How to Effectively Drive Revenue Per Visitor to Boost Online Performance
What’s the main reason we all do conversion rate optimisation?
To try and improve the revenues and profitability of our websites to make more money. Crude as it may sound, it’s why we spend days analysing the data, talking to customers and creating hypotheses.
And we thrive on it. We love creating and scientifically testing those hypotheses through A/B split tests to prove beyond any doubt that one version was more effective than the other. We are also more than a bit addicted to refreshing the live reports for each test to see if we’ve created a winner or not (go on, you can admit it, we’re all CRO addicts here). We get a kick out of carefully watching the conversion rate, willing it to move upwards… it can be quite exciting.
Trouble is, what if we’ve been looking at the wrong thing all this time? What if conversion rate isn’t actually the right metric to track?
Sure, it’s important because it tells you how many of the people you are getting to your site are actually buying something… but it doesn’t tell you how much money you’re making. It doesn’t tell you about your revenue or profitability.
To do all that, there’s a bigger, better metric in town: revenue per visitor.
What is revenue per visitor?
First things first. Revenue per visitor (or RPV) is exactly as it sounds: the amount of revenue made, on average, per visitor to your site. It’s generally used to measure the value of each of your visitors as an average to see where you could concentrate further traffic generation efforts.
For example, if you are looking to invest in further PPC ads, the RPV would tell you what you can afford to pay per click. When you combine it with different traffic sources, devices, landing pages (and many other data points) it becomes particularly useful.
How to measure RPV
To measure your RPV is quite simple, just see how much revenue you made during a specific time frame and divide that by the number of visitors you received in that same time frame.
So, if you get 100,000 visitors per week and you make £250,000 in revenue in that same time frame, your RPV will be £2.50 (£250,000 revenue divided by 100,000 visitors).
These numbers are easily available in Google Analytics (if you have ecommerce tracking set up for your revenue), otherwise you can obviously get your visitor stats from GA and then use your back end system to get the revenue numbers.
Like conversion rates, revenue per visitor will go up and down constantly. So, unless you’re tracking specific spikes/troughs due to a specific short-term event, such as a new marketing campaign or Black Friday, it is best to average this out over some time to flatten any potential outliers.
Why is revenue per visitor important to boost online performance?
Great, you have your magic RPV number…. So what’s next?
Well, let me tell you why you should kick your conversion rate percentages off their pedestal and proudly put your RPV up there instead if you’re looking to boost your online performance.
It’s because your conversion rates only tell half the story.
That is to say, that conversion rates are only one of the two main metrics that matter for your calculation – the other being average order value (AOV).
To illustrate the point, let’s look at some example A/B test results for a fictional company:
As you can see, conversion rates have stayed the same across both versions after a few weeks, so bin the test and run the next one. Right?
Here’s what happens when you bring AOV and RPV into the mix:
So, ignoring the AOV and RPV would have cost this business £22,500. Every 2 weeks. That’s £585k in a year (without even factoring in a Christmas spike)… that’s a lot of free money to leave on the table.
But sometimes things are even less clear cut than that. Here at AWA, we run a huge number of tests and we regularly see a portion of those where the conversion rate actually drops with the variation – but it still makes more money because the AOV has increased so much.
But, it can also work the other way around and just tracking your conversion rate can lead to a false positive (where you think it’s a winner, but it’s not). Take a look at this scenario:
Had you focused on conversion rate alone, you may have implemented the variation, invested in development and QA time, for no additional financial reward. Not the best use of resource I’m sure you’ll agree.
So, as these few examples show, tracking RPV is a much better way of evaluating performance than conversion rate alone. While conversion rate is still important to gauge performance of a page (after all, we do still want more people to buy), it should not be taken in isolation.
Revenue is still king – and RPV is the best way to track that.
Ecommerce performance optimisation strategies to boost RPV
So, now we’ve established what revenue per visitor is and why it’s so important, let’s look at some of the ways in which we have moved the needle and boosted online performance for some of our clients the last few years.
Just before I jump right in, it’s important to note that we don’t do lists of “best practices” to follow here at AWA.
Sure, there are always some things that we might check on a site that we have learned through our combined years of experience, but, cliched as it sounds, every site is different. So something that worked for us a couple of times, may not work for you.
Below I have included a list of themes where we have had success and how we developed these hypotheses – and the results. But this is not a list of tests you need to run; your research will indicate which area has the best chance of giving you a sizable uplift.
