Blog · conversion

12 ways your website talks people out of hiring you

Updated September 28, 2026 · about 18 min read

It happens all the time, and usually by accident — a well-meaning designer who never had to make a stranger pick up the phone. Here's where, and here's the fix.

Why good designers build sites that lose customers

This isn't a rant about bad designers. Most of the sites we fix were built by decent, well-meaning people who know how to make something look good. They just weren't taught what makes a stranger pick up a phone, because that isn't a design skill. It's a sales skill.

A designer's instinct is to make the brand look impressive: a big photo, the logo, a tagline, a menu. A salesperson's instinct is to answer the question the customer walked in with: do you fix my thing, for people like me, and what happens next? Those two instincts produce different first screens, and only one of them converts.

The result is a site that quietly talks people out of doing business with you — not with anything wrong, exactly, but with nothing right at the moment it matters. Twelve of those moments are below. We see them in this order, on most sites, most weeks.

1. It introduces you instead of them

The headline says your name, your founding year and a word like 'quality.' The visitor wanted to know if you fix the thing they have. Say that first. 'Roof leaking? Photos of the problem first, then a price' converts; 'Welcome to Smith Roofing' introduces.

2. It hides the phone number

On a phone the header is prime real estate and the footer is a rumor. If the number isn't tappable in the first screen, some people won't hunt for it. They'll tap the next result instead.

3. It offers five things to do

Learn more, about us, our services, gallery, contact. Five good options produce fewer decisions than one clear one. Choice overload is one of the most replicated findings in the field. One button, one color, everywhere.

4. It keeps the proof on another page

Two hundred reviews on a 'Testimonials' page reach almost nobody. One review beside the button reaches everyone at the moment of decision. Proof belongs next to the ask.

5. It won't say a price, or even a range

'Contact us for a quote' reads as 'expensive, and we'll pressure you.' A range — 'most jobs $X–$Y' — anchors the visitor and filters the wrong ones out. Price conditioning is a sales skill; the site can do it for you.

6. It loads like it's 2009

Google's own mobile research shows bounce probability climbing steeply with each second of load time. Nobody reads a page that hasn't appeared. Heavy sliders, uncompressed photos and a dozen plugins are the usual culprits.

7. It asks for eleven things on the form

Every field costs completions. Ask for a name and a phone number; ask the rest on the call. The form friction calculator on this site shows what each field costs at your traffic.

8. It uses words every competitor uses

'Quality.' 'Trusted.' 'Solutions.' 'Best in class.' If the sentence could be on a competitor's site unchanged, it says nothing. Specific beats superlative: 'photos of every soft spot before we quote' is unarguable.

9. It looks like it was built for a desktop

Most visitors are on a phone. Tiny text, a hero image that pushes everything below the fold, buttons too small for a thumb. Design for the phone first; the desktop will be fine.

10. It has a slider nobody watches

Rotating banners let a committee avoid choosing one message. Visitors don't wait for slide three. Pick the one thing you want them to know and put it in the headline.

11. It talks about features, not the fear

Customers aren't buying a 50-year shingle; they're buying not lying awake during a storm. Name the worry, then show the fix. Marketing psychology calls this loss framing, and it moves people.

12. It goes quiet after the form

They sent the form. Nothing happened for two days. They hired someone else. The first hour decides whether a lead becomes a customer; the site should promise a response time and the business should keep it.

How to check your own site in ten minutes

Open your site on your phone — not the office computer. Pretend you've never seen it. Walk the list:

1. It introduces you instead of them?
2. It hides the phone number?
3. It offers five things to do?
4. It keeps the proof on another page?
5. It won't say a price, or even a range?
6. It loads like it's 2009?
7. It asks for eleven things on the form?
8. It uses words every competitor uses?
9. It looks like it was built for a desktop?
10. It has a slider nobody watches?
11. It talks about features, not the fear?
12. It goes quiet after the form?

