A comprehensive guide to Conversion Rate Optimisation (CRO)
Conversion Rate Optimisation (CRO) is something which should go hand in hand with user experience (UX). Both can contribute to making sites and apps easier for users, while CRO and UX share a lot in terms of goals and techniques.
In this article, I’ll explain CRO, what it can do, and how people can make a start in optimising their websites for conversions.
What is Conversion Rate Optimisation (CRO)?
CRO is a broad term which covers a range of techniques with the same goal. That goal is to increase the number of visitors to a website or app who convert in some way.
Depending on the goals of the site, this conversion could be a sale, a request for a quote, or something more minor like signing up to receive marketing emails.
The reality is that the majority of visits to a website or app will not result in a conversion of any kind.
Some are simply not ready to make a decision, perhaps they’re researching products and services or comparing prices. There’s less that businesses can do about this.
However, in other cases, people have been prevented or deterred from converting because they had problems creating the tasks, or that the process became so frustrating that they decided to abandon the site. Or perhaps they were close to converting but couldn’t find the information they needed.
In the case of problems preventing users from converting, CRO can help. Using conversion rate optimisation techniques, reasons for abandonment can be found and points of friction within the customer journey can be identified.
Essentially, it optimizes conversions by taking account of user behavior and making changes that best fit they way people behave on websites and apps.
Why do you need Conversion Rate Optimisation?
No website experience or user journey is perfect, and nor can it be for every visitor that arrives.
Anything that improves that journey makes it easier for customers to buy or convert from your site in some way.
It has the potential to increase the value you receive from every visitor that comes to your website.
Today’s ecommerce market is relatively sophisticated, with plenty of competition between sites. One way sites can differentiate themselves now is through a better user experience, which makes it easy as possible for shoppers to find and then buy the products they want.
Good CRO here can make customer acquisition cheaper, as it should increase the value per customer acquired, as they’re more likely to convert.
It’s also about increasing profits. Retailers spend lots of time and money creating sites and apps, on marketing to attract potential customers. CRO helps to ensure that more people convert and turn this traffic into profit.
How Conversion Rate Optimisation can work with UX
Though UX and CRO are seen as different disciplines, there are plenty of areas where they overlap, particularly in the methods used.
Common conversion optimisation methods include:
- A/B testing
- Multivariate testing
- Customer journey analysis
- Online surveys and customer feedback
- Session replay tools
- Copy optimisation
- Cart abandonment analysis
- User testing
Many of these methods – such as A/B and multivariate testing, customer feedback and user testing – would be employed by UX professionals and well as CRO practitioners.
UX is, broadly speaking, about using these methods (and others) to make websites, apps and other interfaces as easy as possible for people to use.
The difference between the two disciplines lies in the end goals. For UX, making sites and apps more usable is the end goal. For CRO, the end goal is increasing conversions and therefore profit, often by making sites easier to use.
The biggest common factor may be that good UX tends to correlate with improved conversion rates.
There are some possible exceptions though. For example, in the pursuit of increasing conversion rates, some techniques may not be what users want. For example, inserting pop-ups with offers may be intrusive, but can also help to drive conversions.
However, there’s no serious conflict between CRO and UX. Both seek to make website perform more effectively, by listening to and designing for users, and both have a respect for the use of testing and data rather than guesswork to produce the best outcomes.
Peep Laja sums it up very well:
“As long as we focus on the final outcome that we want to produce, which is, to grow our business, to have happy users and those kind of things. So as long as you know what is the final outcome you want to produce… it doesn’t matter really if you call it UX or CRO or customer experience.”
How to get started with CRO
The best results are likely to be achieved with a planned and structured approach, which starts out with clear goals.
You may arrive at it from different angles. Perhaps you feel your conversion rates could be better, or you are aware of CRO and want to see how it could improve your site’s performance.
Convincing the boss
The first step for many can be persuading key stakeholders in the business that an investment of time and resources into CRO is worth it.
First of all, it may be necessary to make the case for CRO and convince the key people. There’s a robust business case for CRO, and it can be possible to demonstrate the potential with just a little budget.
Once management see the kinds of results CRO can generate, it should convince them to continue or increase their investment.
Take a structured approach
Company culture is important here. While CRO can be effective even when used on an ad-hoc basis, it can be more effective with the right structure and attitudes within a company.
Many businesses still don’t use some of the most common techniques, or are not applying CRO in a way that could produce the best results.
Econsultancy’s Optimisation Report found that 63% of companies said they lacked a structured approach to optimisation, while 77% of agencies said the same about their clients.
In other words, CRO is more effective when key stakeholders buy into it, and the organisation adapts to it.
Focus on the customer/user
Like good UX, effective CEO requires companies to focus on the customer. CRO is only successful when companies are prepared to find out about their customers – how they use the site, the tasks they’re trying to perform, what makes them abandon the site, and so on.
This is the first step. The second, crucial step is to listen to what your testing and optimisation efforts tell you about users, and put these lessons into practice.
