Conversion Rate Optimisation Statistics You Need To Know | Infographic

We spend a huge amount of time, effort and money getting people to our websites. This might seems like a pretty hard task at times but in all honesty, it is the easy part. The challenge comes in when you try to get those visitors to convert.

This is by far the most important part.  The smallest difference in your conversion rate can make a big difference on your bottom line. If you can increase your conversion rate a single percentage point from 1% to 2%, you could double your revenue.

We came across an awesome blog by Startup Bonsai where they pulled together a list of CRO Statistics and we took a few of the key ones and put it into this infographic. They're the stats you need when it comes to winning over the key stakeholders in your business on the subject of CRO and the benefits of an investment into a long term CRO strategy made up of continuous testing and improvement.

Give it a read, share it around and when you're ready to create a strategy and put it into action, come talk to us.

 

 

Keep an eye out for our upcoming infographic on CRO tips and tricks and our next eBook, the first in our 'How To' series, How to Take a Business Online.


Think outside the box

Time to Think “out of the box”

As testers, our passion for quality and taking pride in our work can sometimes be seen to conflict with other project priorities, like the need to deliver at pace, and to rapidly adapt to change. As I alluded to in my last blog, technology is changing so quickly that there is no longer a one size fits all testing approach; no two platforms are the same, and no two customers are the same. At Eclipse, while we have a defined test strategy and some core testing processes on which we base our approach, there is often a need to tailor the way we work.

I was recently discussing the importance of testing Magento Commerce, and the differences in approach to other platforms, when I was asked a question which I was very familiar with; “If we’re taking a platform that is fully tested out of the box, and we’re plugging in already tested extensions, then why do we even need testing in Magento Commerce projects?”

And this is a question that not only applies to this product, but other products or platforms which are sold as fully tested core packages. So why should we test something which is “out of the box”? Great question!

 

What does “Out of the Box” even mean?

Out of the box is a term that is “used to refer to the immediate usability or functionality of a newly purchased product, typically an electronic device or a piece of software.” So by this definition, it’s understandable that people would question why it would need an additional level of testing, but let’s think about a real life example.

 

Building from the ground up…

Let’s consider a scenario where you are investing in a new build home. You have secured your plot, seen your house plans and visited the show home. You have specified all your extra requirements in terms of materials, fixtures and fittings and you wait eagerly for the build to be completed. You are likely to visit throughout the build process to see progress, discuss any issues or changes as they arise, but you are dependent on your contractors to carry out all build work for you to a high standard.

Let’s assume for example that you have selected a standard bathroom suite but opted for a different model or brand of electric shower to those usually recommended. Before you come to move in, you are relying on your appliances to have been installed and confirmed as working. If at the point you come to use your new shower and find that it is working but there’s only cold water coming out of it, you become the person triaging defects to get them resolved.

The point here is that installing a shower is in theory low risk because hey, it’s a shower, we install them all the time, right? It’s only when you come to use it that you find something doesn't work due to a fault with either that specific product, or how it is installed in that particular setting, that you're then going to have cold showers to look forward to when you get the keys!

If you’d have had the option to have some additional testing done to guarantee that everything has been installed properly, is working as expected "out of the box" and fully meets your requirements, would you? You are likely to say yes to gain some additional confidence in your investment and more importantly peace of mind on move-in day.

 

How does this apply to Digital Testing?

In the Magento Commerce example, we offer development services that help clients to get set up with the core product, plus combinations of extensions from varying sources that are specific to the client and their use cases. This development also then includes elements such as data feeds and additional configuration that begins to take the standard Magento product and turns it into one that meets your requirements.  Testing here plays a vital role in verifying these requirements have been met, confirming that your now customised product does what you need it to.

As pointed out in the real-life example, the same applies. If you just install an extension into Magento and assume it’ll work as expected (because of course the extension has been tested before it is released to market) things will be great. Potentially it will and as before, the water will come out, but most of the time it’s not always possible to check every possible scenario due to time or budget constraints. Unfortunately, in the world of testing, we see that a lot of the time basic requirements are not met where things are left to assumptions. Ensuring there is a good solid test strategy defined means that things don’t get left to chance, leaving you with a shock when things go live. We build confidence in quality and reduce these risks throughout your build process.

It also allows you to get into the minds of your core users, which can be defined by identifying your user personas, as Lucy in our Experience team has identified in her post.

