Conversion Rate Optimisation

graph showing upwards curve

Everything below is based on the industry-standard ConversionXL framework (I’m one of the few optimisers in the world to be ConversionXL-certified).

This is a qualification based on conversion rate optimisation processes – no guesswork, just scientific methods for improving your conversions to grow your business.

This is a hefty post, so it’s split into these sections:

“What is Conversion Rate Optimisation?”
“How does it work?”
“How long does it take?”
“How do you decode what to test?”
“How much traffic do I need for A/B testing?”
“What if I don’t have much/any data?”
“What are your rates?”

“Which industries do you work in?”
“What tools do you use and how much will they cost me?”

“What is Conversion Rate Optimisation?”

Conversion Rate Optimisation (CRO) is about growing your business.

Maybe that’s more revenue, more users, more signups. A scientific approach, with analysis, testing and measuring at the heart. Because I use data to find out where your website is leaking visitors (and money). And then work out how to fix the holes. With a healthy flow of the right words in the right place. Alongside generous portions of website usability.

So your website or landing page is optimised based on evidence, rather than relying on “try this and hope”. This is why conversion copywriting works. What’s more, it’s the easiest way to generate ROI. Here’s why.

Imagine your website or landing page gets 1,000 website visitors a week. From those 1,000, you get 10 sales. Giving you a 1% conversion rate.

Let’s say you want to double your sales.You can do this in 2 ways:

  1. Double the number of visitors, so you get 2,000 visits and 20 conversions
    An extra 1,000 people. I don’t know about you, by my knees, back and brain are already aching at the thought of all that effort. You have to go out and find these people that are ready to buy, and get them to visit you. Not easy – it’s usually 4–6 times more expensive to acquire a customer than it is to retain them. You’re looking at a lot of investment (in SEO), time (unless you go viral in some way), and hard labour (blogs, social media, video production, and who knows what else).
  2. Double your conversions, so you get 20 conversions per 1,000
    An extra 10 people. Is that all? Yes. You work with what you’ve got. And optimise it. And once those extra 10 people have converted, they’re more likely to return and recommend you to others. That way your business grows naturally, with no artificial stimulants from temporary, expensive and unreliable SEO tactics.

“How does it work?”

  1. Objectives
    First, we establish your goals. Obviously you want to grow your business, so it’s a case of seeing what needs doing so that you can reach these goals.
  2. Data health check
    It’s scary how many people are making massive strategic decisions based on data that’s wrong. From including bot traffic and not filtering out internal staff, to having the code in the wrong place, or 2 versions of the code on the same page. I get the shivers thinking about it. So I don a white coat and stethoscope, and start inspecting your data. After all, to diagnose a patient, the doctor has to carry out some tests.bed in hospital
  3. Heuristic analysis
    Your website visitors have 3 questions when they land on your website. To convert them, you need to come up with the answers – within milliseconds:– ‘what is this website’
    – ‘what can I do here’
    – ‘why should I do it’Your copy and images need to answer these questions in a way that gets your visitors to take the next step towards your goal.I assess based on the following:– Is it clear?
    – Is everything on the page doing a job?
    – Is there anything preventing or distracting them from taking the next step?
    – Is there anything giving them reason to doubt or trust you?
  4. Hypothesis time
    By now there’ll be some data to dig into. So it’s a case of combining the heuristic analysis with the data. And working out:– What to test
    – The expected result
    – What this will mean for your business
  5. Technical implementation
    I map out what’s needed. That may be building a new page, adding code, or generating a series of analytics reports. If one of the recommendations involves creating or designing a new page or section, you’ll need a developer or a designer to do that.
  6. Run the tests
    These run until they reach “statistical significance”. This is when you can be sure the results are because of what’s been changed, rather than because of chance.
  7. Gather results
    You get these in an Excel spreadsheet. These are split into columns. Depending on the project, these show the test, hypothesis, recommendation, urgency of implementation.

mountain with road leading up

“How long does it take?”

Ideally, a minimum of 28 days. This gives 4 full weeks to see how people behave (think  how office workers behave differently on a Monday morning compared to a Friday night). Naturally, it depends on your business and your schedule, so this is adapted to whatever works best.

“How do you decide what to test?”

I use a simple PIE formula:

  • Potential (how much of an improvement can be made)
  • Importance (how important is the traffic to the specific pages? Are they ready to buy?)
  • Ease (how easy will it be to carry out the test or complete the optimisation)

This varies from project to project, so to weigh each factor correctly it’s a case of assessing your market, competition, and factor in any internal pressures you’re under (such as needing to show positive results to your boss quickly).

“How much traffic do I need for A/B testing”?

As a rough guide:

  • Minimum 1,000 unique visitors to start the test (if you only get 500 per month that’s no problem, we can just run it over 2 months)
  • Minimum 250 conversions per variation before considering declaring a result
  • Needs to run over a full business cycle

What if I don’t have much/any data?”

I can set up your Google Analytics and start getting you data. While I’m doing that, I can get on with stage 3. If you’re just starting, the initial CRO review is based on conversion best practice and what I know works. Any tests can come later, once you’ve built up your customer base and have enough traffic to work with.

“What are your rates?”

Help yourself to either:

  • Fixed rate (where I do a full assessment and give you a full report with recommendations)
  • Ongoing retainer (for a series of tests, to optimise each of your channels and identify common patterns that appear over your business cycles)

clock

Often clients start with Option 1, as a trial run. And then move to Option 2. I’m easy either way.

As for pricing, each CRO project depends on:

  • The nature of your business
    – What type of website (obviously, the more complex the site, the more time it will need).
    – What type of CRO can be done (if you’re running SEO, PPC or digital marketing campaigns this can affect the results, and how the tests are run).
    – How much growth can be achieved through CRO (making sure the potential boost in performance is worth your investment)
  • Your website/landing page performance
    – How much traffic (the more traffic there is, the more time it takes to analyse, assess and measure)
    – How much work it needs, whether it needs a designer or coder to implement changes
  • How much uplift I expect to deliver
    I only take on projects where there’s real potential for growth. After all, CRO is all about results. So if I think you’re better off investing in another part of your business, I’ll say so.

“Which industries do you work in?”

Because the principles of CRO work across industries, I work on most types of businesses. Previous clients include private jet hire, e-commerce, technology, FMCG – so I’m open to most types of business. Although if you sell arms (as in the things that kill people, not the limbs), then I’m not interested.

What tools do you use and how much will they cost me?”

I only tools which:

  • Have minimal impact on your site performance
    (after all, slow speed is a conversion killer)
  • Are free or ridiculously good value
    To make sure you get the most ROI out of working with me
  • Are low maintenance
    No headaches, no problem. Once the tracking code is installed, they run happily in the background

So that means:

Hotjar

This little gem gives you so much insight for so little investment. You get videos showing your users interacting with your site. You can run survey and popups to get instant feedback. You can see heatmaps showing how far people scroll, and where they click. You can create enquiry forms and see how far people get, and where they give up.

Optimizely

Great for A/B testing. And for tracking clicks on any part of your website. Buttons, videos, menus… you name it, Optimizely can track it.

Human beings

Testers can give you real insight into the functions on your site. Although they’re doing this for spare cash, a fresh pair of eyes will always give you real insight. Around 10 testers is usually enough to identify anything we need to fix or test.

And finally…

If you made it this far, then firstly – thank you. I realise that’s a lot of words, for what is a complex service. And secondly, I’ve got one question for you… how can I help grow your business with CRO?

Tell me what you need