The same handful of failures, over and over
The striking result is not that these websites have problems. It is how identical the problems are. Across 15 metropolitan areas, 8 trades and 1,008 independent businesses with no connection to one another, the same short list repeats.
Share of the 997 businesses whose site could be read exhibiting each issue. Findings are reported in aggregate; no individual business is named or identifiable.
The expensive problem is not design. It is the handoff.
A customer finds you, decides you might be right, clicks contact — and sends four sentences with no dimensions, no material, no deadline and no photographs. Quoting now begins with three days of correspondence. And the ones who never wrote back are invisible: they do not appear in any inbox, so nobody counts them.
64% of the businesses in this study do this to themselves every week. It is the single most common finding, and it is not a design problem. A more attractive homepage does not fix it. The form is the product.
Template builders ship a generic contact form — name, email, message. That form is adequate for a restaurant and actively harmful for a fabricator, because the whole job of a fabrication inquiry is to arrive carrying a drawing, a material, a tolerance and a date. Nobody removed the right fields. They were never there.
Who was measured
| Trade | Businesses | Avg. opportunity |
|---|---|---|
| Metal & Fabrication | 296 | 26 |
| Cabinetry & Woodworking | 232 | 30 |
| Apparel, Print & Signage | 198 | 26 |
| Construction & Remodeling | 132 | 23 |
| Home Services & Inspection | 57 | 23 |
| Auto Body & Restoration | 36 | 24 |
| Venues & Events | 32 | 28 |
| Pools & Landscape | 24 | 23 |
Opportunity is a 0–100 measure of how much a site is currently leaving on the table: higher means more unaddressed. Average across the 997 readable sites is 26.5.
What they are built on
777 of 997 sites run on an identifiable template platform. The platform is not the problem; the untouched default configuration is.
The ten-minute self-audit
Every check below is one you can run on your own website today, without buying anything. They are the same first ten questions asked of every site in this study.
- Open your own site on a phoneCan you find the phone number and tap it in under five seconds? Most trade searches happen on a phone.
- Search your business name in GoogleDoes the result headline say what you do and where, or does it say "Home"?
- Click your own contact formCan a customer attach a drawing, a photo, or a measurement? If not, every quote starts with a round of questions.
- Read the first screen out loudDoes it tell a stranger what to do next, or does it describe you?
- Look for a review anywhere on the pageA stranger spending thousands needs evidence somebody trusted you first.
- Check the footer yearAn old year is the fastest way for a visitor to conclude you closed.
- Open your analyticsIf there is nothing installed, you cannot know how many customers the site lost last month.
- Send your homepage link in a text messageDoes a preview image appear, or a blank grey box?
- Find your licence numberCustomers are told to check it before hiring. Not showing it costs the careful ones.
- Ask the last five customers how they found youIf none say the website, that is the measurement you were missing.
If you answered badly to three or more, you are in the same position as most of the 1,008 businesses measured here — which is to say, an ordinary one. None of it requires a rebuild to begin fixing.
How this was measured
Each business's homepage and contact page were retrieved as an ordinary browser would retrieve them, then checked against a fixed list of criteria covering credibility, mobile usability, conversion path, and measurement. No judgement of visual taste was recorded, because taste is not measurable and not the point.
Sites that refused automated requests were excluded rather than assumed broken — a server declining a robot says nothing about what a customer sees. Businesses were identified by trade and metropolitan area from public listings; national chains and directory aggregators were removed so that every entry is an independent local business.
| Scope | |
|---|---|
| Businesses measured | 1,008 |
| Sites their host allowed us to read | 997 |
| Excluded — host refused an automated request | 11 |
| Metropolitan areas | 15 |
| Trades represented | 8 |
| Criteria applied per site | ~30 |
| Conducted | 2026 |
The order that matters
If the study says anything useful, it is that the sequence is wrong almost everywhere. Businesses rebuild the homepage and leave the contact form alone — which is the exact inverse of where the money is.
| Fix | Why it comes first |
|---|---|
| The intake form | It decides whether an inquiry arrives quotable or arrives as a question. Highest return, lowest cost. |
| Measurement | Until something counts inquiries, every other decision is a guess. |
| Proof | Reviews and finished work, near the point of decision. Cheap, and most sites simply omit it. |
| The phone path | Tap-to-call on every page. Trades still close on the phone. |
| The homepage | Last. It is the most visible and the least decisive. |
Want your own site measured?
We run the same checks by hand and send back an annotated review, a homepage concept and a map of where inquiries leak — a fixed price, credited toward the work if you decide to go ahead.
Get in touchMaking the case to whoever signs off
Most of the people who read this are not the person who authorises the spend. That conversation goes better with three things, and worse with a fourth.
Show them their own site, not this study
A percentage about a thousand other businesses is easy to wave away. The same finding on their own homepage, on their own phone, is not. Open the site in front of them and try to do what a customer would: find the phone number, start an inquiry, attach a photo. Whatever fails, fails in front of them.
Give them a number they can check, not one they have to trust
Do not open with a revenue projection. Nobody in a shop believes a marketing forecast, and they are right not to. Use what is countable instead: how many inquiries arrived last month missing information you needed, and how many messages it took to get it. That number is already in the inbox and it belongs to them, not to a consultant.
Make the decision reversible and bounded
A fixed scope at a fixed price is an easier yes than a redesign, because the downside is known before anyone commits. Ask for the smallest thing that would settle the argument — one page, one form, one measurement — rather than approval for everything at once.
What not to promise
Do not tell them a change will increase sales. Nothing in this study establishes that, and neither can anyone who has not measured it on their site. What can honestly be promised is narrower and more useful: that the information arriving with an inquiry can be made more complete, that the number of back-and-forth messages before a quote can be counted, and that both can be measured before and after. If someone promises more than that, they are guessing on your behalf.
Take this with you
We will send the study as a PDF, along with the section above written as a single page you can forward to whoever approves the budget. Leave an address and it comes across.
One email, sent by a person. No list, no sequence, and nothing else unless you write back.