What Is CRO? A Beginners Guide To Conversion Rate Optimization
Today I’m going to be telling you about a way to increase your sales from the traffic your website is already receiving.
I’m going to be explaining a technique called CRO and how it’s transforming the way companies improve their sales online.
Although, there’s a few things you should know…
It’s not a hack.
It’s not a tactic.
And it certainly isn’t a quick fix.
What do we know? We use CRO everyday as part of our lead generation services and we wouldn’t run a campaign without it.
Here’s What We’re Going To Be Covering :
- What Is CRO?
- Why Use CRO?
- Does CRO Work?
- CRO Methodology
- Conversion Research Stages
- Insight Prioritisation
- A/B Testing
- Wash Up
- Rinse and Repeat
Let’s jump in.
What Can CRO Do?
If you don’t know already, CRO stands for Conversion Rate Optimization (or optimisation depending on which side of the Atlantic you are from ? ).
CRO is the method of improving the user experience on a website. The end goal being to increase the amount the amount of conversions your website receives.
A conversion can be classed as any of the following :
- Product Purchase
- Ebook Download
- Newsletter Signup
- Event Registration
A conversion is essentially an event where a user performs an action that is directly related to the objectives for your website.
For example – if you own an ecommerce store, then your primary goal is to generate as many sales at the highest order values as possible.
Alternatively, if you are a lead generation company, then your primary goal is going to be to collect as many leads as possible from as many users as you can.
Optimizing Your Conversion Rate
Your conversion rate (The CR of CRO) is the amount of conversions you receive depending on the amount of visitors that come to your website.
Conversion rate optimizers attempt to increase your conversion rate by optimizing your website.
Side Note : Your conversion rate is calculated by taking the amount of conversions that your website has received over a particular time period, then dividing that by the amount of users that have been to your website.
This is then multiplied by 100 and displayed as a percentage :
How does your conversion rate compares to your competitors?
Average Conversion Rate By Industry : Source
CRO only exists because the sales that websites generate, play such a huge part of most modern businesses.
CRO is a way to gain an advantage online that doesn’t require spending extra money on marketing.
So if you have been wondering – what is conversion rate optimization?, hopefully I’ve answered your question.
Now you know what the CRO acronym stands for and hopefully I have your attention, let’s go into more detail.
Why Use CRO?
If a company is trying to increase their sales online, the number one answer to do something about it is usually to increase the amount of traffic that a website is currently getting.
If more customers come to a website, the website will receive more sales.
Whilst in theory this could result in an increase in conversions, there are two major flaws with this plan :
- Not all traffic will convert at the same rate.
- Acquiring more traffic usually costs more money.
That’s where Conversion Rate Optimization comes in. CRO involves converting more of your existing traffic into customers, rather than having to acquire additional traffic.
Does CRO Work?
Relatively speaking, CRO is still in its infancy. According to Econsultancy, for every $92 spent on acquiring customers, only $1 is spent on converting them.
This is most likely down to companies not knowing or fully understanding about CRO and the benefits that it can have for a business.
But it’s popularity is growing :
More and more companies are adopting an experimentation culture in their organisation and many are reaping the rewards.
Take Booking.com. Well known for running thousands of A/B tests concurrently, their results speak for themselves – They have “conversion levels 2-3x the industry average.” – Source.
Or what about SmileyCookie. They increased their sales by 41% by implementing a conversion rate optimization programme on their website.
A usual CRO programme here at HelloClicks follows a precise methodology that we have implemented in order to see maximum results.
Let’s take a look at our process :
Here are each of the steps in the correct order and a summary of what’s involved :
1.) Conversion Research
This involves using a wide range of techniques (more on this below) to uncover a combination of issues and opportunities on a website.
2.) Insight Prioritisation
Once we have completed our conversion research, we’ll rank each finding in order of their expected impact on the conversion goal we’re trying to improve.
3.) Create Treatment
We’ll then start working our way through each item on the list one by one by creating a variation of the current version to address the issue. This is known as a treatment.
4.) Run A/B Testing
We’ll then test the current version against the treatment to see if we can increase the performance of our objective.
5.) Analyse Results
Once the test is complete, we’ll analyse the results and decide to implement the treatment on the live website or not.
6.) Rinse and Repeat
As soon as the test is complete and we’ve analysed our results, we’ll move straight onto the next test.
We’ll do this until we work our way through our list of insights.
Once that’s complete, we’ll start the whole process again.
Conversion Research and Testing
These are two main parts of a successful CRO programme – research and testing.
The research stage is aiming to find two things.
First of all, we are trying to find user experience issues that people are facing. These issues will be holding back a large percentage of users from completing the main conversion goal.
Secondly, we are trying to find opportunities that we can use to optimize our website as well as other potential areas of the business ( i.e. customer service / sales / purchasing).
The testing aspect of a CRO programme is the point at which alternative versions of the existing web pages are tested against the originals to see if it’s possible to increase the performance of the metrics which are being tracked.
Here are some examples of metrics that we may focus on improving during testing :
- Conversion Rate
- Average Order Value
- Abandoned Checkouts
9 times out of 10, the goal will be to increase the conversion rate. But in certain situations during optimization of an ecommerce website, we may look to focus on increasing different metrics.
Conversion Research Stages
Conversion research takes on two primary stages.
The first is Quantitative Research. This is involves data that is based around numbers, in other words data that can be quantified.