So, before you start to cut down your testing scope, here are some points you should think about if you are asked to reconsider your approach:

  • Any customisation or change to the standard configuration (no matter how big or small) is a change to what is tested “out of the box”.
  • Testing for base or core product packages focuses only on sample use cases – not real-life scenarios specific to those of your business or your customers.
  • Data can impact the way a system behaves, especially in terms of presentation of data, data validation rules and performance.
  • Configuration is always different depending on the client, industry, use case, and even within the same industry no two clients will be the same.

If you’re struggling to define your test approach, why not talk to us? We’re here to advise and guide you on how you can give the right things the right level of testing focus to make sure quality is not compromised.


2 people looking at phone

How to Conduct a UX Audit

What is a UX Audit?

A UX audit is an analysis to discover how users are interacting with your website, product or app. This is usually done to optimise your site for better conversion or better user experiences. The audit will provide recommendations for tests and improvements that can be made to your site. These suggestions are based on the data and research found through the audit. An audit will help you uncover some of the struggle's users are having on your site so then you can enhance the experience.

 

What is the Benefit of Conducting an Audit?

Great user experience is something that customers are coming to expect from brands. Conducting an audit is the first step in improving those experiences. The value of UX can be seen in the numbers “Forester Research shows that, on average, every dollar invested in UX brings 100 dollars in return.” (Forbes.com 2017) Eclipse clients have experienced this themselves for example, on the first round of the conversion rate optimisation (CRO) programme, HM Post Office reported cost savings of over £250,000 within the first 6 months.

 

What to do Before Starting the Audit

Define business objectives – It’s important to understand the goals of your organisation so you have an aim for your audit’s solutions and outcomes that you can measure against the objectives. Without them, the report is subjective and not measurable. Usually, the goals are around conversion rates and user satisfaction.

Decide on your resource – The amount of time and money you can commit to an audit will have an impact on the output. If you choose to do an audit in two days this will not have as many actionable insights but maybe you need a quick turnaround. Make the decision that is right for your business before you get started.

 

The Audit Processes

Find out who the users are – Define who your users are, their demographic, how they’re getting to your site, and what device they’re using to access it. Platforms like Google and Adobe Analytics can be great for getting this information. Alternatively, your business may have a set of user personas, these will be helpful at this stage. If not, you should look at creating some and this post gives you everything you need to think about when creating them.

Analyse user behaviour – This can be done by using tools to track user's behaviour on your site. You may already have tracking and analytics on your site, or you can set it up at this stage. You should look at heatmaps, mouse flows and screen recordings to see current customer flows to identify any patterns or pain points on the site. This can highlight any pages where you see lots of users exiting the site or getting lost or stuck on a journey.

Collect user feedback on your site – Feedback collected from users can be cross-referenced with the behaviour analysis to back up comments and claims made by users. Many tools can be used to do this, for example, Usabilla and Hotjar.

Review performance of the site – Test the loading speed and check for errors. A slow website can be a disaster for your business with users leaving before getting past the first page. Errors and bugs will cause trust issues and identifying these problems so that they can be fixed is vital to improving your website.

Competitor analysis – Look at industry trends and competitor solutions. Seeing how competitors' websites or products compare to yours gives you an idea of what users expect from your site.

Go through the basics – Check the accessibility, for example, are the colours and font sizes appropriate. Making your website or app accessible will make sure it can be used by as many people as possible.

Analyse page hierarchy – Is it obvious what the user needs to do next. Common problems users have on a website are not being able to find the information they want or not knowing what tasks they need to complete. Spending some time looking at the page hierarchy will help you to spot if this is an issue on your website.

Page Analysis – Look at each page of the website and rate it based on best practice solutions. The main areas to focus on are:

  • Header
  • Footer
  • Search
  • Homepage
  • Product listing page (PLP)
  • Product display page (PDP)
  • Basket
  • Checkout
  • Account
  • Accessibility
  • Performance

Break down these areas further within each section for analysis. For an example of this type of analysis, take a look at our report ‘The state of UX’. It looks at some of the big players in the electronics market and how their sites are stacking up.

Opportunities – Look for other improvement opportunities and usability issues that come up on reviewing the site, make sure you capture these as you go.

Review functions – Look at how well functions such as search and filter work. Is there any room for improvement?

Complete report – Once you’ve collected all the data it’s time to put it into a digestible format. This will highlight the key findings and present hypothesis and solutions to test and implemented on your site.

 

What Next?

Hopefully, you found this useful and you now have a set of hypothesises to test but what’s next? Once you have all these ideas it’s time to create a backlog. You should fix any critical usability first. Then categorising all the issues by the effort required to put them in place and by the impact they will have on the business goals. This will determine their priority.

If you would like some help conducting a UX audit or implementing and coming up with solutions for problems that you have found, we would be happy to help. Get in touch and we can find a solution for your business.