Count the yeses. Most sites we see have five to eight. Each one is a specific, fixable thing, and the fix is rarely a redesign. It's a headline. A phone number moved up. Four buttons deleted. A review moved next to the form. Six fields removed. A photo compressed. A promise to call back within the hour, and someone who keeps it.

If you'd rather have a stranger do the walk, the free conversion audit is exactly that — with the findings ranked by what they're likely costing you. You keep it either way.

The three most common combinations

Leaks travel in packs. Three combinations show up on most of the sites we audit.

The "brochure": leaks 1, 3, 4 and 10

A slider, a company-name headline, five menu buttons, and the reviews on their own page. It looks like every other business in town because it was built from the same template as every other business in town. The fix is a single first screen: problem headline, one button, one review, no slider. It takes an afternoon.

The "fortress": leaks 2, 5, 7 and 12

No phone number in sight, "contact us for a quote," an eleven-field form, and silence afterward. This site is protecting the business from its customers. The fix is to lower the wall: number in the header, a price range, a two-field form, and a promised response time that someone actually keeps.

The "gallery": leaks 6, 9 and 11

Beautiful, huge photos that take eight seconds to load on a phone, text sized for a monitor, and copy about materials instead of the customer's worry. Common in remodeling, landscaping, photography and restaurants. The fix is compression, a phone-first layout, and a headline that names the fear ("Tired of the bathroom you're embarrassed to show guests?").

Naming the pack matters because fixing one leak in a pack often does little — a great headline above a hidden phone number still doesn't ring the phone. Fix the pack.

What the fixes look like, before and after

Concrete beats abstract, so here are the rewrites we make most often, in the customer's words.

Headline, before: "Welcome to Mountain View Dental — Comprehensive Family Dentistry." After: "Tooth pain today? We hold same-day appointments for new patients."

Button, before: "Learn More." After: "Book my same-day visit."

Offer, before: "Contact us for a free quote." After: "Get a price in one call — most jobs $4,000–$9,000."

Proof, before: a "Testimonials" menu item. After: "★★★★★ 'They showed me photos of every soft spot before quoting.' — Maria R." directly under the button.

Form, before: name, email, phone, address, city, zip, service, roof type, roof age, insurance carrier, how did you hear about us. After: name, phone. Everything else on the call.

Response, before: "Thank you for your submission." After: "Got it — Mike will call you within the hour." And Mike does.

Every one of those is a copy change, not a code change. That's the point: the site isn't broken. It's just not saying the right thing at the right moment.

Why this is a sales problem, not a design problem

Our founder ran sales teams for multi-million-dollar plumbing and remodeling companies before he ran an agency, and it shows in how we look at a page. A website is a salesperson who never sleeps and never gets to ask a question. Everything a good salesperson does in a first conversation, the page has to do in its first screen.

A salesperson names the problem ("so the roof's leaking near the chimney?"). The headline does that. A salesperson removes risk ("we'll take photos first, no charge"). The offer does that. A salesperson proves it ("here's what we found on your neighbor's roof last week"). The proof does that. A salesperson asks for the next step ("can I come by Thursday?"). The button does that. And a salesperson follows up — the first hour does that.

Tie-downs, trial closes and price conditioning have on-page equivalents too. "Most roofs in your neighborhood are twenty years old, aren't they?" is a tie-down. "How much would it be worth to stop worrying about the next storm?" is a trial close. "Most jobs run $8,000–$14,000" is price conditioning. We write pages with those tools because the visitor is having a sales conversation whether or not anyone is there to have it with them. Our sales training teaches the human version; the page should already be doing the written one.

A note on platforms

None of the twelve leaks is caused by a platform, and none is fixed by one. A hand-coded site can hide the phone number; a Wix site can put it in the header. The platform decides how fast the page loads and how hard the fixes are; it never decides whether they matter.