The term “value proposition” is often totally misunderstood as just being a branding exercise that people don’t really care about and no one bothers to read.
In fact, here at AWA, it’s often one of the first things we look at for our clients – with big results. It’s such a big topic that it deserves its own mammoth blog post, but let me briefly explain why you should focus on your VP:
Firstly, dispel the myth that a value proposition is branding. It’s not. It’s an articulation of why a customer should do business with you. You are answering the question: What’s in it for me?
This is the formula to use:
Value Proposition = Perceived Value – Perceived Cost
Your VP should tilt the balance so that perceived value is much larger than the perceived cost. That’s what people now refer to as a “No Brainer” – it’s so obvious that you should be doing something that it doesn’t even warrant further thought.
That’s the power of a good VP.
There are simple things you can do quite quickly to help shape your own value proposition:
- Survey your customers – there’s many ways you can do this, but an email survey via a platform like Mailchimp is quick to set up and will handily compile all the results for you for easy analysis.
- Ask them what matters – the information you’re trying to get from the survey is what really matters to them when buying from you. What do they like? What do they dislike? And the best question to ask: “What, if anything, nearly stopped you buying from us?”
- Do a mix of multiple choice checkboxes and open-text questions so they have the space to say what they really think – it might be something that you haven’t thought of.
- Look at the patterns and see what matters to them – then put it in your VP, either as a whole sentence or quick bullet points.
- Get your VP on your site in a prominent place (the header/hero section is a good start), weave it into your emails, social media and PPC ads.
To see how effective it can be (and to give you some ideas), have a look at this case study from Paperstone, where we got a 28% revenue increase from working their VP and header.
When you consider the role of the navigation it becomes clear that you need to think about it in a way that can boost or hurt your revenue per visitor.
The navigation is one of the, if not the most, clicked parts of your website. So it makes sense to ensure it is simple and easy to use so your customers can get to exactly what they need easily, quickly and confidently.
If they can’t find what they need, potential customers will just hit the back button to their Google search and browse through the other 843,000 search results.
Here’s some things you could do with your nav:
- Keep it clean and consistent – nothing fancy.
- Divide categories clearly and accurately – and use the language your customers do.
- Make all navigation elements clickable links and double check they are going to the right place.
- Ensure every clickable image has ALT text.
- Ensure your search feature works well – test after test shows that people who use search will convert better (but only if your search gives back good results!).
These were some of the problems that Sunspel were having in their navigation. By researching the customer journeys and watching live recordings of how users were navigating the site, it was apparent that users found that the current signposting was hindering and not helping users find what they wanted.
The photos you use, especially on product pages, can sometimes have a big impact on your ecommerce performance. Customers are now used to being able to see a range of big, HD images that they can magnify – maybe even a video or 360 degree view as well.
Sorry to say, but if your product imagery is a bit tired in this day and age, it’s likely to be hurting your revenue per visitor. We’ve seen many examples of products with better images outselling those with comparatively poor ones, regardless of whether the products are better, worse or the same.
Feedback we were getting from our clients’ customers following all of our surveys was a complaint that the product images were not doing a good enough job of showing the products. The photos were very nice, but the problem was that all users could see was the same single image – in three different sizes of magnification. So we tested a reshoot of some of the most popular products, ensuring there were at least 4 different images of the product. The results confirmed our hypothesis with our version generating an uplift of +10% RPV with 97% confidence.
Unfortunately, unless you sell products from other manufacturers who have already got imagery, the only way to solve the problem is to re-shoot everything. It takes time and budget to do it well and have great shots – but that’s why it’s important to test a sample of improved shots before committing to a whole reshoot.
As this whole list shows, there are many things that influence why someone would buy something from you, but in certain situations the most important of all is your copy.
If you are selling an item that can be bought anywhere, then your product copy is less important than factors such as price, delivery, returns, your value proposition and UX (how easy it is to buy from you).
However, if you are selling something more unique, then copy comes right to the fore. Then, you have to actually sell it. Regardless of your delivery and returns policies, no one will buy it if they don’t like the sound of it from what you’ve written.
One of the best examples from our archive is when we worked with Cox & Cox. They have a great website, which was already quite well optimised, but our research and analysis showed that users were not as engaged with the product pages as we would expect them to be.
Further analysis suggested that the copy was not structured in a way that made sense to the user, leaving people a bit overwhelmed and confused. So, our hypothesis was that a cleaner structure with more persuasive copy would boost engagement and conversions.
