Conversion Rate Optimisation (CRO) isn’t just another TLA (Three Letter Acronym) that us digital marketers like to throw around to make ourselves look smart or cool. When done properly, it is the process of applying scientific methodology and statistical analysis to website experiments with a view to increase the likelihood of a user completing a specific action. In other words, “will changing something on my website result in more leads?”
But CRO is almost never done properly because the changes that are made are too small to matter. We’ll explain why you need to make big, bold changes to your site in order to run a successful CRO campaign.
Performing CRO experiments COULD be an excellent way to increase revenue. Using CRO to increase revenue means you only have to make a change to your website ONCE for a sustainable increase in leads and revenue.
For example: if 10,000 people visit your site every month, and 300 of them become leads, your website conversion rate is 3%. If you then make smart changes to your website which increases the number of leads received to 600, you have successfully increased your conversion rate to 6%
CRO is the gift that keeps on giving: you can reasonably and reliably expect to maintain the number of website leads received each month (assuming all other factors are maintained).
If you were to keep your website the same, but double the number of visitors EVERY MONTH, then you would receive double the leads. But how much would this cost?
Lots of money… This method could end up doubling the Cost Per Acquisition for new customers – can your business realistically sustain this?
Would you want to, even if you could?
Unlikely… so back to CRO.
Before embarking on this strategy, you need to have a few fundamental elements in place:
Baseline data that’s accurate
You need accurate tracking of your website, with enough historical data against which to make comparisons. The amount of historical data you need the amount of traffic to your site. Use this significance tester to see how long you need to run an experiment for.
Clear conversion goals in place
We’re assuming that you want to increase leads or website revenue, so it is essential that goals in your analytics system are properly set up. If there is even a skerrick of doubt surrounding your data, then DO NOT start any CRO yet – you need to get tracking working first.
Clear and simple hypothesis on a change that will directly affect conversions
Your goals must be SMART
Good example: “Halving the number of form fields will increase successful completions by 10%”
Bad example: “Changing the website font will increase conversions by 758%!!!”
The first example is specific and achievable (although slightly optimistic). The second example is completely nuts and should be dismissed. (As an aside, if you ever see “click-baity” studies that promise huge conversion rate improvements, you’re most likely seeing the result of a flawed test – these are a waste of time).
But before you choose your test, think about what type of approach is best for your situation.
Not all CRO tests are suitable for all situations, so you need to choose your approach carefully:
Or as I like to call it: Big Effing Changes!
This is where you test two wildly different ideas to see which performs better. This is one of those times where BIGGER IS BETTER.
You need to be brave – your test should put you and your team outside of your comfort zones.
This approach is the best way to go if you have not done any CRO before.
This is the process by which you make one change at a time that slowly evolves the website over time. This is typically low risk, but you need at least 1,000 conversions per month for any test to yield statistically significant results.
These are small and simple changes that you often read about online, such as changing button colours or background images. This approach is suitable for heavily trafficked sites that have had other major issues resolved.
If you’re running a meek test on a site that receives less than 1000 conversions per month, then this is the equivalent of shuffling deck chairs on the Titanic. You should stop this RIGHT NOW! Anything else is a waste of time and resources.
Be bold! Be brave! Bigger is better! Testing two wildly different ideas is the only way to get meaningful results from a Conversion Rate Optimisation experiment.
Good luck, and happy testing.
If you need a hand with this, don’t hesitate to contact one of our CRO specialists.
Get free digital resources and articles direct to your inbox to boost your business.