The photograph that sells an Austin build has a heritage tree still standing in it
Proof that you designed around what the lot and the ordinance would allow does more work in the first screen than any amount of finished glamor.
Every competitor has beautiful water at dusk. Nobody remembers it. What a homeowner in Tarrytown or West Lake Hills actually wants to see is a finished pool on a lot that clearly had constraints: a mature tree still there, a grade change handled with structure, a deck sized to what the property could take.
Put that in the opening screen with a caption that names the constraint in one line. A tree kept and protected. A slope that needed engineered walls. A narrow lot where the deck had to be planned around cover limits before the shape was drawn.
Say where you build, plainly and near the top. Travis, Williamson and Hays cover a lot of ground and a buyer in Dripping Springs is checking whether you drive that far before they read anything else.
Skip the badge wall. Verifiable license and insurance details, real photographs of real crews, and named towns beat a row of generic trust icons that no visitor has ever clicked.
Build the navigation around the lot, not around your service list
Buyers here sort themselves by property type long before they sort themselves by product, so the site should offer the same doors they are already looking for.
A service list of construction, renovation and service tells a visitor almost nothing about whether you are right for them. Three lot-shaped entry points do: a Hill Country build with slope and access, a central Austin renovation behind an alley or a narrow drive, a new lot in Williamson or Hays County.
Under each door, keep the same structure so the site feels coherent: what the conditions usually mean, how the project runs, what a site evaluation looks at, photographs from that category, and one clear way to start.
Cross-link honestly. A renovation reader may discover their deck replacement raises an impervious cover question. Send them to that explanation rather than hoping the contact form catches them first.
Keep the top navigation short. Five items, one of them a phone number, is a better result on a phone than a dropdown system that requires precision from a thumb.
Let the homeowner upload the survey and half the first call disappears
The one field almost no pool site offers is a file upload, and it is worth more than any other question you could add.
Owners in Austin usually have a survey or a plat from their purchase, and increasingly a site plan from a prior project. Getting it at the inquiry stage tells you the lot shape, the setbacks, the existing hard surfaces and often the trees, before anyone spends half a day driving to Lakeway.
Ask for the address, what they want, roughly when, and the upload. Add one question about access, phrased so a non-builder can answer it: can a machine get to the back yard, and how wide is the gate or the alley. A photo from the back door is worth more than any text field you could write.
Keep the form short enough to complete standing up, and let anything optional be clearly optional. Long qualification forms lose exactly the serious buyer who is doing this on a phone between meetings.
Confirm receipt on screen and by email with a specific next step and a name. Silence after a submission is the most common and most fixable leak in this trade.
The chat widget, the review badge and three tracking tags cost more than your photographs
Most slow pool sites are not slow because of the gallery, they are slow because of everything bolted on around it.
Images are a known problem with a known fix: modern formats, correctly sized files, lazy loading below the first screen, and one carefully prioritized hero image. Do that once and photography stops being the bottleneck.
Third party scripts are the part that quietly returns. A chat widget, a review carousel, a booking embed, a heat map tool and two analytics tags each add requests and execution time, and they usually load before your content does. Audit them, keep the ones that earn their place, and load what remains after the page is usable.
Test on a phone on a mediocre connection, not on office wifi. A visitor in a back yard in Circle C or off a road in Bee Cave has one or two bars, and a site that takes its time will lose to whoever loaded first.
Speed is a sales problem, not a technical one. Every second between a tap and a photograph is a second the buyer spends in a competitor's tab.
Reviews, faces and a real address do the work a referral network used to do
A steady stream of newcomers arrives with nobody to ask about you, so the site has to carry the credibility that word of mouth carries elsewhere.
A household that closed on a house in Mueller or Kyle eight months ago cannot call a neighbor who used you in 2014. They read reviews, they look for a physical presence, and they check whether the people in the photographs look like the people who will show up.
Put a real address and service area on the site. Show the crew, the designer, the person who answers the phone. Name who manages a project once it starts. Publish license and insurance information in a form a visitor can verify rather than a claim they have to accept.
Pull reviews in lightly and link to the source. A fast static block of recent quotes with a link to the profile beats a heavy widget that slows every page it appears on.
Write a page about how a project runs from first call to fill, including the parts that take longer than people expect. Honesty about the front end of a schedule reads as experience to a buyer who has already been told a fairy tale by someone else.
Your equipment page should not assume the reader picks an electricity provider
Austin Energy is city-owned while most of Texas buys power from competing retail providers, so any equipment content written for the state can be wrong inside the city.
Heaters, pumps and automation are worth writing about, because owners ask. The mistake is copying statewide advice about shopping for a provider, which does not apply to an address served by the municipal utility and confuses everyone else.
Write it neutrally. Explain what different equipment does and what drives operating cost, then tell readers to confirm programs and eligibility with the utility serving their specific address, since terms change and boundaries do not follow neighborhood names.
Do not publish plan names, rates or rebate amounts. They date quickly, and a page still promising something that ended will cost you more credibility than it ever earned.
Handled well, the page becomes a small proof of local competence. A reader in Pflugerville or Cedar Park who has already found three contradictory answers online will notice that yours is the one that told them how to check.
Questions we actually get
- Do we need a full rebuild or can we fix what we have?
- Start with a diagnosis. If the site is fast, editable and structurally sound, better photography, real intake and lot-shaped pages will usually outperform a rebuild. Rebuild when the platform makes changes expensive, when performance cannot be fixed, or when the structure cannot support separate pages for the kinds of work you actually do.
- Should we publish prices?
- Publish price drivers rather than prices. Explain how excavation conditions, access, structure on a grade, deck area and equipment change a number. In this market a specific figure is either wrong for most lots or so hedged it is useless, and a clear explanation of drivers filters buyers just as effectively.
- How many photographs should a gallery have?
- Enough to prove range in each category you sell, organized so a visitor can find their kind of lot in a couple of taps. Twenty well captioned projects sorted by condition beat two hundred sorted by year. Caption what the constraint was, because that is the part buyers are reading for.
- Is a chat widget worth it?
- Only if a person answers it during business hours. An unstaffed chat is a slow script and a broken promise. If nobody can respond quickly, put a tappable phone number and a short form in its place and spend the effort on responding fast to what comes in.
- What should we never put on the site?
- Client addresses, recognizable house numbers, gate codes, and any photograph a homeowner has not agreed to. Skip claims you cannot verify and any statement that reads as engineering or legal advice. Let your engineer make engineering claims and point permit and ordinance questions to the city.