Skip to main content


How to tackle Mobile Conversion Rate Optimisation

online-commerceCan you transfer learnings from your desktop optimisation to your mobile site? Do you need to optimise each site individually? These have suddenly become critical questions for online businesses – and here’s why.

Mobile is mushrooming. Analytics data for several of our clients shows that, for the first time ever, sessions on a mobile device have overtaken desktop sessions. Much of this growth is fuelled by increased 4G coverage, which has been found to have a positive relationship with online shopping. As the roll out of 4G continues, you can expect not only more mobile visits to your site, but also increased mobile sales. 4g-growth

Source: Cisco VNI Mobile, 2015

Figure 1: Over the next 4 years, 4G will grow to make up 68% of all mobile data traffic . Recent research suggests that this should have a positive impact on online sales.

While desktop still enjoys the lion’s share of online sales, US data suggests that mobile has started eating into the desktop pie.



As a result, mobile Conversion Rate Optimisation has rocketed up the priority list – leaving E-commerce directors dismayed at the thought of doubling the CRO workload. It’s not surprising therefore that the question we’re asked most frequently right now is whether mobile and desktop websites can be optimised together and how to go about it.

The first step is to assess whether your mobile site is ripe for optimisation – not all are, so you could be wise to stick to desktop only for a while yet. Here are some questions to consider:

1. What is the contribution from mobile to overall revenue?


This graph shows mobile sales lagging behind, which suggests that the best short term business case is to focus efforts on the desktop site, as that’s where the money is. An uplift on desktop will have a bigger overall revenue impact than the same uplift on mobile.

2. What is the growth in mobile sessions for your site?


This graph shows how mobile device traffic has recently overtaken desktop for the first time ever for this particular client. Even if mobile revenue is still small compared to desktop, this could be a signal that optimising the mobile site could pay dividends.

According to Criteo’s State of Mobile Commerce Q2 2015 UK report, optimised mobile sites convert 35% better than non-optimised ones . This represents a significant opportunity cost for non-optimised mobile sites showing increasing traffic.


Source: Criteo

3. Where are you on the optimisation curve?

If you’ve been doing desktop optimisation for a while, and have started seeing diminishing returns, it may be a good time to start focusing more on mobile optimisation. However, if you’re just starting out, you may be able to squeeze more revenue from your desktop site in a shorter space of time.

Assuming your mobile site is a worthy candidate for optimisation, read on to find out where you can make savings, how to avoid duplicating unnecessary work, and what CRO activities you must carry out separately for each device.

Research and analysis: How much of the learning can be transferred?

If you’ve yet to start an optimisation programme:

Don’t think of device-specific optimisation as separate projects. You’ll see the best results by optimising your website as a whole and prioritising opportunities regardless of platform. Initially, you may find that mobile optimisation opportunities are lower down the priority list than desktop opportunities because they would have less of an impact on your overall bottom line. Alternatively, if the greatest opportunities lie on your mobile site, you may be a few months in to your optimisation programme before you’re testing anything on your desktop site. The key is not to go into this expecting to prioritise one device over the other. Follow what the data tells you. Sniff out the biggest opportunities and get the biggest wins.

Take an iterative approach to your research. Start by creating conversion funnels for each device. What do you see? Where are visitors arriving? Where are they exiting? What are they doing in between? You’re bound to uncover lots of insights with this.

Add heatmaps to key pages that you’re interested in finding out more about how visitors are using the page. Use the mobile Qualaroo tool to set up questions on pages where you need qualitative data to uncover why visitors are behaving in certain ways. Run usability sessions with visitors on all devices to really understand the specifics of each visitor set.

This process is less about transferring learnings and more about digging, and then being flexible in where the digging leads you. If you require more mobile specific digging, then build it into your programme. If you uncover something on the desktop site and then realise you need to pose a different Qualaroo question to mobile visitors, set it up. If a finding from a usability session on your tablet site reveals something new and you’re not sure if it’s also relevant to mobile or desktop visitors – pursue it.

Once you have generated your learnings, the next step is to prioritise areas to focus on, so that effort and resource goes into those likely to give the biggest return in the shortest time, regardless of which device the opportunity is on. Time needs to be spent prioritising the issues and preparing a combined optimisation plan with all devices included before working on the creative treatments to address the opportunities and setting up the split tests.