That said, leak six — speed — is where platforms show their character. Hand-coded pages are fast by default. Webflow is fast with care. WordPress and Shopify are fast with discipline. Wix and Squarespace are close on a good day. Our comparison, hand-coded vs WordPress vs Shopify vs Webflow, goes through it honestly, and our platform pages say what we can fix on each: Shopify, WordPress, Webflow, Wix, Squarespace.

What waiting costs, in this context

Every leak on this list is a percentage of visitors who almost called. Add them up and most sites are losing a point or two of conversion to things that cost an afternoon to fix. Our 1% lift article does the arithmetic; the short version is that for a business with a few thousand visitors and a customer worth a few thousand dollars, a single point is tens of thousands a month. Multiply by the months you've meant to fix the site.

That number is the honest frame for the decision, and it's why we lead with the free audit rather than a proposal: once you've seen the leaks ranked and what each is likely costing, the decision makes itself — sometimes toward us, sometimes toward your own designer with a list. Either is fine. Leaving the door locked isn't.

The twelve, applied to one real industry

Abstract lists are easy to nod at and hard to use, so here's the whole list applied to a business we know well: a bathroom remodeling company, the kind our founder has run.

1. Introduces you instead of them: "Welcome to Premier Bath Solutions, serving the Front Range since 2012." The customer is standing in a bathroom they're embarrassed by, or worried their mother will fall in. Say that: "A safer, prettier bathroom in one day — walk-in tubs and barrier-free showers, installed by our own crew."

2. Hides the phone number: a remodeling customer is often older and calling from a phone. The number goes in the header, big, tappable, with "call for a free virtual estimate" beside it.

3. Five things to do: Products, Gallery, Financing, About, Contact. Replace with one button: "Get my free estimate from a photo."

4. Proof on another page: 60 five-star reviews on a Reviews page nobody visits. One of them — "installed in a day, my dad can shower alone again" — goes under the button.

5. Won't say a price: "Every project is different." True, and useless. "Most walk-in tub installs run $X–$Y, and we'll tell you where yours lands on the first call." That's price conditioning, and it filters tire-kickers while reassuring real buyers.

6. Loads like 2009: a 4 MB hero photo of a shower. Compressed to 120 KB, it looks identical and loads in a second.

7. Eleven-field form: name, phone, "what would you like to change?" Everything else — measurements, tub type, financing — belongs in the estimate tool or the call.

8. Words every competitor uses: "quality craftsmanship," "customer satisfaction." Replace with the unarguable: "our own crew, not subcontractors; one day; a photo estimate before anyone visits."

9. Built for a desktop: the older customer is on an iPad or a phone. Larger text, bigger buttons, nothing that needs a hover.

10. A slider: gone. One picture of one finished bathroom, with a real customer's name.

11. Features, not the fear: "acrylic wall systems with lifetime warranty" becomes "no more stepping over a tub edge — and no more worrying about the fall."

12. Goes quiet after the form: "Thanks!" becomes "Got it — Dana will text you within the hour to set up your free virtual estimate." And Dana does.

That's one afternoon of copy, one hour of image compression, and one conversation with Dana. No redesign, no new platform. The door opens.

How to write the headline (the one thing that moves most)

If you fix only leak one, you've done most of the work, so here's the method we use to write a first-screen headline.

1. Write down the moment the customer searches. Not the demographic — the moment. "It's 9 p.m., the water heater is leaking, the kids need baths." "My mom slipped last week and I'm terrified." "The roofer's quote seemed high and I don't know why."
2. Name the problem in their words. Use the words they'd say to a friend, not the words on your invoice. "Water heater out?" not "Water heater replacement services."
3. Promise the result and remove the risk. "Same-day repair, price before we start." Result plus risk removal in one line.
4. Make it unarguable. If a competitor could put the same sentence on their site, sharpen it until they couldn't. "Photos of every soft spot before we quote" is a claim only a company that does that can make.
5. Read it aloud on your phone. If it takes more than four seconds to say, cut it.