Man on laptop

What is User Testing and Why Do You Need it

Anyone who creates a product or designs a website, whether that be for information purposes or as an eCommerce experience, wants it to be the absolute best it can be. That’s why we put it into the world. No one sets out to create the worst possible experience for their end-users but how do we know it is the best possible experience?

Sure, we can look at it as individual developers or designers and maybe bring in the wider design and development teams but this verges on self-gratification and doesn’t really give you much beyond the fact that you all think you did a great job.

This might sound obvious but surely engaging with the people that are either using or you want to use the product or service you’ve developed is the answer to finding out if it is any good? Well, you’ll be surprised just how many businesses choose not to do this and it is blinding their future development and improvements.

This idea of talking to the end-user isn’t new. Market research and focus groups have been the grounding to some of the biggest advertising campaigns and certainly changed the destiny of several feature films released over the years. What is new is the technology that is being used to achieve this information and the personalisation of that testing to achieve much more detailed data.

 

What We’re Talking About is User Testing.

User testing is when you analyse and measure the experience of a user that is engaging with your product or website either as a whole, or just a portion of it.

In its most simplified explanation, it’s testing and quantifying how someone uses your product, service or website. This in most cases is different from what you think they should be doing with your product or website.

User testing records and observes a set of user’s performing tasks with your product or website to find errors, frustration points and potential areas for improvement to your future product or website development. It can also involve asking a user how the quality of the experience was for them.

Often done with users in location, user testing can also be done remotely which is especially handy when you consider the current various states of lockdown, we find ourselves in. This should not be seen as a blocker to taking to your users.

 

Why User Testing

Outside of getting the valuable information you need from the end-user, several other benefits come along with user testing. At the root, user testing can help save you money, save you time and increase user satisfaction for your business, product or website.

 

Time-Saving

It may seem like an odd suggestion to think that adding another process to your development or design journey is going to save you time, but when you look at the long term the time saving is abundantly clear.

User testing saves you from making mistakes now, rather than later. The longer you stay in isolated product development and planning with your product or website, the longer it will take you to fix issues and functionality down the line. This is especially true in applications and web design, the more complexity and features you add into a product or website, the harder it gets to fix newly discovered problems. And so, it kind of goes without saying but the harder it gets, the more time it takes to unpick and fix them.

 

Money-Saving

Much like the idea of saving time, the suggestion that engaging in user testing which has a cost to it can save you money might sound ludicrous. It isn’t.

If we look back at the long-term benefits, it is clear to see that identifying potential issues early allows them to dealt with before they come intertwined into other developments or feature enhancements.

Rewriting code or redesigning layouts are far from free. Every backtrack or re-work takes hours out of moving forward and those hours have a cost attached to them and they can be substantial especially when you are starting all over again. Finding a problem before a product is completed is almost always cheaper than fixing it later.

And even though you will need to spend a little to undertake the user testing, doing it with a team of experts who can get you the best possible results the first time is another way to ensure that the process is saving you money, making it as cost-effective as possible. Remove the need to try and try again through a period of self-learning and gather valuable insights fast.

 

Increased User Satisfaction

This is a more obviously understandable benefit. Taking feedback and working it into the development and usability of your product or website is going to ensure that when others engage, they won’t be faced with the same issues those in user testing experienced.

Products or websites that focus on user experiences have been proven to increase customer satisfaction. It shows that you care deeply about the experience your users have with your product or website and improving customer satisfaction with enhanced user experience will likely have an impact on sales and your bottom line.

 

User Testing Works

If you need more of a reason to undertake user testing, you only have to look at the results. Usertesting.com has many case studies that highlight the direct benefits of taking this process on.

Zillow increased engagement and conversions by over 8%, Evernote increased user retention by 15% and Walmart Canada increased onsite revenue by 13% as a result of tailoring experiences.

And at Eclipse we’ve been performing user testing for many years and have worked with some of the most well-known brand names in the UK and across the world. Some of our existing clients have found it so beneficial that it is now just included as part of their regular process.

 

How Can Eclipse Help

Hopefully, you can now see the benefits that user testing will bring to your business, product or website and you’re wondering what you need to do to get it going.

The good news is that we’ve developed some ready-to-go packages for remote user testing that we can roll out for you in no time at all. All you need to do is take a look at what is on offer, explore even more of the benefits that can be achieved with user testing and then reach out to our Experience team so that we can make arrangements to gather those valuable insights for you.


Group of people looking at view

How To Create A User Persona

User personas are a great tool to improve designs and communicate design decisions. Let’s take a look at what user personas are, why they are useful, and how to create your own. 

