Drive growth with conversion rate optimisation.
Your conversion rate measures how many users are acting on your website. The ‘conversion’ is anything you want a user to do when they visit a page. Maybe it’s making a digital purchase, signing up for an email newsletter or clicking a download button.
While there are many ways to calculate a conversion rate, most commonly, we are looking at the ratio of users who act after actually arriving on the site. Select a timeframe, divide the conversions by the total number of visitors, and multiply by 100.
Conv. Rate = (conversions/total visitors) * 100
If your page had 6000 visitors over a week and 300 took a conversion action, then your conversion rate would be 5%. Assuming you have some kind of site analytics setup, we would be looking at conversions/sessions or if you are looking directly at an advertising platform this might be conversions/clicks.
If your page has multiple conversion points, e.g., a form, web chat and a phone number, these would be added together then divided by total visits. It’s important to only select relevant conversion points here. While adding in elements like button clicks, time on page goals, scroll depth, etc. might make your conversion rate look better, you’re really just diluting your actual goal and moving further from business objectives.
A high conversion rate can indicate several things, like:
- Your website is attracting highly relevant traffic
- A competitive offering
- Well-designed landing pages
- Clear conversion points
Let’s assume you’ve got the first two covered and focus on points three and four.
While conversion rates can indicate effectiveness, even with analysis you might not be able to understand exactly what is driving or blocking. And that is where Conversion Rate Optimisation (CRO) comes in.
It might just be the link that’s missing between a site visit and the flip to a conversion.
What is conversion rate optimisation in SEO & digital marketing?
Conversion rate optimisation is the process of analysing your website and identifying opportunities to improve it so that more visitors take the desired action. You can think of CRO as targeted user experience design (UX) that doubles down on driving interaction.
If you think about the journey a user takes to eventually convert, there are a lot of opportunities for drop-off. Overall, marketers seem comfortable with the concept of driving relevant traffic to a website, whether that’s organically or via paid channels and they often look at conversion rate as a reflection of relevancy. While that’s certainly a significant factor, getting a user to take a conversion action takes so much more than relevance alone.
Why is conversion rate optimisation important?
No matter how well-targeted your digital marketing is, all the time and money you’ve spent will be for nothing if you aren’t generating conversions. Conversion rate optimisation is crucial for ensuring your other digital marketing investments pay dividends.
If you’re running a mature digital marketing campaign, doubling your conversion rate or even increasing it by a small percentage is not a simple task. There is only so much traffic coming from your target audience, and you’ll encounter diminishing returns if you cast the digital marketing net wider.
CRO has the advantage of multiplying your existing efforts and driving even more value.
Say you’re currently spending $50k a month and driving 5,000 conversions at a $10 CPA (cost per acquisition or conversion). By increasing your conversion rate by 50%, you’re now driving 7,500 conversions at a $6.66 CPA. Improving your conversion rates effectively gives you a discount on all your other marketing efforts.
More volume AND a lower cost per conversion? What’s not to love?
How does conversion rate optimisation Work?
So, now you understand the goal and the benefit of CRO, but how does it work?
CRO is all about analysing your path to conversion, identifying roadblocks and smoothing them out. These roadblocks are sometimes obvious, a form that’s hard to use key text that doesn’t stand out, but other times the issues users are having on a page aren’t that obvious.
A way to better understand user interaction is by using heat mapping to highlight what is impacting decisions as well as low areas of activity that are important but that users aren’t reaching or engaging with. Understanding how a user is engaging with your site is invaluable and heat mapping can identify ways to design a better conversion experience and drive change based on first-party user data, not just best practice approaches.
After understanding the roadblocks, you generate a variant that fixes the issue. This might be an entirely new page, a small adjustment to a layout or even something as simple as icons and headings.
This variant is then tested against the existing, or ‘control’ element or page, and we measure which one is better at driving the target conversion actions. This is typically facilitated through a layer of software such as Google Optimize, Optimizely, A/B Tasty or VWO, which segment your users and serve the variant and control in equal numbers.
After receiving your results, you implement or keep the better-performing variant and then start all over again. This is achieved by either iterating on the same page or looking at other problem areas on your site and developing new solutions to roadblocks.
What is a good conversion rate?
A good conversion rate is one that is higher than before your CRO project was completed!
While there is plenty of info out there on ‘good’ conversion rates and industry benchmarks, your conversion rate is, for all intents and purposes, unique to you. Comparing yourself to other sites will only make you rest on your laurels or chase a number for the sake of it.
If your product is 2x more expensive than the industry average with no appreciable difference, would you expect your conversion rates to be the same? What if your brand is an industry leader? What if your site takes 30 seconds to load? What if your header has a picture of a cute puppy? Comparison is a slippery slope.
