The Website That Looked Fine and Converted Almost Nothing: A 10-Point CRO Audit From 1.2% to 4.8%

Your website is generating traffic from Google, from ads, from word of mouth. But a frustrating number of those visitors are leaving without contacting you. That's a conversion problem, and it's almost always fixable.

Picture a mid-sized roofing and trades business operating across Auckland and the Bay of Plenty. Good reputation, steady Google Ads spend, roughly 1,000 website visitors a month and yet the phone barely rang because of it. Twelve enquiries a month. A 1.2% conversion rate. The owner's conclusion was that the traffic wasn't good enough. The website hadn't been checked in over a year.

This is a composite example based on the patterns we see across website conversion audits for NZ service businesses. It's not one specific client, but a realistic scenario built from the issues we repeatedly uncover. In many cases the website itself isn't the problem, it's how the web development has been implemented. Small, targeted improvements to performance, user experience, and conversion-focused web development often move the needle further than a full redesign. Here's what the audit uncovered, and what happened after those issues were fixed.

1. The Site Was Slow, and Nobody Waited to Find Out

Fix your page load speed first

The homepage took 5.8 seconds to load on mobile. A quick check at pagespeed.web.dev put the mobile score at 41. Portent's research shows conversion rates fall by roughly 4.42% for every second of load time between 0 and 5 seconds; most visitors arriving from paid ads were gone before the page had even finished rendering.

The fix was unglamorous: compressed images, deferred scripts, and a hosting upgrade. Nothing about the design changed. See how we approach this in our website performance optimisation services.

Tip 1: Fix page load speed first. It's the single biggest lever, and it's rarely a design problem.

Quick check: Curious what your own score is? Test it free at PageSpeed Insights it takes about 30 seconds.

2. Visitors Couldn't Tell What the Business Actually Did

Make your value proposition instantly clear

The homepage headline read “Quality Solutions for Every Roof.” Technically true, and useless. A visitor landing from a Google ad for “emergency roof repair Tauranga” had no way to confirm, in the first five seconds, that they were in the right place.

The headline changed to something a first-time visitor could act on immediately naming the service, the location, and the outcome, instead of a generic tagline. Nothing else on the page moved.

Tip 2: Say what you do, who you help, and what they get in the first five seconds not the third paragraph.

3. The Big Blue Button Went Nowhere Useful

Remove friction from your CTA

The CTA said “Contact Us,” sat below the fold on mobile, and led to a form with nine fields including budget, timeline, and “how did you hear about us.”

It became “Get a Free Roof Inspection,” moved above the fold, and the form dropped to three fields: name, phone, and a one-line description of the problem. Everything else moved into the follow-up call. The result lines up with a HubSpot study of over 40,000 landing pages, which found that cutting a form from four fields to three lifted conversions by nearly 50%.

Tip 3: Make the CTA specific and visible above the fold.

Tip 4: Cut the contact form down to what you actually need up front.

4. There Was Nothing to Trust

Add trust signals near your CTA

No reviews on the page. No past jobs shown. Just a stock photo of a roof that wasn't theirs.

A Google review count, three real job photos, and one named testimonial went in directly beside the new CTA. This is the same pattern that shows up across our case studies: trust signals work best when they sit at the exact point of hesitation, not buried on a separate “About” page.

Tip 5: Put proof next to the decision, not somewhere the visitor has to go looking for it.

5. The FAQ Section Didn't Exist

Address objections proactively

Every question a nervous first-time customer has about how much does this cost, how long will it take, do you actually turn up went unanswered anywhere on the site.

A short, honest FAQ block went in: starting prices, a simple three-step process, and realistic turnaround times. Visitors stopped needing to call just to ask questions the website could have answered.

Tip 6: A good FAQ section removes hesitation before it becomes a lost enquiry.

6. Nobody Knew Where People Were Actually Dropping Off

Use heatmaps to find where you're losing people

Until this point, every change was a reasonable guess. Installing Microsoft Clarity free, five minutes, no developer replaced guessing with evidence: most visitors scrolled past the CTA before it ever appeared, and the third form field was where most people abandoned.

That single piece of data confirmed two of the fixes above were the right ones, and ruled out a full redesign that had been under discussion for months.

Tip 7: Heatmaps tell you what's actually happening often not what you assumed.

7. One Headline Test Outperformed Everything Else Combined

A/B test your headlines

With the basics fixed, an A/B test ran two homepage headlines against each other for three weeks: the company-centric original against a customer-outcome version. The second won by 34%.

It was the single largest jump of the entire audit, bigger than the speed fix, bigger than the form change and it cost nothing to run.

Tip 8: Test headlines. The impact-to-effort ratio beats almost anything else on this list.

8. The Phone Number Wasn't Actually a Phone Number

Make your phone number click-to-call

On mobile, the header phone number was plain text not tappable. For a trades business, where a large share of buyers would rather call than fill in a form, that's a lead lost to a missing hyperlink.

One line of HTML wrapping the number in a tel: link fixed it in fifteen minutes.

Tip 9: Click-to-call on mobile is a small technical fix with an outsized effect on calls.

9. Every Page Was a Dead End

Improve your internal linking

Blog posts ended with nowhere to go. Service pages lacked relevant internal links, leaving visitors with no clear next step after they finished reading. As a result, many simply left the site.

Every page was updated with a relevant call to action and links to related content, including our SEO services for NZ businesses for those looking to improve their online visibility. A well-planned internal linking strategy guides visitors to the information they need and helps search engines understand how a site is structured strong internal links distribute authority, improve crawlability, and support better rankings while creating a smoother user experience.

Tip 10: No page should be a dead end. Give every visitor an obvious next step.

Run the Same Audit on Your Own Site

Use this as a checklist against your own website:

• Mobile PageSpeed score above 70
• Headline names what you do, who you help, and the outcome within 5 seconds
• Primary CTA is specific, visible above the fold, and low-commitment
• Contact form has 4 or fewer fields
• A trust signal sits directly next to the CTA
• FAQ section covers price, process, and timeline
• Heatmaps installed (Microsoft Clarity free)
• Headlines are being tested, not assumed
• Phone number is click-to-call on mobile
• Every page has an obvious next step

Small improvements can create meaningful results without rebuilding your entire website.

Final Thoughts

None of the ten fixes above required a rebuild. Most were hours of work, not weeks: a headline rewrite, a shorter form, a tappable phone number, a free heatmap tool. What changed the result wasn't a bigger marketing budget or more traffic it was removing the friction that was already quietly costing enquiries every single day.

That's the pattern worth taking from this: conversion rate optimisation is rarely about doing more. It's about noticing what's currently stopping visitors from taking the next step, and clearing it. Run the checklist above against your own site, fix what you find, and keep testing conversion rate isn't a one-time project, it's an ongoing habit.