 

The user persona

User personas are profiles that capture the most important data from a user base. This data is then used to outline the archetypes of your common users. These are usually one-page documents. They are used as a reference point to remind you who your users are and give you a way to communicate this to others. They help you to understand who you're designing for. Giving you an insight into what the behaviour and thought process of that user is. Explaining why users take certain actions in your product, and what they're hoping to do when they use it. 

People across the business in different roles should use user personas. Stakeholders, senior managers, and product owners can all use personas. They can use them when analyzing the software for new features or prioritizing bug fixes. Designers, copywriters, and developers can all use user personas too. They use them to find the best approach and solution that fits the needs of the user.

 

Why user personas are useful

User personas are beneficial because they are a way of empathising with the people who use the software. They put the user at the centre of the design process. If you are always looking through the eyes of the user you’re more likely to create a solution that works for them. 

They are specific and this gives you clarity about who to design for and who to prioritise. You are thus able to meet the needs of groups of users with similar characteristics. As opposed to creating generalised solutions that don’t fit exactly with any users’ needs.

Designers often fall into the trap of designing for themselves. Having a user persona holds you accountable for your design decisions. If they aren’t in the best interest of your user persona then they may not be the best solution. You can also use them to communicate and justify those decisions to stakeholders and clients.

Another benefit of creating personas is that they anthropomorphise data. Adding human characteristics and behaviour to data it makes it easier to understand and remember. It’s great having a ton of useful data but if no one ever references or uses it then it’s worthless. A concise digestible format that doesn’t include unnecessary information, will help you get the most out of your personas.

 

Creating a user persona

Begin by segmenting and creating hypotheses about the different types of users. You must separate your users by the different roles they have. Let's take the example of project management software. This would have users with two different roles. One user is the admin that would be creating tasks and reporting on the project. The other user is the employee who would be using the tool as a reference and to track time. These two roles need separating as they have big differences. Within these roles, you can begin to create more nuanced profiles. Through a combination of analytics and research, you can validate these hypotheses or disprove them. Both are equally illuminating.

Take a look at our example of an admin persona for a project management tool:

Do your research

User research is critical to understanding the experience of users. There are many different ways in which you can collect and present feedback. Likely, you're already getting a lot of feedback if you have a product that's up and running. 

Interviews and observation are the most common research methods. Both of these research methods are qualitative. They need analysis that looks for patterns and commonalities between users.

At this stage, it is critical to learn about the motivation of your users—what problems are they looking to solve when they come to your site? Different personas will have different reasons to use your site. You can start by developing hypotheses about what drives each user to engage. 

 

Your research should focus on the following aspects of the user experience: 

Bio: What does this person do? Are they always rushing around with lots of things on their mind? Are they worried? Planning an adventure? 

Motivation: What drives your user to interact with your product? What are they hoping to get out of it? Why are they using your product instead of a competitor’s - or nothing at all? 

Pain points: What are the challenges users are facing? Is your product helping them solve these or aggravating them? Are there any obstacles they have to face when using your product? 

Mental models: How does your user conceive the problem that your product addresses? What concepts and connections come naturally to them, and what do they need teaching? 

Personality traits: Is this user more of an introvert or extrovert? Are they an influencer or a follower? Are they loyal to brands or are they more fickle and drawn towards other features or lowest cost? 

Internet usage: Internet usage is an indication of online behaviour. Are they a regular online shopper? Do they mainly browse at home or during lunch breaks? Are they particularly active on social media? 

Brands: Are these users used to dealing with high-end premium brands or value brands? Or is it a mixture of both? Be specific to the actual brands this user buys from to understand their brand relationship. 

 

What to look out for

There are some risks associated with user personas. They are usually from not following the user research process correctly or thoroughly enough. One of the main areas where people tend to go wrong is by using too much second-hand information. This could be from stakeholders within the business or stereotyping. In these cases, research and data are not informing the personas.  This type of information runs the risk of missing out on actually user insights. You may be attributing characteristics to your users that aren’t accurate. Using these inaccurate personas when developing the product will mean you won’t meet the needs of your actual users. 

 

Conclusion

User personas are a great tool to guide your design decisions. They can be used to inform other models such as user journey maps, usability reviews and user stories. The entire business should make use of the personas. This will help to create the best possible experience for your users. The key to successful user personas is good quality research. There are risks to creating personas. You can mitigate these risks by following the correct research processes. 

There’s also no such thing as ‘done’ when it comes to the personas. You should revisit, review and update them regularly as economic and social climate changes users' online behaviours.

