What methods are companies using to improve conversion rates?

Exactly what is being optimized? And which methods are proving more difficult to implement than others?

These are some of the questions asked in Econsultancy's seventh Conversion Rate Optimization Report, in association with RedEye.

Here are a few highlights from the survey of nearly 900 cient-side and supply-side marketers.

What methods are used to improve conversion rates?

Multivariate testing (MVT) continues to be the method bringing the most significant improvement and increased satisfaction rates, consistent with last year’s results. 

However, multivariate testing is still struggling for adoption, currently used by only 30% of respondents.

A/B testing is unsurprisingly at the top of the pile, currently used by well over half of respondents.

Customer feedback also features prominently, something which Dan Barker, independent consultant, finds encouraging.

Customer feedback is the heartening point here: it’s ranked almost as highly as A/B testing as a method companies use to understand and improve their results.

In the past sometimes optimization was seen as almost synonymous with A/B testing, and it’s therefore good to see other tactics have grown to similar levels of use.

Customer feedback is also positive as, unlike A/B testing, it can capture the one or two really unhappy VIP customers and allow you to act on them, as well as the broad-brush ‘aggregate level’ results A/B testing allows.

Which of the following methods do you currently use to improve conversion rates? (Company respondents = 449)

conversion rate tactics

What do companies test?

The chart below shows some interesting trends in testing. A large majority of respondents are testing those factors that are perhaps easiest to test (calls to action - 83%, page layout - 77% and copy - 72%).

Other areas that have a large impact on conversion, such as checkout process (46%) are still being tested by less than half of respondents. We may speculate this is due to limitations of technology (CMS, third party testing tool or in-house tech).

Only 32% of respondents are testing site search, an area that Suniel Curtis, Head of Analytics at Hays, thinks could offer competitive advantage.

It is understandable that the complex area of search functionality is only tested by a third of the respondents and I understand not everyone has search.

With the vast choice consumers now have, optimising search and filtering is becoming a big issue so I see this rising as an area to test.

A further section of the report looks at the number of tests run per month and finds the optimum number (for clients’ perception of both ‘satisfaction’ and improvement in conversion rate) is three to five. Anything more or less and improvement falls.

As was the case in last year’s survey, there’s evidence that improvement and satisfaction increase as the number of testing methods used increases.

However, even among companies using more than seven methods, only 41% are satisfied with conversion rates.

Specifically for your website, what do you test? (2015 company respondents = 335) 

testing areas

Which methods are easiest to implement?

Website personalization and multivariate testing top the pile for ‘difficulty to implement’.

Of course, personalization is a broad church, but it’s perhaps no coincidence that two of the lesser seen methods of improving conversion rate (see chart above) are deemed difficult to implement.

Though few find usability testing ‘very difficult to implement' (8%), it’s surprising to see 44% deem the tactic as ‘quite difficult', given how mature this discipline is.

A further question in the survey asked about team structure and investment in conversion rate optimization.

The results show that organizations with dedicated resource are 16-20% more likely to see improvements in conversion rates.

Of course, there are a lot of qualifiers here, but this is a reassuring correlation all the same, given not all those investing in CRO have seen an uplift.

How difficult is it to implement the following methods for improving conversion rates? (Company respondents = 207) 

difficulty of conversion optimisation methods  

For more on conversion strategies and tactics organizations are using, in addition to the tools, processes and resource employed for improving conversion rates, download the seventh Conversion Rate Optimization report, in association with RedEye.

Ben Davis

Published 11 November, 2015 by Ben Davis @ Econsultancy

Ben Davis is Editor at Econsultancy. He lives in Manchester, England. You can contact him at ben.davis@econsultancy.com, follow at @herrhuld or connect via LinkedIn.

1244 more posts from this author

You might be interested in

Comments (8)

Save or Cancel

Auke Vos, Owner at Umigo

Interesting read, but it's a bit comparing apples and oranges. You are comparing research methods and validation methods. AB and multi-variate testing are, and should be used as tools to validate findings/hypothesis that where discovered during research and are not methods in itself to increase conversion rates.

almost 3 years ago


Darren Ward, Director of Product Marketing at User Replay Ltd

It is interesting that the focus in the article is on A-B testing, MVT etc which is really all about getting people into the funnel. However, companies pay much less attention than they should about what happens when you actually get those users into the funnel.

However, some of the biggest conversion uplifts and revenue benefits I have seen are from removing the obstacles that customers hit when they are trying to buy - these could be technical issues, usability problems, application logic problems and issues with 3rd parties such as payment engines. More to the point, once you have discovered these issues - how do you quantify their impact on conversion. This is what CEM tools such as UserReplay are fantastic for and provide all the tools to be able to do this.

