Conversion Rate Optimisation (CRO) can deliver remarkable improvements to the performance of each marketing channel.
Ultimately, it delivers revenue growth and increases ROI.
However, despite the overwhelming benefits, especially in the online betting space, the majority of websites do not apply CRO processes to improve conversion.
Conversion Rate Optimisation – Where to Start
So, where to begin?
Ok, so I am going to start with some sports betting and casino issues you should be aware of.
- Your bonus can make a bad funnel look good. A generous price boost in your sportsbook or a no deposit bonus on your Casino will make even the worst landing pages and registration forms convert well.
- Page speed is a factor. Outside of the page design, page speed has a massive impact on conversion rate. Always check the load times of variations in your tests to make sure there are not huge differences. This can invalidate your test.
- Segment your results by device type.
- Proximity to an event can affect conversion rates. For example, ten minutes before a race you will see a smaller percentage of people drop from the conversion funnel. Therefore, conversion rates tend to be naturally higher just before an event starts (PPC Managers – bid modify)
CRO Pre-Test Research
The first step is to look at your user journey and identify problem points. Then depending on the areas identified you may want to roll out a test on live traffic or do some preliminary user testing first (on staging environment).
If it is something like moving the “Join Now” button above the fold the releasing the test straight to the live environment should be ok.
However, in the case of changing the registration form, you should do some user testing on the changes before deploying to your live environment to test.
A/B Tests
On the face of it, A/B tests are simple. Just test two variations against each other and keep the winner. Rinse and repeat by replacing the losing page and letting evolution create the ultimate conversion machine.
However, it’s not that simple. Each test should have a hypothesis.
You need to do preliminary research in order to identify where users are dropping off and what might be causing issues with conversion. This will give you insights which you can use to create your hypothesis.
For instance, after reviewing Google Analytics Click Events, looking at heat maps and viewing recordings of user actions you see that many users do not click on the “Call to Action” button because it is located below the fold on mobile devices.
So, your hypothesis is – The “Call to Action” button located above the fold on mobile devices will get a higher click-through rate than the control (button located below the fold on mobile devices).
A/B Test – How do you know what is the best version?
The accepted standard is that if a version has a conversion rate and the test shows 95% significance (1/19 favourite) then you should select it as the winner.
Now, we have all seen 19/1 horses win a race so just because your A/B test shows 95% significance doesn’t mean that it is the best version for sure. It means that there is a good chance that it is the best version, but there is a 5% chance that the difference between pages was down to natural variance.
Try running a series of A:A tests and see how that will mess your head up. This is when you run identical pages against each other. (Sometimes an identical page will convert better than it’s twin!)
On a lot of occasions, you don’t reach 95% confidence. There is just not that big of a difference between the two pages. Some practitioners recommend dropping the significance threshold or designing your tests so that there are likely to be large differences between variations.
Because of this, you should try getting the biggest possible sample size for your A/B tests. Now, this can be difficult if you are testing single marketing channels or expensive if you are testing online casino PPC traffic.
Multivariate Tests
Multivariate tests require a lot more traffic. As a result, it can take a long time to get significant results. Because of this, for most tier 2/3 bookmakers A/B testing is the best approach.
Look at Multivariate tests as A/B/n tests. You are testing multiple “static” versions against each other.
It is sort of like a league format with a number of teams and multiple games to determine the champion, compared to the one on one boxing style “if you beat the champ you are the champ” of A/B testing.
Affiliates Will Appreciate You More
By implementing a conversion rate optimisation programme your affiliates will also benefit. They will see better value in sending traffic to your site than to your competitors.
When affiliates are looking at driving revenues from their website real estate, you will be eligible for improvement in position. Or, in the case of a CPA affiliate, can compete for better CPA deals.
Cost of Implementation Versus Benefits
The cost of implementing findings from tests must be considered.
- What are the potential benefits of implementing the changes?
- What resources are required to implement the changes?
- Am I likely to see a positive ROI?
- What am I “saving” in ad spend?
Online Betting – Using Conversion Rate Optimisation to Get the Most from Your Advertising Budget
Paid Search is hugely expensive with some of the highest cost-per-clicks in any industry.
I believe that Conversion Rate Optimisation should be compulsory for any paid search programme in betting. The costs of implementation are often dwarfed by the improvements in CPA that can be achieved even with modest changes to landing pages and registration forms.
Below is an example of how after a cycle of conversion tests throughout the conversion funnel resulted in an improvement of 7% on landing page CTR, 20% on form completion rate and a 16% deposit rate improvement can reduce CPA per depositing customer by £175.
So, in this example, the operator saves £175 in ad spend every time they acquire a new customer (based on their old pre-test CPA).
| Budget – £ | 100,000 | ||
| Average CPC – £ | 65 | ||
| Old Version | New Version | Improvement | |
| Number of Clicks | 1,538 | 1,538 | |
| Landing Page Conv. Rate | 70% | 75% | 7% |
| Web Form Visits | 1,077 | 1,154 | |
| Form Completion Rate | 35% | 42% | 20% |
| Registered Customers | 377 | 485 | |
| Deposit Rate | 50% | 58% | 16% |
| Depositing Customers | 188 | 281 | 49% |
| CPA – £ (Depositing Customer) | 531 | 356 | -175 |
| Saving on Customer Acq. (1 month) | 49,143 |
Above The Line/Sponsorship Campaigns and Conversion Rate Optimisation
Sponsorship, T.V., Print and In-Shop campaigns are expensive. Spending all that money to get people to visit a website with gaps in the conversion funnel is crazy.
Optimise conversions from “Direct” traffic to make the most from your ATL and sponsorship spend.
Most of the time this traffic class will land on your homepage. You will find that you can achieve substantial increases in registrations and revenue by optimising the home page.
However, patience is required and you must resist the temptation of testing everything all at once. Remember that you are optimising a thought sequence and not a website design.
Conversion Rate Optimisation – Well Known Brands Convert Better
An important thing to note is that a strong brand can improve a poor conversion funnel. Branding increases conversion and in many cases papers over the cracks.
Be aware of this whenever analysing funnels of Tier 1 operators.
Tier 2s and 3s along with start-ups have to work much harder.
Conclusion
In summary, Conversion Rate Optimisation (CRO) is the best way to increase revenue and increase new customer numbers.
On top of this, it is also a great way to help improve customer retention metrics.
Ultimately, Conversion Rate Optimisation should be a priority for your marketing team.