The results backed up our findings, with the test showing a 36.7% uplift. For speed of implementation, we only tested on a small range of items, but the concept was quickly rolled out site-wide.
To see more on this one (and some of the tests we did) you can read the full Cox & Cox case study.
I would suggest you go back to the top of this section and re-read the bit about value proposition. Your product copy should be like a mini VP, based around that product.
- What does it do?
- Why is that unique?
- Who is it for?
- What does that do for me?
Be specific and back up your claims with facts.
For example, what works best:
“A lovely soft jumper”
“Our super soft jumper – made with 100% cashmere – has become an instant top seller”
Don’t be afraid to make claims or sell the value of the product. Use whatever evidence you have to reinforce the main selling points and let people picture having it themselves, rather than browsing for it online. One of the best ways to do that is through using carefully planted testimonials, that reinforce whatever key claim you’re making (more on this further down).
For an excellent insight into copy that converts, you can read this article from Mel, our Head of Creative, “How to Use a Copy Review To Increase Your Conversion Rate”.
As we demonstrated in the examples at the top of this article, one of the best ways to increase your revenue per visitor is to increase the average order value.
And one of the best ways of doing that is by upselling your customers.
Got a pair of shoes in the basket? Maybe I could persuade you to buy wet weather repellant spray and some insoles…
Got a new football in your basket? Maybe you’ll need an air pump to inflate it…
Booking some flights? Maybe you’d like car hire and a hotel at your destination…
We’ve all seen these types of offers hundreds of times, but there’s a really good reason why some of them work and some are just plain annoying: relevance.
I’ve been on sites shopping for my 6 year old niece and been presented with Prosecco bath bombs and extra large G&T glasses. Maybe they made the assumption that I have young kids so I might need a large drink every now and then… but probably not.
Relevance is key. If it fits really well with what I am buying (and it isn’t too much extra money) there is a good chance I will add it to my basket too.
One of my favourite examples comes from our work with Kitbag, the sports / football kit ecommerce store. When selling a football kit, the margin comes from any upsells you can generate by the time the clubs, sponsors, league and manufacturers have had their cut – and the best upsell is name printing on the back of the shirt.
Our analysis showed that users were not clear as to how they could personalise, add their name or their favourite player’s name, to the back of the shirt. Our hypothesis was that by making the UX on this key upsell much simpler and intuitive that we could increase the number of people who personalised their shirt and therefore increase basket AOV.
It worked. Not only did 14.4% more users complete their order, but the AOV jumped by a whopping 40.6%. That’s two big wins for their revenue per visitor!
Delivery (and returns)
There are certain ‘checklists’ that online shoppers have before they will enter their payment details. Without them being satisfied that all their boxes are ticked, the sale won’t happen.
The biggest can often be about delivery. And depending on what you are selling, it can have a huge impact on your revenue per visitor.
Firstly, it has to be clear and easy to find. You would be amazed how many ecommerce sites have it buried somewhere in the 3rd page of their checkout and wonder why people are dropping out. Even if you do not have a great delivery or returns policy, you still have to label what they are clearly. The only thing consumers hate more than paying over the odds for these things, is not being able to find out what the prices are in the first place…
But that’s not the only potential problem with delivery. One of our clients, Thompson & Morgan had a big problem about how their products were delivered – not about the cost or time it took. What’s more, they didn’t even know it was an issue until our mammoth research phase was in it’s very final stage.
Thompson & Morgan sell seeds and seedling plants to their customers and it turned out that one of the biggest reasons why conversion rates on plants was fairly low was because customers were worried about the condition of the plants when they were delivered. Would the plant still be alive? Would it have been knocked about it transit?
When Johann, the lead optimiser on the project visited the warehouse, he was talking to the head of dispatch who told him about the 6-point checklist they go through when packaging every plant to ensure it arrives in top condition. The website did not mention any of this and when we simply weaved that checklist into the site, user anxiety dropped enormously – and conversions went up.
You can watch the full case study for Thompson & Morgan here.
Finally, it’s also worth ensuring that your returns policy is just as clear. The nature of any mail order business is that people cannot feel the product before buying it. This causes anxiety, which will hurt your conversion rate and AOV. However, a clear, friendly returns policy can help to relieve that.
Personalisation or being relevant to your website visitor is key. Research from Smart Insights suggest that around half of us spend more in a personalised experience.