Companies really should be paying more attention to the experience users have in the funnel - or else those investments in MVT, A-B etc could be wasted.

almost 3 years ago

Edmund Jones

Edmund Jones, E Commerce Marketing Manager at Digital Marketing Pro

Thank you. Amazingly there is no mention of testing the online value proposition in this blog or many others that I have seen that discuss CRO.

Many of us try to improve results by changing page elements, incentives etc., when the first step should really be to focus on strengthening the core value proposition. Why is it that such a fundamental factor that effects the probability of conversion is not even mentioned? Are digital marketers being relegated to operational pawns in the organisations' marketing machine ? Are they not seen to be able to contribute to strategic marketing discussions about OVP? And should digital marketers take responsibility for this state of affairs?

I would argue that identifying , expressing and testing alternative versions of the online value proposition should be the starting point of any CRO process. In fact value proposition for your brand and your expression of that value are at the heart of marketing and have always been. Below are some ideas that spring to mind when thinking about strong value propositions:

Differentiated from competitors
Meaningful to customers
Meets minimum , non negotiable requirements for the category/ product
Is expressed in a clear and memorable manner
Is constantly evolving yet tied to the fundamental essence of your business
Relative to a space , context, time

almost 3 years ago

Landon Donovan

Landon Donovan, Digital Marketing Consultant at Picreel

Ben, thank you for the detailed, informative and interesting post!
I’ve been doing lots of conversion rate optimization with my clients this year and I it’s tough. It’s mainly because most of these people want the definite answer, like in math. They want to know that the problem is solved. But you never really know if you’re reached maximum or if you need to proceed or just try something really crazy and out of the box. Here I found some great data about the most useful and doable methods of CRO. Perhaps, it’ll help to learn tactics we should focus on. Also, I think that exit intent technology software should be in this list because this method has helped me for a couple of years. I was using picreel.com to get the result, for example. But, there are others that prove to be useful.
Thanks for the post and new ideas!

almost 3 years ago

Annabel Paton

Annabel Paton, Group Marketing Manager at PureClarity

Really interesting read, Ben. It's surprising to see that only 22% of respondents are currently using web personalization however, these statistics show that the next 12 months are going to be extremely exciting in the industry with 56% planning to use personalization in the future. At PureClarity, we're starting to see an uplift in the number of customers who approach us with a clear understanding of the importance of a personalized strategy.

I'm extremely surprised that 45% of respondents believe it to be very difficult or quite difficult to set up event or behavioural triggered emails and almost 80% believe the implementation of personalization to be difficult. This just shows how it's up to solutions such as PureClarity to educate retailers in the onboarding process and eliminate those barriers to invest in such powerful software.

One method I would highlight as key to any ongoing ecommerce optimisation strategy is on-site search. The importance of an optimised, smart search can often be forgotten in favour of optimising other areas of the website. Research has shown that visitors who use search often convert at a higher rate and even spend more - it's definitely not an area to be overlooked!

Thanks again for the post, I look forward to reading more.

almost 3 years ago

L.M.L. Beerthuyzen

L.M.L. Beerthuyzen, CEO at 1972

Nice post! In my opinion A/B testing is an overall method to validate your insights. You first start with online surveys, heatmaps, usability research etc. With this data you can create a hypothesis, which you should test. So A/b testing is always the last step of optimization.

You can a/b test everything; every method like personalisation, or any finding out of your research and surveys.

almost 3 years ago

Mike Smee

Mike Smee, Business Development Country Manager (UK) at Devatics

I find it really interesting that website personalisation is viewed as the most difficult thing to implement. But I think also that this maybe comes from a lack of awareness of solutions on the market...At one end of the scale I worked with personalisation solutions that require deep site integrations via APIs etc requiring significant IT input / sponsorship and at the other end, very effective solutions that simply involve dropping javascript tags into pages for the collection of behavioral data and the "squirting" of content / offers / messaging into the journey. Both approaches produced similar results, however time-to-market, cost, complexity massively reduced using the second approach (literally a few days). I think the future of e-commerce is all about agility and simplified implementation processes, low pain of change etc. Technology vendors who cannot adapt to this way of working will not survive.

almost 3 years ago


Robert Moore, Marketing Manager at KSA

Price is perhaps overlooked. The marketing department do not have enough influence over this crucial factor. I have tested so many different variables over my career and as the OP said value proposition is the MOST important

almost 3 years ago

Save or Cancel

Enjoying this article?

Get more just like this, delivered to your inbox.

Keep up to date with the latest analysis, inspiration and learning from the Econsultancy blog with our free Digital Pulse newsletter. You will receive a hand-picked digest of the latest and greatest articles, as well as snippets of new market data, best practice guides and trends research.