This can require significant change within organisations. Replacing guesswork with structured testing and optimisation requires key stakeholders to trust the results of CRO efforts, and to be prepared to invest the resources in the right area.
Where can you improve?
It’s good to start by identifying key opportunities for improvement on your site.
There are a range of sources to look at for this. Analytics may provide some insight into areas which are underperforming, perhaps product pages with lots of views but low numbers of conversions, or checkout steps with higher than normal dropout rates.
This data can give you a starting point of where to look, but there are other sources too. Customer feedback and on-site survey tools can help you pinpoint problem areas.
Common issues harming conversion rates
It’s good to examine common areas and problems when looking to improve performance.
Here are some possible areas:
- Forms. Most online transactions involve tackling forms at some point. Checkout is the most obvious, as well as forms for quotes or bookings. A well-designed form which is easy to understand and complete, means more shoppers will finish the process. On the flipside, poorly designed or confusing forms can cause users to abandon in frustration. When looking at forms, there are lots of potential for tripping up users – poor labelling, lack of feedback on progress, poor error messaging, lack of mobile optimization, and more.
- Lack of key information. There are certain things that customers need to know before they can commit to converting, or that they need to be reassured about. For example, information on delivery costs and timescales, as well as other areas such as ease of returns can be a key factor in the customer’s purchase decision.
- Choice paralysis. If shoppers see too many products on a website, it can be difficult for them to make a decision. The challenge is to help shoppers filter and sort to narrow choice, and to find the information they need to make a decision.
- Distractions on web pages. Too much visual noise or information to wade through and confuse users.
Once you have some ideas of where to look, it’s time to start using some of the CRO methods we mentioned earlier.
A good way to start is to drill into some of the possible issues you’ve identified, by observing how people are using your website.
Session replay tools and heatmaps
Session replay tools can capture actual people using your site, enabling you to follow their journey, see where they click and view any problems they may encounter.
For example, if you’ve identified people dropping out at a certain stage of checklist, session replay may enable you to pinpoint it to a tricky form field, or see that people aren’t viewing your CTA.
Heatmaps can tell you where people are clicking on your pages, the page elements that grab their attention (as well as those they miss) and how far they scroll down a page.
Both can be useful, and offer a good level of insight. However, while they can show you what visitors are doing, they can’t tell you why.
Usability testing can uncover problems and pinpoint issues but because the user is being observed, and able to give feedback.
This means is should help to take the guesswork out of CRO, cutting through opinion to deliver lights direct from people using the site.
It can answer key questions about how people interact with your site, the things they are looking for but can’t find, the major issues which get in their way, and how easy it is to locate key features and information.
The answers to these and other questions can help to confirm areas of improvement, and an also uncover new areas to address.
Once you have uncovered issues which are affecting conversion rates, the next step is to make improvements, and test to see whether your hypothesis was correct, and they do deliver better results.
Some improvements may be obvious. A form field that doesn’t work on mobile, or on certain browsers doesn’t need testing, it just needs fixing.
However, there are other areas where ideas and improvements should be trialled before implementation. This is where A/B and multivariate testing can help.
A/B testing is a way to make improvements by testing different versions of a page or app against each other to see which version produces the best results.
A/B tests may involve two completely different pages, or perhaps a page where just one or two elements have changed.
In the test half of the incoming traffic is sent to one version of a page, half to another. Depending on the goals of the page you can determine which was most successful.
Here’s a typical A/B testing process:
- Identify possible improvements. Use analytics and other data to find pages to improve, perhaps those with lower than average conversion rates, or high bounce rates.
- Select goals. Which metrics do you want to improve on the page? Perhaps you want to increase engagement on the page, or improve add to basket rates.
- Select a hypothesis. Generate ideas for improving the page and solving any issues you’ve uncovered.
- Create variations to test. Perhaps you feel the CTA could be more prominent, so you could test different placements or designs.
- Run your tests.
- Analyze the results. Did your hypothesis work? Did it improve performance? If not, even failed tests can deliver insights.
A/B testing allows you to test two variations, but multivariate testing can test changes to a range of features on a page.
This allows you to test a greater range of variations to find the most effective combination for the page, as well as measuring the effect of all the changes together.
The drawback, when compared to A/B testing, is that MVT requires high numbers in terms of traffic and conversion to deliver reliable results.
It can also be harder to understand what works and why, as the number of variations make it harder to isolate individual tweaks.
For both A/B and MVT, the key is to have enough data to achieve statistically significant results.
Like UX, good CRO is about using a disciplined, step by step process of research and testing to deliver insights and improvements.
To be truly effective, it requires senior stakeholders to be willing to listen to the user and deliver changes based on the findings of tests.
Also like UX, the work of a CRO is never finished. It should be a continuous process of analysis and testing. User behaviour and technology changes all the time, so no web page will be perfect for ever.
Graham Charlton is Editor in Chief at SaleCycle and former Editor of Econsultancy and ClickZ. When he’s not creating content, he can be found listening to vinyl, spending time with the family and enjoying the odd glass of red wine.