Why Is Your Website Not Converting?
And how to fix it, step by step.

What does "converting" actually mean for your website?

Before I fix a website that isn't converting, I ask one boring question: converting into what? Most sites I audit can't answer it, and that's usually the first leak.

When a business owner tells me "my website isn't working," they usually mean the phone isn't ringing. But the site might have six different buttons, three forms and a chat widget, and nobody knows which one is supposed to matter. If you don't decide what a win is, you can't tell if you're winning.

Primary and secondary conversions

I split conversions into two groups. The primary conversion is the money action: a quote request for a roofer, a consultation booking for a lawyer, a trial sign-up for software, a purchase for a store. Secondary conversions are smaller steps that show interest: downloading a guide, joining an email list, using a price calculator, watching a video.

What a conversion rate is

Conversion rate is conversions divided by visitors. If 1,000 people visit a page and 20 request a quote, that's a 2% conversion rate (example numbers). The number on its own means little. What matters is whether it goes up after you change something, and whether the leads are any good.

I'm careful with benchmark numbers you'll see online. They mix industries, traffic sources and definitions of a "conversion." A local emergency plumber getting search traffic will convert very differently from a software company getting cold social traffic. Your own past results are the benchmark that counts.

First check: is it a traffic problem or a page problem?

Half the "my website isn't converting" calls I get aren't about the website. They're about who is arriving. This is the first fork in the road, and getting it wrong wastes months.

Diagnose your site: the triage table

This is the table I use on a first audit call. Find the symptom that sounds like you, then jump to the part that covers it.

Where to look

• Visitors per month: if it's tiny, conversion rate swings wildly and tells you nothing yet.
• Traffic sources: Google search, ads, social, direct and referrals each behave differently.
• Landing pages: which pages people arrive on first. Often it isn't your homepage.
• Device split: if most visitors are on phones, judge your site on a phone.
• Engagement: how long people stay and how far they scroll on key pages.

Wrong visitors: message and audience mismatch

You can't convert people who were never going to buy. A lot of "low conversion" is actually the right page meeting the wrong crowd, or the right crowd meeting the wrong page.

Common mismatches I see

• Ranking for information searches ("how to unclog a drain") and expecting emergency calls.
• An ad promising a discount that the page never mentions.
• A local service with no clear service area, so out-of-town visitors bounce or waste your time.
• A premium brand attracting bargain hunters with "cheap" keywords.
• One generic homepage for every ad, post and email.

Message match

Message match means the page repeats the promise that got the click: the same words, the same offer, ideally the same look. If someone clicks "Free roof check before storm season," the page should say "Book your free roof check before the next storm" in the headline, not "Welcome to Ridgeline Roofing, family owned since 1998."

The 5-second test: is your headline and offer clear?

Show your homepage to someone for five seconds, then hide it. Ask what you do, who it's for and what they should do next. If they can't answer, that's your conversion problem.

A headline formula I use

[Result the customer wants] + [for who / where] + [without the thing they fear]. For example: "Fast, upfront-priced AC repair in Denver, without the surprise bill." It isn't poetry. It works because a stressed homeowner understands it instantly.

The offer is more than the headline

An offer answers: what do I get, what does it cost me (money, time, effort), and what happens next? "Contact us" is not an offer. "Get a fixed-price quote in 24 hours" is. If your offer is weak, no design will save it.

Too many choices: competing CTAs and cluttered navigation

Every extra link is a door out. When a page asks visitors to call, book, chat, follow on Instagram, read the blog and sign up for a newsletter, most people pick the easiest option: leaving.

What I usually remove

• Social media icons at the top of the page (move them to the footer).
• Menus with 9+ items. Group them or cut them.
• Pop-ups that fire the second the page loads.
• Sliders that rotate three different messages.
• Two or three different button colors fighting each other.

Homepages vs landing pages

A homepage serves many visitors, so it can offer a few paths: "Get a quote" as the main button, "See our work" as a quieter link. A landing page for an ad should have one goal, often no main menu at all. That's why I build separate landing pages for ad traffic instead of sending clicks to the homepage.

Weak or hidden calls to action

I've audited sites where the only way to get in touch was a tiny "Contact" link in the top corner. The business owner thought it was obvious. Visitors didn't.

• Put the main CTA in the first screen on desktop and phone.
• Repeat it after each major section, so people never have to scroll back up.
• On phones, a sticky bottom button or click-to-call bar works well for local services.
• Add one line under the button that lowers risk: "No obligation. Takes 60 seconds."

No trust: missing proof, faces and guarantees

People don't hand over their phone number to a website they don't trust. Trust is a conversion element, not a nice extra. And no, the fix is not fake reviews.