Three headlines we'd put on real sites tomorrow: "Roof leaking? Photos of the problem first, then a price." "Tooth pain today? Same-day appointments for new patients." "Arrested in Denver? Here's what happens in the next 48 hours — and what to do now." Notice none of them contains a company name. The name goes in the logo, where the customer will find it after they've decided to keep reading.

Measuring whether the fixes worked

Fix the twelve and you'll feel the phone ring more. Feel isn't a number, so here's how to know.

Before you change anything, record one full month: visitors (by source), calls, forms, chats, bookings. That's your baseline, and it's the thing most businesses skip and then regret.

Change the first screen and the form in the same week — they're one fix — and leave everything else alone for a month. Compare conversions to baseline. If traffic changed a lot in the meantime, compare rates, not counts.

Then speed, measured before and after on PageSpeed Insights. Then the follow-up promise. One change per month, each measured against the last.

If you have more than a few thousand visitors a month, you can do this properly with a split test — half the visitors see the old first screen, half the new — and our A/B significance calculator tells you when the difference is real. Below that traffic, before-and-after months are honest enough, as long as you write the numbers down and don't move the goalposts.

The businesses that keep doing this — one fix, one measurement, one month — end up with sites that quietly out-perform everyone in their category. Not because of a redesign. Because of twelve afternoons.

Fifteen small things that also cost calls

The twelve are the big ones. These are the small ones we fix while we're in there — each worth a fraction of a point, and fractions add up.

• A "Home" link in the menu that goes nowhere useful; make the logo the home link and give the slot to the offer.
• Hours that aren't on the page, or are wrong. If you're open Saturdays, say so in the first screen — it's a differentiator.
• A map with no "Get directions" button. On a phone, the button is the conversion.
• A chat widget that says "leave a message" at 2 p.m. on a Tuesday. Either staff it or use a scripted assistant that answers and hands off.
• Stock photos of people who are obviously not your crew. Customers can tell. One real photo of one real employee beats ten handsome strangers.
• The word "submit" on the button. Nobody wants to submit. "Get my price," "Book my visit," "Call me back."
• Address without a service area. "Denver" is not "we come to Arvada." Say where you go.
• A PDF menu, price list or brochure on a phone. Put it on the page as text.
• Autoplay video with sound. It's the fastest way to make someone close a tab in a waiting room.
• A cookie banner covering the phone number. Move the banner or shrink it.
• A "we've moved to a new site!" post from three years ago. Delete it.
• Broken links in the footer. Machines notice, and so do careful buyers.
• A 404 page with no phone number. Even lost visitors can call.
• Reviews with no names or dates. "Great service — J." reads as invented. Use first name and last initial and the month.
• Nothing that says what happens after they call. "Here's what to expect on your first call" removes the last bit of fear.

None of these is worth a project on its own. Together they're an afternoon, and they're the difference between a site that feels like a business and one that feels like a template. They're also on the free audit checklist, ranked, so you know which ones matter for you.

One more, because it's the one we're most often asked about: the assistant. A scripted guide like Scout on this site isn't a gimmick when it's built right — it answers the question the visitor was too shy to call about, at 11 p.m., and hands them to a human in the morning. It's leak twelve, solved for the hours you're asleep. We build them for clients; the roofing and clinic sites in our work each have one, and they're trained on the business, not on the internet.

What we'd do first

If we could change one thing on most sites, it would be the first screen: the headline names the problem, the phone number sits in the header, one red button says what happens next, and one line of proof sits beside it. That single screen fixes leaks one, two, three, four and eight at once. Then the form. Then speed. Then the follow-up.

It all starts with strategy — and then actually starting. Being found is the other half, handled by our sister brand Eye To Ad Media; there's no point being found by people your site then talks out of calling. Fix the door first. Then open the road to it.

Questions people ask