Why Your Contractor Website Isn't Converting
Traffic is rarely the problem. The problem is that nothing on the page answers the question a homeowner is actually asking.
A long read on why home improvement websites lose interested homeowners, and the specific structural changes that recover them.
The homeowner is not shopping. They are managing risk.
Consider what a $22,000 kitchen remodel actually asks of a person. Strangers in the house for three weeks. Dust in every room. A payment schedule that overlaps with a mortgage. A decision that will be visible every single day for a decade, and visible to every guest who walks in.
Against that, a website that opens with a rotating banner and the sentence 'quality workmanship at competitive prices' is not merely bland. It is unresponsive. It answers a question nobody asked while ignoring the three that are actually running through the homeowner's head.
Once you accept that framing, the redesign priorities invert. The gallery is not a portfolio flex; it is evidence. The warranty is not fine print; it is the answer to 'what if this goes wrong.' The form is not a formality; it is the smallest possible commitment you can offer somebody who is not ready to commit.
The three questions every page has to answer
Have you done this exact job, in this kind of house, near me? Homeowners pattern-match against their own property. A ranch in their zip code beats a mansion two states away, every time. Organize proof by home type and neighborhood, not by chronology.
What happens if something goes wrong? Warranty terms, licensing, insurance, who to call, and what a change order actually looks like. Companies fear that this invites problems. In practice it signals that you have handled problems before.
What is the next step, and how small can it be? A homeowner nine weeks from signing will not fill in eleven fields. They will look at photos, check a price range, and maybe send one message. Give them a step sized for where they are.
Sites that answer those three questions above the fold consistently outperform sites that lead with a slideshow.
Where the traffic actually leaks
In audits, the same handful of leaks appear over and over. They are boring, and they are expensive.
The six most common leaks
- 01A homepage-only conversion path, when most paid and organic traffic lands on interior pages
- 02Forms that ask for address, budget and timeline before establishing any trust
- 03Phone numbers that are images, or not tappable, on mobile
- 04No after-hours capture, on a category where most research happens at night
- 05Price silence, which sends the homeowner to a competitor who gives a range
- 06A gallery of twelve untagged photos with no context, location or scope
Proof beats polish
A slightly imperfect photo of a real tear-off, taken by a foreman at seven in the morning, outperforms an expensive stock image every single time. It is legible as truth. The stock image is legible as marketing.
The practical problem is that the best work lives on crew phones and never leaves them. Fix that with a standard rather than a campaign: same three angles, before and after, one wide shot with the house visible, uploaded to one place at the end of the job.
Within a season you have hundreds of assets covering exactly the home types in exactly the neighborhoods you serve. That library feeds the website, the Business Profile, and every retargeting sequence you will ever run.
Speed is a trust signal, not a technical metric
Homeowners do not experience 2.4 seconds as a Core Web Vital. They experience it as competence. A site that stutters on a phone in a driveway suggests a company that will also be slow to call back.
Most contractor sites are slow for two reasons: uncompressed project photography, and a stack of plugins bolted on across five years of small requests. Both are fixable in days, not months, and both usually move the conversion rate before any design work happens.
What to change first, in order
Do not redesign everything. Rebuild the estimate path — the pages traffic actually lands on, the request flow, and what happens in the first ten minutes after a submission. Measure that for thirty days before touching the rest of the site.
This sequencing matters because it produces evidence. If the estimate-request rate moves on the same traffic, the rest of the project has a business case that survives a bad month.
The short version
- Design for risk reduction, not service description
- Answer 'done it near me', 'what if it goes wrong' and 'what is the small next step' above the fold
- Organize proof by home type and neighborhood, not by date
- Fix the estimate path first and measure for 30 days before a full redesign
- Speed is read as competence; compress images and cut plugin bloat early
Wonder what your marketing looks like under the microscope?