How to improve your website conversion rate

Most business sites convert around 2% of visitors. Here's what actually moves that number: clarity, speed, trust and friction, minus the dark-pattern nonsense.

Updated 21 September 2026

On this page

Conversion rate is the share of visitors who do the thing you want: enquire, book, buy. Most business websites sit around 2%. Moving that number isn't magic or trickery. It's a handful of unglamorous things done properly. Here's what actually moves the number.

Say what you do in the first five seconds

The most common conversion killer is a homepage that makes people work out what you even offer. Visitors scan, they don't read. If your headline says "Transforming experiences for a connected world" instead of "Custom websites for Australian businesses", you've lost them.

Lead with the specific thing you do, who it's for, and what they get. One clear sentence beats a paragraph of mission statement. If a stranger can't tell within five seconds whether you're for them, nothing else on the page matters.

Make the next step obvious

Every page should have one clear primary action. Not five competing buttons, one. A visitor who's interested shouldn't have to hunt for how to contact you. The call to action should be visible without scrolling, repeated as the page gets long, and worded as a benefit ("Get a free website audit") rather than a chore ("Submit").

Decision paralysis is real. Two options convert better than ten.

Speed is a conversion feature, not a tech detail

Every extra second a page takes to load costs you visitors, and the drop-off is steep on mobile, where most Australians browse. Google's own research on why speed matters ties slower load times directly to higher bounce rates and lost conversions. A page that takes four seconds to appear has already lost a chunk of the people who clicked. If you've run a PageSpeed test and aren't sure what the numbers mean, here's how to read a website speed test before you act on it.

This is one reason we build custom sites rather than on WordPress (here's the full reasoning): plugin-heavy sites on shared hosting are slow by default. We serve pages from a global edge network, so they load fast everywhere, including for the customer checking you out on their phone at a traffic light. Fast sites convert better, full stop.

Earn trust on the page

People don't enquire with businesses they don't trust, and online they decide fast. Give them reasons.

Show real proof: specific results, named clients, genuine reviews. "Trusted by businesses Australia-wide" means nothing. "Lifted enquiries 3x for a named trades business" means something.

Show your prices. Hiding them behind "contact us for a quote" reads as evasive. Real numbers upfront convert the people who were never going to call just to find out whether they can afford you.

And show a human face. Who you are, where you're based, who actually does the work. Faceless sites feel like a risk.

Strip the friction out of forms

Every field you ask for is a chance to lose someone. The classic mistake is a contact form demanding name, phone, company, budget, and a 200-word project description before anyone's even spoken to you. Ask for the minimum you need to start a conversation, usually a name and one way to reach them, and get the rest later.

Same goes for checkout on an online store. Forced account creation, surprise shipping costs at the last step, and clunky payment forms are where carts die.

Can a small site run an A/B test?

Not usefully, in most cases. A site with a few hundred visitors a month cannot reach a result within any reasonable time, so count enquiries and ask customers instead. Take a site that converts at 2%, where you would be pleased to find a change that lifts it to 3%. That is a 50% improvement, larger than a button colour will produce. Evan Miller's sample size calculator puts the need at roughly four thousand visitors per version, so about eight thousand people through the site before the answer can be believed. At 350 sessions a month that is about two years, and a realistic effect of a few percent needs far more.

What usually happens instead is a fortnight-long test where one version is ahead and gets called. At that volume the result is noise, and decisions made from it do not move the number. Above a certain volume testing is the best tool there is, for example on an online store with real ad spend behind it. Below it, the useful work is observation.

  1. Count every enquiry for four weeks, in one place. A form submission recorded properly, calls from the site attributable, and one place where every enquiry lands. Until enquiries are counted when they arrive, you are optimising something you cannot see. Which leads came from your website, ads and Google Business Profile shows how to start with no tools.
  2. Ring the last five people who enquired, including one or two who did not buy, and ask what nearly stopped them. Five answers will not survive statistical scrutiny and are still the richest data about your own site. Typical findings are that visitors thought you worked only in the one suburb named on the site, or that they rang to ask something the site answered but buried.
  3. Watch some recorded sessions. Most analytics tools record real visits, and twenty minutes of watching a phone session shows where people give up.
  4. Fix the leaks that need no evidence. A slow site, a form that asks for a project description and a budget before anyone has spoken to you, and hidden prices are costing enquiries regardless of what a test would say. Why your website is slow and what it costs covers speed.

The catch is that these fixes come without a chart that has a winner on it. The evidence is a slow drift in enquiries over the following quarter. If counting shows traffic but few enquiries, the site is the problem. If it shows almost nobody landing on the site, conversion is not your bottleneck, which is the order of operations in what to do first for small business SEO.

This is also why a conversion rate optimisation consultant is often a poor fit for a small business. The discipline was built for high-traffic ecommerce, with test infrastructure and an iteration cycle. Applied to a few hundred visitors, it produces recommendations from the consultant's judgement rather than from your visitors' behaviour.

What we don't do

We don't use dark patterns. Fake countdown timers, manufactured scarcity, guilt-trip opt-outs. They nudge the number up short-term and torch your reputation. A high conversion rate built on clarity, speed and trust compounds. One built on tricks gets you a chargeback and a one-star review.

If your site is getting traffic but not enquiries, the leak is usually one of the five things above, and it's fixable without a full rebuild. See how we approach conversion and analytics.

Common questions

What is a good website conversion rate?
Most business websites sit around 2 per cent, so anything materially above that is doing well. The benchmark matters far less than the trend on your own site, because conversion rate varies enormously by industry, by traffic source and by what you decide to count as a conversion.
What improves a website's conversion rate fastest?
Saying what you do in the first five seconds, then making the next step obvious. Most sites leak at the top because the headline describes the company rather than the problem it solves, and leak again because no single action on the page is clearly the one to take.
Does page speed affect conversion rate?
Yes, directly. Every extra second before the main content appears costs you visitors, and the drop-off is steepest on mobile, where most Australians browse. It is also the one fix that helps every page at once, which is why it comes before rewriting copy.
Can you improve conversion without running A/B tests?
Yes, and below a few thousand visitors a month you have to: a test needs thousands of visits per variation before it says anything reliable. Observation replaces it. Count enquiries at the point they arrive, ring the last five people who got in touch, and fix the leaks that do not need evidence.

Want a website that turns readers into customers?

A custom site, hosting and upkeep for one monthly fee, with your leads and Google results in a dashboard you can check any time.