If you’ve already started optimising your desktop website:

Although some findings apply universally across devices, visitor behaviour, psychology and intent are key areas that are likely to differ. Optimising your mobile site is not, therefore, as simple as doing the research, analysis and testing for one device and transferring learnings or improvements to another. True, some learnings can be transferred from one device to the other, but there are some aspects of CRO that need to be carried out independently.

Key techniques that require additional analysis are:

  • Conversion funnels mapping where visitors are dropping out in the conversion process from homepage to order confirmation page will need to be created for your mobile, desktop, and tablet websites. Another key area to look at is landing pages, where are visitors landing, where are they coming from? This analysis allows you to easily identify any similarities and differences in visitor journeys according to device.
    In the UK, 45% of ecommerce transactions involved multiple devices . This is a blind spot with traditional analytics packages. If a user visits your mobile site to browse and then return later to convert on desktop, that customer will be counted as two different users and the conversion will be attributed to desktop. Tools like Woopra aim to overcome this by uniting data from different sources at an individual user level.
  • On-site surveys – although there are some key questions you want to ask visitors regardless of device, the trick is to set them up as individual questions, targeting each device, so that you can analyse the response according to device type. In addition, there are likely to be some different questions you want to ask depending on what you find from your conversion funnel analysis. If a higher percentage of visitors are dropping out between landing and product pages on your mobile site that any other device, for example, you need to find out why.
    In one study, we asked where the user was and what they were doing, to get an understanding of their context. They were clearly distracted: answers leaned towards “cooking”, “watching TV” and “on the loo”. Furthermore, 70% of this user base told us that they were visiting the mobile site to browse, rather than to buy, and 40% were likely to buy in store or on the desktop site. This helped us to quantify the potential of mobile optimisation for this client.
  • Heatmaps – if your website looks the same across devices, then you can set up one heatmap and filter the results by device to conduct the individual analysis. If, however, your site looks different according to the device used to view it, then you’ll need to set up individual heatmaps to analyse where your visitors are clicking, scrolling and viewing.
  • User testing – In order to identify key usability themes and issues, we recommend running at least 5-8 qualified sessions per device – desktop, tablet and mobile. You may also want to further break down mobile into different segments according to device size – running 5 sessions on a small screen and 5 on a large screen.

Split Testing: Will improvements deliver across devices?

You may find that some optimisation opportunities apply across devices. For example, you may have a common theme of visitors complaining of a lack of product copy or images on product pages regardless of which device they are using. Such a finding may represent a big opportunity for you, but don’t fall into the trap of thinking that the treatment to address this issue will be the same. Improvements that work on mobile may not work on desktop or vice versa. Ideally, you want to run different tests on different devices. This is particularly the case if you are testing a radical new design.

For one of our clients, we ran a redesigned homepage split test. Although it delivered 34.7% increase in Revenue Per Visitor (RPV) on the desktop site (read the full case study here), RPV for visitors viewing the site on a mobile actually decreased.

There could be many reasons as to why this may be the case. Perhaps the new design was too cluttered for the smaller screen size. Did visitors not use their mobile to browse in the same way desktop visitors did? Without further research we can only infer. What we do know is that when the winning variation was rolled out onto the live site, it was important that only the desktop site was made live as the mobile homepage converted better as it was.

Top tip: Don’t assume that you’ll get a similar uplift on your mobile site as you are your desktop site. If you have already started optimising your desktop site and have run split tests across all devices, go back and segment your results by device if possible. (Optimizely has this feature, but on some platforms it needs to be set up in advance before starting the tests).

Interested in taking the next steps in mobile Conversion Rate Optimisation?

If optimising your mobile, or desktop, website is high on your list of priorities this year, get in touch today. Our team of optimisers have experience increasing sales and revenue for a wide range of desktop and mobile sites – and if you’d prefer to do it in-house we also offer training and mentoring to make sure you get the best results possible. 


Keep up-to-date

People from Facebook, FarFetch and RS Components receive our newsletter. You can too. Subscribe now.

Interested in turning experimentation and testing into an advantage for your entire business?