bubbles, review carousels, financing calculators, video backgrounds and a tag manager holding six years of dead tags. Each one is a request to somebody else's server before your page finishes drawing.
Audit them on value. If the chat widget produces inquiries you can trace, keep it and load it late. If nobody has answered it in a month, it is a tax on every visitor.
Handle images properly rather than removing them. Modern formats, sized for the container, lazy loaded below the fold, and no rotating hero carousel. A gallery done correctly can be the fastest part of the page.
Test on a real phone on cellular, away from office wifi, in a neighborhood with older infrastructure. The number you should care about is inquiries, not a score in a testing tool.
Publish nothing about historic review, settling or an insurance claim you cannot stand behind
Three claims get window dealers into trouble on the web, and all three are avoidable with one sentence each.
On historic districts, say review requirements generally vary by district and scope and are worth confirming with the city before ordering. Do not restate district rules on your site, and never imply an approval is assured.
On settling and subsidence, never diagnose. Explain generally what can cause a door to bind, say a window company cannot rule out a structural cause, and point readers to a licensed engineer and to their insurer. Region-wide context is fine. A statement about a specific property is not.
On insurance, promise nothing. Do not suggest a product will produce a premium result or a claim outcome. Homeowners should be told to speak to their own insurer, and the site should say so in plain words.
Display license and insurance details accurately and keep them current. Use unretouched before and after photography of your own work. A gallery that looks slightly rough is more persuasive than one that looks assembled.
Questions we actually get
- How many fields should the quote form have?
- Around six or seven for this trade. Name, phone, email, street address, opening count, wall type and a photo upload cover almost every case. Every extra required field costs completions, so anything that does not change dispatch or pricing should be optional or asked on the phone.
- Will asking for a photo reduce the number of submissions?
- Slightly, and the tradeoff generally favors asking. A photograph of an older opening tells an estimator more than several questions could, and it filters out the inquiry that was never a real job. Keep it optional if you are nervous, and make sure it works from a phone camera roll with no extra steps.
- Do we need a separate page for every city we serve?
- Not every city, but at least one substantial page per side of the bay, and more only where you have real work to show. A page for St. Petersburg with actual Pinellas jobs on it does work. Twelve near identical pages naming every community in the county tend to help nobody and can drag the rest of the site down.
- Should we put prices on the site?
- Ranges tied to a defined scope can help, as long as they are yours and you can defend them. Say clearly what is included and what is not, particularly stucco or trim repair after a full frame replacement and any permit handling. A bare number with no scope attached invites an argument at the kitchen table later.
- Can we say on the site that we handle the permit?
- You can say what you generally do as part of a job, worded carefully. Say that permitting is typically required for replacement work, describe your usual role in it, and note that requirements vary and are worth confirming with the county or city. Avoid guaranteeing an approval or a timeline you do not control.
