Animated character working on website

Four good ways to increase website conversion rates

30th April 2021

7 minute read

A prospective client asked us yesterday “what are your top four ways of increasing conversion rates?”. I didn’t have a good answer for this, as each market, audience and website are contextually different, and we tailor our advice for your specific situation. Sometimes there are flaws in the UX, sometimes the products are being mis-sold, sometimes the context in which the products are being sold is wrong. There is no ‘one size fits all’ approach, which is why we spend our time analysing and understanding a site before we give recommendations.

Having said that, here are four approaches that will help you when considering why your site isn’t selling as much as it should, or could, be. 


First you need data. But do be careful about data for the sake of it; you can prove anything with enough data. The key thing when looking at analytics on a website is to drop any of your preconceived notions about what your customers want, and definitely drop any ideas of what you want to sell. Don’t try and mould what the data is telling you to fit what you already think you know; instead, look at the data and then ask yourself a set of questions and see what the answers are.

Maybe first look at the conversion rate of visitors to customers – in other words, the number of people who visit your site who then become customers. Let’s imagine this is around 5%. Now if, like most FMCG/Challenger Brand websites, you have a range of products this conversion rate will be a blended rate of all of them, so you’ll have some which convert a lot higher and some a lot lower. Don’t mistake the blended rate for a number that means anything – it may be that you’ve a product that sells really badly which is dragging your numbers down, or one selling well which is pulling them up. Instead look at the conversion rates per product and see what the range is.

Next look at the abandonment rate of your products, per product. Which ones are people adding to the cart and then abandoning the most?

Armed with these two pieces of information, go onto your website and try and understand what the data is telling you, in context. Understanding why one product is selling really well and one isn’t, in the context of your website, will tell you a lot about your customers and what they really want. Use the data to understand your customer’s behaviour in the context of what you’re selling. 

Don’t be afraid to shoot your darlings, or don’t try to optimise the wrong thing.

Say you’ve a product you really like, but it’s just not selling. The rest of the products are fine, but this one – people just don’t seem to like it. Or you’ve a page you like, or there’s an image which you think is great. You don’t want to drop it because of all the time you’ve spent on it, so instead you spend months tweaking and testing and nothing is working. What do you do?

Drop it. Or at least, test a version of the site without it on. Your aim with CRO isn’t to optimise the sales of every product, nor is it necessarily to prove that you’re right; more it’s to prove that you’re wrong. Dropping that product that nobody wants might stop traffic going to it. That traffic may go to one of your higher-converting products instead, which then might make you more money. We’ve advised a client to drop all but four of their products as short-term this made them more profitable and long-term made for a better business strategy (for various reasons specific to their products and their plans for product expansion). 

Remember always that you’re not your customer. Just because you think something is great – what do your customers think? You might be spending time trying to optimise the sales of a product that people simply don’t want, and trying to force something on them that they obviously don’t want turns you into an Apprentice contestant, and nobody wants Alan Sugar’s finger pointed at them. (I nearly wrote something much cruder there, which I’ll now leave to your imagination.)


Put hotjar on your site, and watch what customers are doing. This seems so obvious, but so many people don’t do it. You can get a free account, and you can watch actual video recordings of people going through your website. This forces you to see things through their eyes. 

Ever been at someone else’s house, or a new AirBnB and trying to find the fridge? Opening every drawer and cupboard until you chance upon it, and then asking yourself ‘why on earth did they put it there?’. Well, that’s what customers are doing on your site. To you things make perfect sense, because you’ve been using the site for ages. For a fresh customer, starting without those preconceptions, it’s a whole new world. Watching what they’re doing on the site is essential for finding those problems, and to see that the things which you think are super-important your customers couldn’t care less about. For example, you’ll almost certainly find that the lovely rotating banners that you pestered your agency to design and got some great photography for and argued for weeks about the text on are completely ignored. 

Another top tip – watching where their cursor spends time is a good proxy for where their eyes are spending time; in other words, what they’re actually reading and are interested in. 

Don’t be afraid to test dumb things.

People test obvious things – I’m an advocate for testing things which may be a bit dumb. If you’ve read Rory Sutherlands fantastic book Alchemy, for instance, you’ll know that asking dumb questions actually can bring more insights than asking the ‘clever’ ones. This can also be thought of as going back to first principles, and dropping the assumptions that have got you where you are now. So as a thought experiment – imagine a website that sells vitamins, and the USP upon which they sell their vitamins is that a customer will be more hydrated after taking them than they would be by drinking water alone. Why are people buying the product? Well, because they want the super-hydrating vitamins of course. What a dumb question. 

Well no, not really. They can’t measure how hydrated they are, after all. So maybe they’re buying or visiting the site because of something more emotional – perhaps they want to feel like they’ve got the smartest vitamin product around, better than those boring ones at Boots that everyone else buys. In other words, they want to feel smarter about their health than other people. They know that athletes need to stay hydrated, and perhaps that’s how they want to see themselves. Or that your brain works better when you’re hydrated. 

So maybe focusing the site less on how much Zinc they’re getting in each tablet and more on conveying the ‘cleverness’ of the vitamins (or perhaps more specifically, the cleverness of the people who take the vitamins) will end up selling more, as it will help the customer to emotionally feel that they are who they want to be. Maybe the site isn’t selling vitamins at all, and maybe the customers aren’t coming there because they want vitamins.

If all this sounds like hard work, well, it is. It’s not just ‘make the button bigger, it’s about understanding your customers and why they do what they do. That’s why agencies like us exist. This hard work is important though, because turning visitors into customers is vital for growth, in a world where so many FMCG products are relatively easy to replicate, getting customers ahead of your competition is vital. 

(By the way – if you do want some ideas to try without first understanding what’s happening on your site, this site has quite a few that might be useful.

Written by Rob Dobson

Rob Dobson has been working in digital and building websites for 20 years. From designing and developing the world’s first internet bank in 1999 (, he founded Northern Comfort in 2010.

See how we can help you.