That’s not to say that what your competitors, industry or indeed websites in general are doing is not important. After all, there’s a reason why websites don’t have trailing cursors or use Flash anymore – it’s important to stay up to date. But really, CRO is an introspective process, and you should be focused on increasing your conversion rate, not comparing it to others.
Is CRO Relevant to My Business?
Yes.
Unless you have a conversion rate of 100%, you should be doing CRO.
While mature business and marketing operations stand to make the biggest potential gains, new businesses or sites that incorporate CRO into their thinking will benefit. Without historical or real-time data to analyse, you can test elements with variations of designs to optimise a site build. Bear in mind there’s nothing to stop you from launching a website with split testing enabled, too.
Don’t rely on tips, guesses or unverified tweaks to figure out what works for your site. The conversion opportunities you miss out on are too valuable to leave your CRO to chance.
How do you optimise your conversion rate?
There are a few different ways to approach a conversion rate test, but most of them boil down to comparing a variant (or set of variants) to the original or control. This would typically involve these steps:
- Identify a conversion action of interest
- Identify a page/element to test
- Develop a variant or series of variants of the test page/element
- Set a sample size
- Run a test splitting traffic between the control & variant(s)
- Analyse the results
- Implement the best variant/keep the control based on performance
There are several ways to implement this, but let’s look at simplified A/B tests.
Example: We identify that users arriving on page A often leave the site before performing their desired conversion, filling out a contact form. We want to reduce the number of people leaving and increase our conversion rate on page A.
Having a look at Google Analytics, we see that page A receives 1,000 visits a month, and 10% of users take action. With our heat mapping we see that only 30% of users make it below the fold (first visible section of the page), and so don’t even see our contact form.
Charged with this information, we investigate direct competitors and note that some of them have moved their contact forms to the header. We develop a hypothesis that by moving the form into the header, we will get greater visibility and, therefore, increase the number of users who complete our form.
We redesigned the page, moving the form into the header, and also updated some of the text, images and icons above the fold to give as much pertinent information to the user in the section of the page they actually view.
We decided to test 2,000 users over two months, with 50% of the traffic going to the control (A), and 50% to the variant (B).
After the test has concluded, we note that page B had a conversion rate of 16% while page A had a conversion rate of 11%. This improvement is statistically significant, and we can be confident it was due to our change, not random chance. Note that we are comparing our uplift against the conversion rate of page A in the same period at 11%, not against our historical measurement of A at 10%.
We implement page B on the website.
How long should conversion rate optimisation experiments run?
The length of your test is determined by two main factors: the number of samples and the variance between the variant and the control.
Imagine a test that has sent two users: 1 to the control and 1 to the variant, and the variant has converted their user, while the control has not. If you look at the results, the variant has a conversion rate of 100%, and the control has a conversion rate of 0%. But with a sample size of 2, it’d be hard to say whether this was representative of the population or just random chance.
What if you let your test run a little longer? 99 more users come to the control and don’t convert, while 99 users come to the variant and convert. You’re still looking at the same conversion rates for each page, but now, with a sample size of 200, it’s more likely that your variant is the better performer.
If you’re looking at marginal performance increases, say comparing 14% conversion rate to 15%, it’s important to keep collecting data to justify the result before implementing changes. If no clear stand-out performer emerges, a test with a new variant is your best next step.
Put your site to the test
By using CRO to test and refine how visitors interact with your site, you can make data-backed decisions that boost your ROI and underpin growth more quickly and cost-effectively than any other tactic you can employ.
| Definition | Description |
|---|---|
| Sample & Population | The sample is the portion of the population that we are testing in our CRO experiment. While it would be more accurate to test every user when you make a change to your site, its impractical and cost prohibitive. |
| Confidence | This is represents how confident you are that the results you record are not due to random chance. It’s crucial to understand that this is not the chance that your result is reflective of the population, but rather the chance that your result was not a fluke. |
| Hypothesis | A proposed explanation made on the basis of limited evidence that’s used as a starting point for further analysis. |
| Heatmapping | Recording user behaviour to identify ‘hot-spots’ and ‘cold-spots’ of activity on a given page, this could cover how far down a page a user goes, what elements they click on, and how much time the spend on certain sections of a page. |
| Statistical significance | A threshold that, when crossed, determines that an outcome is the result of an attributable cause rather than chance. |
| Control | The original element or page that we are testing against, it’s crucial that we use this to ensure we have a reliable comparison for our variant performance. |
| Variant | An adjusted element/page that is designed to perform better than the control. |
| A/B/n test | An A/B or A/B/n test compares performance of the control (A) to a variant (B) or number of variants (n). This covers 95% of tests and is the easiest way to think of CRO testing. |
| Multivariate test | Similar to A/B/n tests, a multivariate takes combinations of variants and tests these against each other, and the control to determine the best combination. o E.g., if you had 3 header images and 3 button colours you wanted to test, you would test all possible combinations to find the best 2 elements. |
Written by
Peter Dimakidis
