What to measure, where visitors drop off, and which fixes are worth making first
A professional services firm with 40 staff spends on search and content, watches the traffic graph climb, and still fields the same number of enquiries it did two years ago. The marketing report looks healthy. The sales pipeline doesn’t. Somewhere between a visitor arriving and a visitor making contact, something is falling over, and nobody in the business can point to where. If you want to improve website conversion rate without simply buying more traffic, the first job is to work out which part of that journey is failing and why.
In this article
What a conversion rate tells you
Your conversion rate is the share of visitors who do the thing you want: submit an enquiry, book a call, download the capability statement, start a quote. The figure itself matters less than what sits behind it, because an average across the whole site hides almost everything useful.
Split it apart and the picture changes. Visitors arriving on a service page from a branded search behave nothing like visitors landing on a blog post from a broad informational query. Lump them together and you get a number that moves without telling you why. A site can show a falling conversion rate purely because a new article started ranking and brought in early-stage readers who were never going to enquire that week.
So before changing anything, agree what counts as a conversion and measure it by page and by source. That’s unglamorous work. But you can’t fix a problem you only see as a monthly average.
Why this matters more than more traffic
Traffic has a price. Every additional visitor from Google Ads costs money, and every additional visitor from organic search cost time and content investment to earn. A lift in conversion applies to all of them at once, including the ones you’ve already paid for.
Conversion work also tends to uncover something bigger. When you examine why people aren’t enquiring, you often find the business has never been clear about who it’s for or why it’s the better choice. That’s a positioning problem wearing a website costume, and changing the form button won’t touch it.
For an established business, the enquiries you’re missing are usually the expensive ones. A buyer researching a large, considered engagement reads more pages, asks more questions and needs more proof than someone buying a product off a shelf. If your site answers half their questions and then presents a contact form, they’ll go and read a competitor’s site instead.
Where conversions are usually lost
Most sites lose people at one of four points, and it’s seldom the one the business suspects.
The page doesn’t match why they came. Someone searching for a specific problem lands on a general services page and has to work out whether you do what they need. Every second of that work costs you. Check your top landing pages against the search terms bringing people to them.
The site is slow where it matters. Speed isn’t just a technical score. Google’s web.dev case study on Vodafone found that a 31% improvement in Largest Contentful Paint led to 8% more sales in an A/B test, alongside a 15% improvement in their lead to visit rate. That’s one global telecommunications business testing one change, so treat it as evidence of the direction, not a formula. It makes the point that loading performance sits upstream of everything else on the page.
The two failures nobody wants to look at
The proof isn’t there. Established businesses often undersell themselves badly. The work is strong, the clients are happy, and the site says “we deliver tailored solutions”. A buyer comparing three suppliers needs to see what you’ve done, for whom, and what changed. Vague claims read as weakness.
The next step is too big. A contact form asking for company size, budget range and project timeline is a commitment. Early-stage buyers won’t fill it in, so you only hear from people who were already sold. Offering a smaller step alongside it, a short call or a specific piece of information, catches the people who aren’t ready to talk price.
What to check before you change anything
Give yourself four questions to answer with data instead of opinion.
- Which pages receive the most qualified traffic, and what’s the conversion rate on each?
- Where in the journey do people leave, and does that differ by device?
- Is the enquiry form working, and does every submission reach your CRM?
- Of the enquiries you do get, how many are the kind of work you want?
That last question catches a trap. A site can convert well and still be failing, because it attracts the wrong people. A jump in enquiry volume made up of price shoppers costs your team time and returns nothing.
The order the work goes in
There’s a sequence that saves money, and skipping it is the most common reason conversion projects deliver nothing. Our website conversion strategies article goes further into which changes to make; this is about what comes first.
Start with measurement, because without it you’re guessing. Then look at intent and content, because they decide whether the right people are landing on the right pages. Then performance and friction, because those are the mechanical blockers. Only then test variations, and only on pages with enough traffic for a test to mean anything.
Most businesses with modest traffic can’t run a valid A/B test. That’s fine. Qualitative review, session recordings and conversations with recent enquirers will tell you more than a test that never reaches significance. This whole sequence sits inside how we think about the relationship between a business and its buyers.
When to bring someone in
Handle it internally when the problem is obvious and contained: a broken form, a page that loads badly, a missing phone number. Those are worth fixing this week.
Get help when you’ve made changes and nothing moved, when the measurement itself is unreliable, or when you suspect the issue isn’t the website. A business whose enquiries stall after first contact has a follow-up problem, and no amount of page work will touch it.
Frequently asked questions
There’s no single benchmark worth chasing, because it varies enormously by industry, traffic source and what you count as a conversion. Compare your site against its own baseline over time instead of against a published average.
That turns on your traffic volume. A site with high traffic can read a change within weeks, while a lower-traffic site may need a quarter before any shift is meaningful.
Yes, and that’s usually the better starting point. Most gains come from clearer messaging, better proof and removing friction, none of which require a rebuild.
Yes. Every improvement in conversion multiplies the value of the ad spend you’re already making, so fixing the page first makes the same budget go further.
Start with the page that already has the traffic
Pick the single page receiving your most qualified visitors and work out what someone arriving there is trying to find out. Read it as a buyer comparing you to two others, and ask whether it answers their questions or simply describes your services. That one page will tell you more about why enquiries aren’t coming than any sitewide report.
Redfox is a Sydney-based digital growth agency. We work with established businesses whose digital presence isn’t doing justice to what they’re capable of, and we find where growth gets stuck instead of assuming the answer. If you’d like a second set of eyes on where your site is losing people, get in touch. Book a discovery call.
