Web Design for Construction Companies: What the Site Actually Has to Do

A construction company's website has one job: convince somebody who already has your name that you are a real firm worth calling. Almost nobody finds a general contractor by browsing. They get a name from an architect, a developer, a past client or a supplier, and then they look you up to see whether the name holds. The site is the inspection, not the introduction.

That single sentence changes what the site has to be good at. It is why a beautiful, slow, vague website loses work that an ugly, fast, specific one wins. And it is why almost every article about web design for construction companies is aimed at the wrong target: they are written by agencies selling design, so they talk about looking modern. The question that matters is functional. What does the site have to do?

This is written from the build side. We are a software studio, not a design agency, and we spend more time on what a page has to accomplish than on what it looks like. If you would rather start with the pipeline problem than the website problem, read how construction companies actually get leads first, because the website is the last step of that process rather than the first.

Start by being honest about who is on the site

Before any of the specifics, the audience question has to be settled, because it determines everything else. There are four people who open a contractor's website, and they want different things:

• The referred prospect. Somebody gave them your name. They are verifying that you exist, that you build the kind of thing they need, and that you are not going to be an embarrassment to the person who recommended you. This is the majority of your valuable traffic.
• The general contractor or developer shortlisting subs. They are checking scope, capability, licensing, bonding and whether you work in their area. They are often on a deadline and comparing you against two other firms in the same twenty minutes.
• The candidate. Skilled trades are hard to hire, and people research an employer before applying. A careers page that looks abandoned costs you applicants.
• The cold searcher. Genuinely looking for a contractor with no referral. Real, but a much smaller share than most marketing advice implies, and the hardest to convert.

The mistake is building for the fourth reader and hoping the first three are impressed. A site optimised to attract strangers reads as marketing. A site optimised to survive scrutiny reads as competence, and it happens to convert strangers better too.

The five things the site actually has to do

1. Prove the work is yours

The project gallery is the highest-value part of a construction website and the part most often ruined. It gets ruined in a specific way: the firm does not have good photographs, so the site gets filled with stock imagery, renderings, or three photos of the same job from 2019.

A prospect can tell. Stock construction photography is recognisable — the hard hats are too clean, the site is too tidy, nobody is standing in mud. The moment a visitor suspects the imagery is not yours, everything else on the page loses credibility, including the claims that are true.

What a project entry needs, in rough order of how much it matters: photographs of the finished work, the location at city level, the scope in plain words, the approximate contract value or size, the duration, and the delivery method. A named client and a quote are worth more than all of it, and are worth asking for at handover while everyone is still pleased with you.

The practical fix is a process one, not a design one. Somebody has to take photographs at closeout on every job, with a real camera or a recent phone, in decent light. Twelve projects documented properly will outperform forty documented badly, and the site should be structured so that adding a project is a five-minute task rather than a call to a developer.

2. Match how people actually search

Contractors search and get searched for by place, and the place that matters is where the work is, not where the office is. A firm headquartered in Katy that builds across greater Houston is invisible to somebody searching for a builder in Sugar Land if the only geographic signal on the entire site is a footer address.

The usual solution is an Areas We Serve page with a list of thirty city names. It does not work, and it can look like spam. What works is a smaller number of genuine pages, one per market you actually operate in, each carrying something real: the projects you have completed there, the permitting or inspection quirks of that jurisdiction, the trades you self-perform there, and a local point of contact.

For a Texas contractor the natural cuts are the metro markets rather than the municipalities — greater Houston, Dallas–Fort Worth, Austin, San Antonio — with sub-pages only where you have enough completed work to justify one. Three substantial market pages beat thirty thin ones, and the test is simple: if the page would be useless to a reader who already knew you worked there, it is not a page.

3. Load on a phone with two bars

This is the requirement most often treated as a technical nicety and it is not. Your site gets opened outdoors, on a phone, on a congested or weak connection, by somebody who is standing on a job site or sitting in a truck. A homepage carrying a full-screen autoplaying video of cranes is close to unusable in those conditions, and that visitor does not wait.

Google publishes fixed thresholds for this under Core Web Vitals, and they are measured against real visits rather than a lab test. A page passes only when it meets all three for 75% of visits:

The usual culprits on a construction site are unoptimised job photographs uploaded straight from a phone at full resolution, a heavy page builder, a stack of third-party scripts, and hero video. Job photography is the specific one worth calling out, because the thing that makes the gallery valuable is also the thing most likely to make the site slow. Serving correctly sized, modern-format images is the fix, and it is a build decision rather than a content decision.

4. Capture scope, not just a name

A quote form with three fields — name, email, message — is the default on almost every contractor website, and it converts an enquiry into an unqualified obligation. Somebody now has to call, discover the project is out of scope or out of area, and lose an afternoon.

