Small business help, guides and calculators Since 2004

Why your website visitors aren’t buying – and what you can do about it

If you’ve been involved in running a website, you’ll know that it’s relatively “easy” to get eyeballs on your site. Throw money at it and they will come.

But, it’s getting those visitors to take action, to buy from you – that’s the hard part. So it’s no wonder that a website’s ‘Conversion Rate’ is one of the most keenly tracked metrics in online business. It measures the percentage of visitors who take a desired action, which in e-commerce means making a purchase.

Improving your conversion rate can mean more sales without having to pay for more traffic, so finding ways to convert more of your website visitors into buyers is a key plank of ecommerce success. Our guide to finding customers and making sales looks at the wider challenge of winning business.

Conversion rates vary widely by sector, product, device and the kind of traffic a website attracts. A site selling expensive equipment will behave differently from one selling everyday products. Rather than assuming a particular percentage of visitors should buy, start by measuring what happens on your own site.

Under pressure to grow revenue, many marketers increase their traffic acquisition budgets. If you are thinking about spending more on advertising, our guide to marketing your business online and offline covers some of the options. However, this can make the problem worse: unless those new visitors are warm prospects, your conversion rate may drop even further.

A more effective approach is to improve the ability of your site to convert your visitors, rather than buying more traffic. This is what Conversion Rate Optimisation (CRO) is all about. You use data to ascertain where and how to tune your site to improve sales.

Here are five things you can start doing today to turn more of your website visitors into buyers:

1. Stop obsessing about conversion rate

This may sound counter-intuitive, but overall site conversion rate can be a misleading indicator. For example, not everyone is on your site to buy something today, or even at all.

Focus instead on factors that you can control, like the ones outlined below.

2. See where your visitors are dropping out

You may already have useful data from an analytics service such as Google Analytics 4 (GA4). Its events and funnel exploration reports can help you see where visitors abandon important steps, such as adding a product to a basket or completing checkout. GA4 calls important actions, including purchases and enquiries, key events.

Here’s the trick: compare different groups of visitors instead of relying on the site-wide average. For example, how do mobile journeys compare to those made on computers? Are people finding you through search more likely to buy than those arriving from social media? Which segments are most relevant to your situation?

3. Understand why they are dropping off

GA4 can show you where visitors are dropping off, but it won’t tell you why. This is equally important to know, because only then can you start addressing the issue.

The way to find out why visitors leave is to ask them. On-site polling can be useful. A short question can reveal problems that the numbers alone cannot show.

Services such as Hotjar (part of Contentsquare) and Qualaroo can help you gather feedback, although you don’t necessarily need specialist software to ask customers what nearly stopped them buying.

These polls can be annoying, so don’t overdo it. Run one at a time, or two in different areas of the site.

Open-ended questions are more insightful. One of our favourite questions is posted on the payment confirmation page and asks: “Was there anything that almost put you off buying from us today?”

If you have live chat on your site, read through the transcript logs from time to time for hints on what may be getting in the way of potential customers.

4. Usability testing

Some of our most powerful insights have come from watching users interact with the website. This is known as usability testing.

For a free starting point, Microsoft Clarity offers heatmaps and session recordings that can help you spot areas where visitors struggle. A service such as UserTesting can help you recruit participants and watch how they use your site. You can also start more simply by asking a few people unfamiliar with your website to complete a realistic task while you observe. Give them something specific to do, such as finding a product and checking the delivery cost, and see where they get stuck.

If you’re using a research panel, make sure the participants resemble your likely customers. Otherwise, their feedback may not tell you much about the people you’re trying to reach.

5. Test your proposed improvements

It’s a good idea to test your interventions on a live audience to check their impact, before committing the changes to your site.

An A/B test shows the original page to one group of visitors, while another group sees a version with changes. Comparing the results can help establish whether the change actually improves performance, rather than simply looking better to the team. Some testing platforms let you make changes without editing the site’s code.

You’ll find many case studies of triple digit improvements, but it’s generally not a good idea to only shoot for home runs. Incremental improvements can stack up to a meaningful result.

The caveat is that it can be difficult to draw reliable conclusions from A/B testing on low traffic volumes. How many visitors do you need? There’s no single threshold: it depends on the existing conversion rate, the size of the improvement you are trying to detect and how many people see the test. If you have relatively little traffic, direct customer feedback and usability testing may be more practical places to start.

Established testing platforms include VWO, Convert and Optimizely. Choose a tool based on your own requirements and budget, rather than assuming expensive testing software is essential.

This guide was originally written for ByteStart in 2017 by Johann van Tonder, co-author with Dan Croxen-John of E-Commerce Website Optimization.

Editor’s note (November 2026): We’ve updated this article to reflect current analytics and testing tools, removed outdated service recommendations and refreshed the conversion-rate advice. The original author’s structure and practical guidance have been retained.