websites conversion rate optimisation

The most common conversation I have with owners running ads goes like this: "The clicks are cheap enough. The leads just aren't there." Nine times out of ten, the problem isn't the ad account. It's the 8 seconds after the click, on the landing page.

The maths of website conversion is unforgiving. A site converting at 2% needs twice the traffic — twice the ad spend — of a site converting at 4%. Doubling conversion rate is often easier than doubling traffic, and it's always cheaper. The seven fixes below are the ones that keep showing up in every conversion audit I do, ranked by the ratio of effort to impact.

None of them require a rebuild. All of them can be done inside a fortnight. Most owners see a measurable lift within the first month, and the compounding effect on ad ROI usually pays for the changes ten times over inside a quarter.

Before the fixes: measure what's actually broken

Every conversion project should start with three numbers: current conversion rate, mobile-vs-desktop split, and where users drop off (a heatmap or scroll-depth report). Without those you're guessing. GA4 handles conversion tracking, Microsoft Clarity is free and gives you the heatmap. Twenty minutes of setup, then you'll know which of the seven fixes below will move the biggest number.

The 7 fixes, ranked

1. 1Speed. Compress hero images. Lazy-load everything below the fold. Get LCP under 2.5s. Nothing else matters if this is broken.
2. 2One clear H1 that names the outcome, not the service. "Get quoted in 30 minutes" beats "Sydney's #1 plumbing service."
3. 3Phone number top-right on every page, tappable on mobile.
4. 4Social proof above the fold — real photo, real name, real number outcome.
5. 5A form under 5 fields. Every extra field cuts conversion 8-10%.
6. 6Sticky CTA on mobile — the button follows the scroll.
7. 7Trust bar — insurance logos, warranty, years in business — in the hero band.

Fix 1 in depth — speed

Every second of load time above 2.5s on mobile costs you 10-15% of conversions. The two culprits in 90% of the sites I audit: an oversized hero image (5MB when it should be 200KB) and third-party scripts loading synchronously (chat widgets, tag managers, analytics duplicates). Compress images to WebP under 300KB, defer non-essential scripts, and cache aggressively. Owners who do just these two things typically see a 20-30% conversion lift with no other change.

Fix 2 in depth — the H1 that names the outcome

"Sydney's #1 plumbing service" is a claim. "Emergency plumber to your door in 60 minutes — flat-rate quote before we start" is an outcome. Visitors don't buy claims; they buy outcomes they can picture. Write the H1 as the sentence a customer would say to a friend describing what they just got from you.

Fix 3 in depth — the phone number that never hides

Around 55% of service-business enquiries still come by phone. If the number is hidden in a footer or requires a click to reveal, you're forcing every phone-preferring customer through friction. Top-right, tappable, always visible, on every page. This one change alone routinely lifts total enquiries by 15-25% on sites that had the number buried.

Fix 4 in depth — social proof that isn't stock

"John S., very happy customer" convinces no one. "Sarah in Bondi — 'Fixed our leak in 45 minutes, quoted $180, charged $180'" convinces almost everyone. Real names, real suburbs, specific outcomes. If you can add a photo of the customer or the completed job, better again. If you have Google reviews, pull the actual review text through with schema so it appears both on-page and in search results.

Fix 5 in depth — the five-field form

Every field is a chance for a visitor to give up. Name, phone, suburb, service, message — done. Job title? Company size? Preferred contact time? Cut them. If your CRM demands them, capture them on the follow-up call. The primary job of the form is to get you the phone number so you can have the qualifying conversation.

Fix 6 in depth — the sticky mobile CTA

Mobile visitors scroll. Desktop visitors scan. A sticky CTA bar at the bottom of the mobile viewport — "Call now" plus "Get a quote" — puts the action button under the visitor's thumb regardless of where they are on the page. This is the single easiest change on the list and it typically lifts mobile enquiries by 10-20% on its own.

Fix 7 in depth — the trust bar

In the hero band, before anything else scrolls into view: license number, insurance status, years in business, satisfaction guarantee, industry association memberships. This addresses the anxiety of "is this a real, accountable business" that every serious buyer has before they enquire — and it kills the objection before it forms.

What to stop doing

• Sliders / carousels (nobody clicks the second slide)
• Stock photos of people who don't work at your business
• "Contact us" as your main CTA (name the outcome instead)
• Video backgrounds that autoplay
• Cookie banners that block the fold

Sliders in particular have been publicly killed by every serious CRO study for a decade and are still on 40% of the sites I audit. They kill your LCP score, they hide your best offer behind a click no one takes, and they signal "designed in 2014." Replace them with a single strong hero and move on.

The 14-day conversion sprint

1. 1Day 1-2: Install heatmap + conversion tracking. Get baseline numbers.
2. 2Day 3-5: Speed fixes. Compress images, defer scripts, cache aggressively.
3. 3Day 6-8: New H1, trust bar, phone number top-right and sticky.
4. 4Day 9-11: Rewrite form to five fields. Add real social proof.
5. 5Day 12-14: Ship. Measure the new conversion rate against baseline.

Two weeks of focused work, no rebuild required, and most sites in the client book lift conversion by 40-80% by the end of the sprint. The site pays for itself in the first month of the higher conversion rate feeding back into cheaper ad performance.

When it really is time for a rebuild

• The current site is more than 5 years old and built on a platform your developer can't easily edit.
• Mobile performance is genuinely unfixable (chronic layout shift, unresponsive templates).
• The brand and offer have moved on so far that the site no longer describes the business.
• SEO is being actively hurt by the technical foundation (bad Core Web Vitals, thin content templates you can't extend).

Even then, do a two-week conversion sprint on the existing site first. It buys you cashflow to fund the rebuild and gives the new site a benchmark to beat.

Frequently asked owner questions

How much conversion lift can I realistically expect?

Starting from a typical small-business site (~2% conversion), a proper CRO pass usually gets to 5-8% within a month. Top-quartile service-page performance sits at 10-15%. Beyond that requires a strong offer, mature brand and heavy social proof.

Do I need a designer or a developer?

Both, briefly. A CRO-focused designer to draft the new layout, a developer to implement it cleanly, and someone who can measure the result. Or hire an agency who does all three end-to-end.

Should I A/B test everything?

Only once you have enough traffic to run tests to significance — usually 5,000+ visitors per variant per month. Below that, just ship changes based on best practice and measure the aggregate effect over 30-60 days. Small-business A/B testing is mostly theatre.