The counter-argument is that longer forms reduce submissions. They do. That trade is usually worth taking for a contractor, because the cost of a bad lead is measured in hours of somebody senior's time rather than in a wasted click, and because a serious buyer will answer questions that a tyre-kicker will not. This is our own operating view rather than a research finding, and the way to settle it for your business is to instrument the form and look after ninety days.

The fields worth having, in the order they should appear: project type, location, approximate size or value, target start date, whether drawings exist, and how they found you. That last one is the cheapest piece of business intelligence you will ever collect, and almost nobody asks.

5. Make the credentials checkable

Trust signals on construction sites tend to be decorative — a row of association logos in the footer, greyed out, linking nowhere. The version that works is specific and verifiable: licence numbers for the trades that require them, the bonding capacity and the surety, insurance limits, safety record, and the associations you are genuinely a member of, linked to a profile that proves it.

Texas is a case where getting this right takes a little care, because licensing is not handled in one place. Some trades are licensed at state level while general contracting is generally handled through city and county registration and permitting, and the requirements differ between jurisdictions. Confirm what applies to your trades and your cities with the Texas Department of Licensing and Regulation and the building department in each city you work in, then display exactly those numbers. We have deliberately not summarised the rules here, because a licensing claim that is slightly wrong on your own website is worse than no claim at all.

The general principle: every credential on the page should be something a sceptical general contractor could verify in two minutes without calling you. That is what a credential is for.

What this costs when it is wrong

The failure is quiet, which is why it persists. Nobody emails to say they chose somebody else because your site looked dated, and no analytics report has a line item for the referral who checked you out and did not call. The loss is invisible and it compounds, because the referral you lost was the highest-intent lead you were ever going to get.

This is the argument for treating the website as infrastructure rather than marketing spend. It sits at the end of every other channel — the referral, the truck livery, the bid board listing, the search result — and it either converts that attention or wastes it. A firm doing eight figures with a site built in 2016 by a cousin is losing more work to it than to any competitor's marketing.

How to tell whether yours does the job

A short and slightly uncomfortable audit. Do it on your phone, on mobile data, not on the office wifi:

• Open your homepage and count the seconds until you can read something useful. Over three, you have a problem.
• Find your project gallery. Count the projects with real photographs of work you actually did. Under ten, that is the first thing to fix.
• Search for your firm plus a city you work in but are not headquartered in. If nothing of yours ranks, you have no service-area coverage.
• Fill in your own quote form as a prospect would. Notice how much you would still have to ask on the phone.
• Look for your licence numbers, bonding capacity and insurance limits. Time how long it takes to find them.
• Open the careers page. Ask whether you would apply.

Most contractors fail three or four of these, and the fixes are not evenly expensive. Photography and form fields are cheap and immediate. Performance and service-area structure usually mean a rebuild, because they are consequences of how the site was constructed rather than what was put in it.

Where a rebuild is and is not worth it

Not every site needs replacing. If the current one is on a reasonable platform, loads acceptably and simply lacks content, the answer is a photographer and an afternoon, not a project. Rebuild when the platform itself is the constraint: when adding a project means calling an agency, when the page builder makes it impossible to pass Core Web Vitals, or when the structure cannot accommodate service-area pages without becoming a mess.

The other trigger is integration. Once the site starts feeding a CRM, a bid tracker or an estimating workflow — rather than emailing a form submission to an inbox that three people ignore — it stops being a brochure and becomes part of how the business runs. That is usually the point at which the build pays for itself, and it connects directly to the back-office automation that has been moving fastest in construction over the past year.

Sources and method

• Core Web Vitals thresholds: Google's published good-score boundaries for Largest Contentful Paint, Interaction to Next Paint and Cumulative Layout Shift, each assessed at the 75th percentile of real-user visits. Documented by Google at web.dev.
• Texas licensing: deliberately not summarised. Requirements differ by trade and by city, and we could not verify the current position against a primary state source at the time of writing, so this guide tells you where to check rather than telling you what the rule is.
• Form-length and channel claims are stated as Inferya's operating experience where no primary source is cited, and are labelled as such in the text rather than dressed up as research.

If a figure in this guide is not attributed to a named source, it is not presented as a fact. That is deliberate: the construction marketing literature is full of conversion statistics that trace back to a single undated agency blog post, and repeating them would make this page less useful, not more.

The next step

If the audit above turned up three failures and you already know which one is costing you work, that is the place to start — not a redesign. Most of the value in a contractor's website is recoverable without touching the visual design at all.

We build sites like this: fast by construction, structured around service areas and real project records, with forms that capture scope and feed something useful. If you want a straight assessment of what your current site is costing you, tell us about the business and we will come back with a specific answer rather than a proposal deck. Our web development work is where the technical detail lives if you want to see the approach first.