Hopefully, you’ve found this useful and feel inspired to create your own, but drop us a line if you’d like a little help – we’d love to talk to you.


Website Speed Tips

6 Tips to Increase Your Website Page Speed - Part 1

As anybody who knows me will tell you, I’m a huge self-confessed web performance geek, I love websites that download fast, I believe that fast internet is a basic human right and I think hotels that claim to have WiFi but instead have LiFi (WiFi that doesn’t quite seem to download anything) are criminal and should be boycotted.

This means that I’m always on the hunt for a good website performance technique and I thought I’d share 6 of the best web performance techniques that work every time.

Reduce the size of your images

The most commonly overlooked technique to get a faster website, can not only give you some of the biggest rewards but it’s also the easiest technique to implement.

There are a few areas in which you can reduce the size of your images:

Reducing the aspect ratio
This sounds very obvious but a huge amount of websites I’ve optimised and found that their main banner at the top of every page is 5000 pixels wide but is being displayed at 20% that. Just by reducing that you can save yourself a huge amount of bytes and therefore save your users from a huge download for no real benefit.

Make sure you’re showing the right image for the device
If your customer is viewing your product on their mobile, show them a version of the image that’s no wider than the device. When CSS is used to reduce the size of an image, it’s still the same size to download, just the browser then has to process it to the size required by the CSS, adding more time. Consider looking into srcset which is becoming more widely supported.

Compress the image

Tools like TinyPNGTinyJPG and ImageOptim can save your users from many kilobytes of download without reducing the visual of your image in any way your users will notice. This means that you can save a large amount of bytes on your page, without actually impacting your users. If you can have this built into your media upload, you don’t even need to think about it.

Reduce the ‘quality’ of the image

By reducing the quality of our images, we’re really talking about simplifying the amount of colours being used. This means that we can not only reduce the file size but also making tools like Gzip more effective. Other more advanced techniques of this include reducing the quality significantly to the point where you can see but saving the image at double the resolution, this has been proven to deliver even smaller file sizes without impacting the asset visually.

Remove old and unneeded code

Technical debt is an unavoidable part of any development, making sure you go through and remove old CSS, HTML and JavaScript is hugely important, though.

A few common areas that get overlooked regularly:

Removing vendor files

Pulled out an old version of a carousel or a split testing tool, make sure to pull out the assets as well, that includes the JS and CSS that goes with it.

Removing losing variants of split tests

Very often I’ve seen examples where 2 or more versions of the same element have been created and tested. After this has been analysed and a winner decided, the traffic is pushed down one stream to the winning variant only for the old versions to be forgotten about. Make sure to remove any old code associated with the losing variants of the split test.

Not including the entire framework

This is the most common cause of removing unneeded code, when people include frameworks like Bootstrap or Zurb, they very often include the entire framework, this means all the JS and CSS that build elements that the website might never even use. You can build these files up as and when you need them by not including the unneeded Sass files or only downloading part of the JS. For instance, the entire bootstrap CSS and JS is 668KB whereas the bootstrap grid is just using the grid is 13KB. You can use the customize tool on the Bootstrap website to do this for yourself and tailor the download to your needs.

Prioritise your assets

This is equally as important as anything else, make sure you’re downloading your assets as and when you need them. The simplest case of this is making sure to put your CSS in the head of your file and the JS in the bottom.

Even more than this though we can look at preloading pages and lazy loading.

Preloading

Preloading is when you download an asset before the user requires it and store it in cache or local storage, this means when they get to the area of your website where you think they might need it it’s already available and you don’t have to download it. A good example of when this could work is if you have a booking flow on your website, your booking flow will probably have some extra functionality, so before the user clicks on the search button on your booking engine, you could preload some of this functionality meaning you can load the page quicker and hopefully convert that user.

Lazy loading

Lazy loading is the other side of this, delaying your assets on the page the user is currently looking at until they actually need them. If for instance, you’ve got a gallery on your page, you don’t need to load all of the images the moment the user lands on the page. A more optimal way of doing this would be to download nothing at first, this allows for the first render to happen. Once we’ve displayed the page we download the first couple of images, and maintain being 1 or 2 images ahead of the image in view. This means you’re not downloading images that your user might not ever see but still make your users feel everything is instant as they slide through your gallery.

This is part 1 of a two part series, read part 2 here.

Everything I’ve said about are all web performance techniques that actually work, I’ve implemented them, I’ve seen the results and I can vouch for them.

I’d love to hear your stories implementing some of these results, equally, I’d love to hear some of your web performance techniques that do or don’t work, so please reach out in the comments.