Sometimes it takes time and some technology to get personalisation right – you have to know who they are and you have to have some data on what they like to do. There are many ways to do personalisation – some of them are very expensive, such as investing in a bespoke platform to do it for you. And others are free, but do involve some effort.
For a detailed look at the options, see this previous blog post How to personalise your website – no matter what your budget.
Personalisation can take many forms and is often a source of inspiration for us when thinking about hypotheses.
Some are quite simple, such as the geo-personalisation that we routinely adopt for Canon Europe. All of their European ecommerce stores are 99.9% the same (except translations and currencies) to make management much simpler. But there are some differences. Such as when we discovered that Paypal credit was a big plus in Germany when buying DSLR cameras, but not so much in other territories – so now Germany sees a slightly different checkout with Paypal much more prominent.
Other implementations of personalisation can be more complex than this, of course. But the point is that all personalisation needs to come back to what is working best for each of your audiences – and giving them that experience. Personalisation technology, like marketing automation, is only as good as the content and thinking you put into it.
And you will only find what those winning experiences are by research, analysis and testing lots of informed hypotheses.
Everyone knows that having positive user reviews can help increase your revenue per visitor. But few people know that using those reviews at the exact right time and in the exact right way can boost those results even further.
Picture this: you use a tool like Feefo or Trust Pilot to encourage your users to leave reviews. Fortunately, your reviews are very good so you put up 4.9* rating on your site and even have a carousel of the latest reviews pride of place in your customer journey. So far, so good.
Now think about how much more effective it would be if you could cherry-pick the most relevant reviews to show at each part of your customer journey. If you know users are worried about returns, show reviews that celebrate your no-quibble returns policy. If they are worried about how comfortable your shoes are to wear, have a review that talks about the “unbelievable comfort straight out of the box” on the product page. There are countless examples, but hopefully you get the idea.
Now, as with all of these approaches, there is no “copy and paste” template that will work for every website, so let me give you an example of how we have done it with great success.
We recently worked with Discover Car Hire, a price comparison site for rental cars, who were quite new on the market and had not yet built a trusted brand status among consumers. Prices were good and the service was good, but people were reluctant to put their trust in a new brand – after all, your hire car not being at the airport to meet you can get your holiday off to a bad start.
We analysed their Trust Pilot reviews in detail and found quite a few talking about the quality of service, reliability, how they have become people’s most trusted car hire firm and so on. We then used these reviews (not a carousel) on the main checkout pages to reassure new customers. This one small test increased conversion rates by 17.55%.
So, whatever other metrics are important to your organisation, it’s revenue per visitor that most ecommerce businesses should be judging themselves on, not just good old conversion rates.
Of course, there are other metrics that a business will be interested in, such as operational efficiency and cost savings that are worth just as much to a P&L sheet (and we’ve done tests for each of those with some clients), but we find RPV is a great place for optimisers to focus.
After all, everything we do is about driving further revenue from the people who are already coming to a website. Our reason for existing is that we passionately believe it is far more profitable for a business to concentrate on converting the people they are already paying advertising dollars to attract – not just throwing money at attracting more people.
That’s what our own conversion rate optimisation services are based on.
Finally, a bonus point: The only time we say you should always listen to best practise…
You know I said that you should not listen to best practice because every website is different? Well, there are two ‘golden rules’ when running any optimisation strategy – and you should always stick to them. They are:
- Do your own research. Research your analytics, your heatmaps, your customer surveys to try and find the answers you need to create relevant hypotheses. Use your data to draw conclusions and come up with ideas… it’s why blindly copying what others do is a road to more losses than wins. As an idea, we use up to 32 different research methods for every site we work on.
- Test everything. Never presume that something that has worked for one person will work for you and just roll it out without testing. To highlight this is a brilliant story about a car hire company that we were working with. Their own UX team had redesigned their checkout and wanted to make the full roll out straight away as it was obviously so much better. We persuaded them to test their version against what they already had – so they would have the hard numbers to back up their assumption. You guessed it, it was a complete failure – if they had rolled that out without testing, they would have lost over £7.5M in the rest of that year alone!
That’s it. These are the only two things we strongly urge you to do every time.
Let me know how you get on with some of these ideas and what metrics matter most to your business in the comments below.
To learn more about how to make your conversion optimisation programme a success and ensure your website is easy to buy from, read our ebook below:
People from Facebook, FarFetch and RS Components receive our newsletter. You can too. Subscribe now.
If your CRO programme is not delivering the highest ROI of all of your marketing spend, then we should talk.