Trust signals that work

• Real customer reviews, with names (with permission) and specifics.
• Photos of the owner and team, not stock smiles.
• Before/after photos or case studies of real jobs.
• Licenses, insurance, certifications and memberships.
• Clear contact details: phone, email, address or service area.
• A specific guarantee: "If we can't fix it, you don't pay the call-out fee."

If you're new and have no reviews yet

I'm in this position myself with my own site, so I don't pretend. Show your work instead: demos, process, photos, credentials. Explain exactly what happens after someone contacts you. Ask your first clients for honest reviews and add them as they come. A clear, honest page beats a page stuffed with suspicious five-star quotes.

Slow pages and Core Web Vitals

A slow page doesn't feel slow to you because you've visited it a hundred times on office Wi-Fi. Your customers are on a phone, in a parking lot, with one bar of signal.

Core Web Vitals in plain English

• Use modern image formats (WebP/AVIF) and export at the size you display.
• Remove tracking and chat scripts you don't use.
• Avoid autoplay video backgrounds on mobile.
• Lazy-load images below the first screen.

Mobile problems that kill leads

For most local businesses I work with, the majority of visitors are on phones. Yet most owners review their site on a laptop. That gap hides a lot of lost leads.

My mobile checklist

• Body text at least 16px; no pinching to read.
• Buttons big enough for a thumb, with space between them.
• Phone number is a tap-to-call link.
• Headline and main CTA visible without scrolling.
• Forms use the right keyboards (numbers for phone and ZIP).
• No pop-up that's hard to close.
• Menu is easy to open and close.

Forms that scare people off

Forms are where conversions actually happen, and where many of them die. People start, see question seven, and quietly close the tab.

How I slim down a form

1. List every field and ask: do I need this before the first call?
2. Delete or postpone anything that isn't essential (budget, company size, how did you hear about us).
3. Start with easy, low-risk questions ("What do you need help with?") before contact details.
4. Use buttons or picks instead of typing where possible.
5. Show a friendly error message right next to the field, not at the top.
6. Say what happens next: "I'll reply within one business day."

Multi-step forms often feel shorter even when they ask the same number of questions, because each step is small and the first one is easy. That's the idea behind the quiz in my roofing funnel demo.

Pricing confusion and hidden costs

"Call for pricing" is one of the most common reasons visitors leave without calling. Not because they can't afford you, but because they're afraid of the awkward conversation.

• "Most AC repairs fall between X and Y" (use your real ranges).
• Packages with clear inclusions: basic, standard, premium.
• A simple calculator that gives a ballpark before the form.
• A list of what changes the price: size, materials, urgency.
• Financing options, if you offer them.

Copy that talks about you instead of the customer

Count how many sentences on your homepage start with "We." Then count the ones that start with "You." Most sites I audit are heavily weighted to "We."

A simple page story

1. The problem in their words.
2. The result you deliver.
3. How it works, in three steps.
4. Proof: reviews, photos, credentials.
5. Answers to their worries (FAQ).
6. The offer and CTA again.

Design that looks dated or untrustworthy

Design isn't decoration. Visitors judge credibility from how a site looks, very quickly. A dated or messy site makes even a great business look risky.

• Too many fonts or colors fighting each other.
• Low-contrast grey text that's hard to read.
• Stock photos that look like every other site.
• Walls of text with no headings or breaks.
• Broken links, old copyright years, blog posts from years ago.

No path for "not ready yet" visitors

Most visitors aren't ready to buy today. If your only option is "Book a call," you lose everyone who's still researching, which is most people.

1. VisitorBrowsing, not ready
2. Small stepQuiz, calculator or guide
3. Follow-upEmail or SMS with useful info
4. ReadyBooks or buys when it's time

Missing or broken tracking: you can't fix what you can't see

On most audits, the first thing I find isn't a design problem. It's that the business has no reliable way to tell how many leads the site actually produces.

• GA4 installed on every page, ideally through Google Tag Manager.
• Key events: form submit, thank-you page view, click-to-call, booking, purchase.
• Phone calls tracked (tap-to-call clicks at minimum).
• UTM tags on ads, emails and social links.
• A test lead submitted every time you change the form.

Reading the data: GA4, heatmaps and session recordings

Numbers tell you where people drop off. Recordings tell you why. I use both before I change anything.

• People clicking on things that aren't links (they expect them to be).
• Most visitors never scrolling past the first screen.
• Rage clicks on a broken button.
• Visitors hovering over the price, then leaving.

Slow follow-up: leads that convert, then go cold

Sometimes the website converts fine. The leads just sit in an inbox for two days, and by then the customer has hired someone else.