The second stage is Qualitative Research. This type of research is the opposite of quantitative research as it’s non-numerical based. It’s a way of us understanding user behaviour and the reasoning behind the decisions they make on our websites.
Let’s go into a bit more detail.
Quantitative Research Techniques
As I just mentioned, Quantitative Research is data that we can interpret based on numbers. As you may of guessed, analytics data will play a bit part here.
We will be using platforms like Google Analytics, Mixpanel and Kissmetrics to try to gain insights into what users are doing whilst they are using our website.
Here are some insights that we will be able to collect :
- The users journey through a website
- Where users are abandoning a website
- Which traffic sources are generating the most conversions
- How are different devices converting
With this data we can start to build a picture of the user’s journey, pre-click to purchase. Additionally, we’ll be analysing users that abandon the website before buying.
We can also begin to find out if there are any technical issues on our website that are causing users to leave.
For example we may see that mobile devices have a really high bounce rate and low conversion rate, so we may do some further analysis to find out more.
Look at the huge difference here between desktop and mobile – needs some investigation
Qualitative Research Techniques
This part of our research is much more user-centric. Rather than numbers, it actually involves people.
We will be using a variety of techniques that will allow us to observe user behaviour and how these people people ‘use’ our websites.
Some examples of the various ways of performing our research are:
- Performing User Testing
- Customer Interviews or Surveys
- Watching User Session Recordings
- Analysing Heatmaps
This data will usually be our most valuable. The reason being is that Quantitative data only tells us what users do, not why they are doing it. Once we know why users are doing something, we can try to form a resolution to try and fix it.
Combining The Two
More often or not, we’ll have theories behind why we ‘think’ something is happening or why users are doing things a certain way. What we will do is combine the two types of research to back any theories up with actual data.
For instance, we may find that during our analytics analysis, a higher than average number of users are abandoning the shopping cart on mobile devices.
We can then take this data, do some web form optimisation by performing a site walkthrough in all major mobile devices and browsers to see if we can find an UX issues or bugs.
Or vice versa. We may find that during user testing, a high percentage of users are finding it difficult to progress to the next step on during our checkout process.
We can then go into our analytics tool to see if we can find any data to back our findings up.
André Morys of konversionsKRAFT told us this :
Expert Insights - André Morys
We strongly believe in the combination of qualitative + quantitative methods.
Throughout or conversion research, we will be taking notes of our findings that we feel have a strong likelihood of causing users to abandon our website without converting.
We aren’t going to test fixing all of the issues at once. Why? Well if we did that, we wouldn’t know which ‘fix’ played the role in any increases or decreases in our conversion rate. Instead we will be testing one item at a time, to see if we can improve our websites performance.
But how do we know where to start?
First of all, we will rank all of the issues based on their potential impact of increasing the metrics that we’re trying to optimise.
Once we have our list of prioritised findings, it’s time to start working towards trying to fix them. This is done through something called A/B testing.
A/B testing involves taking one or more variations of a webpage (hence A vs B) and testing them against each other to see which one performs the best.
The same volume of traffic will either see version A or version B of the page.
Before we start testing we will need to calculate the amount of traffic and conversions our tests will need as well as the duration that they run for. This is to ensure that our tests our accurate.
There’s a lot of people and companies out there that “think” they are doing A/B testing and are seeing increases in their conversion rates. This is due to a lack of knowledge in statistics.
Peep Laja of ConversionXL told us this :
Expert Insights - Peep Laja
Learn basic testing statistics. You will ruin your CRO efforts if you don't understand stats. You will end up running so many inconclusive experiments, and end up getting so many false positives.
Peep recommends this this article to get started with statistics.
So the inaccurate tests companies are running, could actually be doing more harm than good. Ensuring that test results are valid is number one priority.
The most common scenario is that ‘testers’ are deciding a version is a winner based on too few conversions and too little traffic.
There are other type of testing available to use. If you would like to find out more about them, take a look at this post.
We’ve got to the point where we’ve run our test and it’s time to review the results.
We take a look and it looks like it saw an increase – “Great, onto the next experiment” I hear you say….
Hold up one minute.
Don’t take those numbers at face value, it’s time to look at them in more detail.
In order to know if we truly have a winner or loser, we need to segment the results.
We need to compare various segments to make sure that we have a true winner.
Some example segments to compare :
- Mobile Vs. Desktop
- iPhone Vs Android
- Chrome Vs Safari
- Paid Vs Organic
By segmenting the data, we can find out so much more about our experiments.
Sometimes if we have a losing test, we can find areas where there was actually an increase in performance. This can then give a foundation for a future experiment.
Rinse and Repeat
OK so washing up and rinsing may sound more like housework than CRO, but stay with me.
Once your test has finished and you’ve analysed your results, it’s time to move onto the next test as soon as possible.
The aim is to run as many tests as possible (also known as testing velocity). Doing so will allow you to see lifts quicker and learn much faster than doing stop-start testing.
By now I hope i’ve been able to answer what is CRO.
CRO is an extremely interesting and exciting area. It’s a way to help you make more money for your business, whilst at the same time learning about who your customers are.
In this post, we’ve covered a summary of the core fundamentals of CRO and I really hope you’ve been able to gain a firm understanding into what it is all about.
If you have any questions or if you have any comments, please feel free to type away in the comments!