Jump to content
Slate Blackcurrant Watermelon Strawberry Orange Banana Apple Emerald Chocolate Marble
Slate Blackcurrant Watermelon Strawberry Orange Banana Apple Emerald Chocolate Marble
Sign in to follow this  
Rss Bot

Master the science of conversion rate optimisation

Recommended Posts

Conversion Rate Optimisation (CRO) is the process of maximising conversions from existing traffic. For example, if you get 1,000 visitors per day on average, and have a 1% conversion rate, that means 10 of those visitors will convert. CRO is the process of trying to optimise your site, to improve that rate of conversion to 20 users (2%) for every 1,000 visitors on average, for example. The success of your ecommerce website depends on mastering CRO.

Imagine that you’re the owner of your town’s first supermarket and you notice that your large store is selling less impulse products (think chocolate and soft drinks) per customer than your smaller corner shop competitors. This is despite your supermarket offering a much larger and higher quality selection. Why would this be the case?

After examining your competitors, you realise that the majority of their impulse sales come from displaying their impulse products within eyesight of the checkout while waiting to pay. As such, you trial placing the chocolate near to the cashier and notice a major boost in sales. 

Following your first success, you then take your improvements one step further and trial different chocolate brands, sizes of package and flavours at each checkout. You monitor which three products sell the most so that you eventually place the most successful products across all cashiers to further increase sales.

This is a real world equivalent of Conversion Rate Optimisation and supermarkets across the world are still trialling new layouts and approaches to increase sales of key products all year round to maximise profits from their existing footfall. You should get in the habit of applying the same logic and mindset to improving your website and attempt to convert more visitors into customers. 

CRO doesn’t just apply to ecommerce sites and product sales but also increasing user signups, newsletter signups or any other KPI that increases revenue, user experience or engagement.

  • Illustration: Neo Pheonix

01. Select your tools

There is a large selection of tools to choose from to achieve your CRO goals. Depending on your technical ability and complexity of A/B tests required, a simple visual editor to change some content, visual styles or CTA (Call To Action) elements may be all that is required. On the other hand, if you need more comprehensive A/B tests that justify an integrated service within your application, enabling you to A/B test the back-end logic or serve totally different experiences to your users, then a more advanced tool will be needed.

02. Identify your KPIs

Before you get caught up in deciding which CRO tool to use you should clearly define the KPIs that you wish to improve. For an eCommerce site, these are likely to be as simple as more clicks on the Add to Cart button, fewer exits from checkout pages and more newsletter signups. For a SaaS website, the KPIs might be more signups or more upgrades to higher membership types, for example. For a blog these may be as simple as more page views per visit or more clicks on the external ads.

03. Set up analytics and identify your baseline

The next step is vital to be able to quantify your results. The easiest way to do this is to set up goals and events within Google Analytics. Once you have completed this step, you will have a firm and reliable baseline to evaluate your first round of A/B tests against. Another benefit is that it is very motivating to know (and brag about) an actual percentage increase, thanks to your hard work.

Now for the frustrating part. Before you start your first round of A/B tests, you need to collect enough data from analytics to be able to identify if your improvements are actually working. If you have a busy site you can have this data within a day or two but be prepared to wait a week or two if you have a new website with lower traffic levels. Try to wait until you have data from 1,000 visits before making any major changes.

04. Identify weaknesses and plan A/B tests

Now that you have reliable baselines set up for your KPIs, you have much more useful data available to you within analytics and may even be able to spot major weak points in your site. Is there a specific point during the checkout that users leave? Are less than 0.2% of people viewing the product page clicking the Add to Basket button? Is a very low percentage of first-time visitors to your site failing to click the Register button in the top-right corner?

Once you’ve identified your biggest and most urgent weak points, evaluate the page, try to find the cause of the poor performance and plan your A/B tests. Is the Add to Basket button the same colour as all the other buttons on the page and therefore hard to spot? Is it below the fold and users have to scroll further down to see it? 

Think of A/B testing like a science experiment back in school. First you need to create yourself a hypothesis and then get to work proving it wrong or right. For example: “I propose that if I make my ‘Add to Basket’ button stand out more by making it a different colour with slightly bigger text and padding, then more users will notice it and be persuaded to begin their checkout journey."

05. Experiment

Now that you have your baseline performance established and a hypothesis to prove, you’re ready to create some A/B tests and put them in the wild. We introduce some CRO tools in the box on the left if you’d like a starting point for available tools. Just remember to be patient until you have enough data to make an informed decision. Around 1,000 visits should be enough to clearly see the difference in performance between them.

06. Evaluate and repeat

Finally, your first A/B tests are complete! Compare your results from the 1,000 (or more) visitors to your baseline conversion rates and identify the variation with the highest conversion rate. Put that version permanently live on your website and reap the rewards of your hard work. 

Now go back through the process, find the next biggest problem within your website, ensure your baseline data is collected, evaluate the page to see what may be the problem, create a hypothesis that you anticipate will solve it and then experiment with options!

This article was originally published in issue 301 of net, the world's best-selling magazine for web designers and developers. Buy issue 301 or subscribe to net.

Eager to keep upping your digital marketing game?

An image displaying the speakers appearing at Generate London and providing a link to buy tickets.

Sarah Parmenter is giving her talk Digital Marketing Strategies for the Busy “Web Master” at Generate London from 19-21 September 2018

If you're interested in learning more about marketing, make sure you've picked up your ticket for Generate London from 19-21 September 2018.  An award-winning designer with clients including Adobe, Ellen Degeneres, Apple, Blackberry and News International, Sarah Parmenter will be delivering her keynote – Digital Marketing Strategies for the Busy “Web Master” – in which she will discuss the idea of quarterly website design reviews with a “design once use everywhere” mantra.

She will also dig into the ever-changing world of Instagram algorithms, Facebook marketing and topical social media takeaways for immediate implementation.

Generate London takes place from 19-21 September 2018. Get your ticket now.

Related articles:

View the full article

Share this post


Link to post
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

Sign in to follow this  

×