1. Form submittedInstant thank-you page with next steps
2. Within minutesAuto SMS / email confirmation
3. Same dayPersonal call or reply
4. Days 1-7Helpful follow-up until they book or reply

• Send leads to a CRM or shared inbox, not one person's email.
• Use the thank-you page to offer booking right away.
• Stop automated messages once someone books or replies.

A 30-day fix plan, in priority order

Here's the order I work in when a client says their website isn't converting. Quick, cheap fixes first; big rebuilds only if the basics don't move things.

After 30 days, compare leads and conversion rate with the month before (keeping in mind traffic changes and seasons). If things barely moved and the data shows deeper problems, that's when a rebuild or a dedicated landing page makes sense.

A/B testing without fooling yourself

A/B testing sounds scientific, and it can be. But with small traffic it often produces confident-looking answers that are just noise.

• Test big things first: offer, headline, page layout, form length. Not button shades.
• Decide what success means before starting.
• Run for at least one or two full weeks.
• Don't stop the moment one version looks ahead.
• Check lead quality too, not just lead count.

Before/after walkthroughs with my demos, and when to hire help

To make this concrete, here's how I'd apply this guide to three kinds of pages, using demos I built. All brands are fictional, and the "before" problems are typical ones I see, not real clients.

Local service lead page

Typical before: a generic homepage, a tiny contact link, a 10-field form and no prices. After: a problem-focused headline, a four-field estimate form in the first screen, real-looking trust strip, a repair-or-replace helper and a sticky mobile call bar.

Software trial page

Typical before: feature lists and jargon, a "Request a demo" button and no pricing. After: an outcome headline, email-only sign-up, feature tabs that show the product and a clear pricing toggle.

Ad-to-booking funnel

Typical before: ads pointing to a homepage with no tracking. After: an ad, a matching opt-in page, a three-question quiz, booking on the thank-you page and automatic SMS and email follow-up, with tracking at every step.

When to hire help

• You've fixed the basics and conversions still haven't moved.
• Your site is hard to edit, slow or built on an old platform.
• You're about to spend real money on ads.
• You don't have time to set up tracking and follow-up properly.

When you hire someone, ask to see pages they've built, how they'll measure success and what they'll change first. Be cautious of anyone guaranteeing a specific conversion rate. I don't, because results depend on your traffic, offer and follow-up as much as the page.

My conversion fix checklist

• One primary conversion named for each key page
• GA4 and conversion events tested with a real lead
• Traffic sources and device split checked
• Headline passes the 5-second test
• One main CTA, visible on phones without scrolling
• CTA text describes the result
• Real reviews, photos and credentials (nothing invented)
• Page loads fast on mobile
• Tap-to-call phone number
• Form cut to essential fields, with next steps explained
• Price ranges or starting prices shown
• A smaller step for visitors not ready yet
• Instant auto-reply and same-day follow-up
• Monthly test lead scheduled

Conversion terms in plain English

A/B test

Showing two versions of a page to similar visitors to compare results.

Above the fold

What's visible before scrolling.

Bounce

A visit where someone leaves without interacting.

Call to action (CTA)

The button or link asking visitors to take the next step.

CLS

Cumulative Layout Shift: how much a page jumps while loading.

Conversion

A valuable action: a call, lead, booking, sign-up or purchase.

Conversion rate

Conversions divided by visitors.

Conversion rate optimization (CRO)

Improving a site so more visitors take the key action.

Core Web Vitals

Google's measures of loading, responsiveness and visual stability.

CRM

Software that stores leads and tracks them to a sale.

Drop-off

The point where visitors abandon a page, form or funnel.

Event

A tracked action in analytics, like a form submit.

Friction

Anything that makes the next step harder.

GA4

Google Analytics 4.

Google Tag Manager

A tool for managing tracking tags in one place.

Heatmap

A visual of where visitors click and scroll.

Hierarchy

Design that makes the most important things stand out first.

INP

Interaction to Next Paint: how fast a page responds to taps.

Landing page

A page built for one offer and one action.

LCP

Largest Contentful Paint: how fast the main content appears.

Lead

A person who shared contact details or asked to be contacted.

Lead magnet

Something useful offered in exchange for contact details.

Message match

A page repeating the promise of the link that brought the visitor.

Micro-conversion

A small step showing interest, like using a calculator.

Multi-step form

A form split into short steps.

Offer

What a visitor gets, what it costs them and what happens next.

Session recording

An anonymous replay of a visitor's session.

Social proof

Real reviews, testimonials and results from others.

Speed to lead

How quickly you reply to a new lead.

Sticky CTA

A button that stays visible while scrolling.

Trust signal

Anything that shows the business is real and reliable.

UTM parameters

Tags on links that show where traffic came from.

Value proposition

A clear statement of the result you deliver and for whom.

Frequently asked questions