Do you have a software business? The first thing that people see about you is your website. Are you investing enough resources on making sure they have the best possible experience?
I’m guessing that more than 98% of people coming to your website leave without giving you their email address.
You know why? It’s because you didn’t convince them that your solution was right for them.
It means that you’re letting 98% of your traffic go to waste. Your business is doing okay, but what if you could convert 3% or 5% of your visitors to leads?
What if you could convert 10% of them? Your business would grow tremendously! You’d get more leads without spending more on marketing (thus reducing CAC).
To do that, you need to improve the conversion rate on your website. You need to understand your visitors and solve issues on your website, and this means doing some research.
CRO (Conversion Rate Optimization) does exactly that. However, it raises tons of questions… Are we ready? Do we have enough traffic? How do we do this the right way?
This article will give you answers to all these questions. You’ll finally be able to find out if you have what it takes to really improve your website’s conversion rate.
The Traffic Question
When launching experiments on your website, the test duration is influenced by two things: the number of people coming on your website, and the number of conversions.
If you get 500 visitors and 10-20 conversions per week, optimizing your website will take you weeks. However , it doesn’t mean that you can’t run a CRO program on your website.
You just have different tools at your disposal. If you want to learn more, read how you can do CRO on low-traffic website.
If you want to see how this plays out with your own numbers, you can try it out on the Sample Size Calculator.
Another question is: Are we solving for local maximum?
If you already have a conversion rate from visitors to lead of 1.5% or 2%. Raising that by 10% will have results (you’ll get 10% more leads).
However, are you optimizing the best lever there is? If you only have 500 visitors per week, you should probably work on acquiring more traffic before you can optimize your website.
Focus is extremely important within small companies. Discover your best lever so that you don’t spread yourself too thin.
If you have low-traffic or a bad conversion rate from free to paid, maybe your time would be better spent working on these areas than on your website.
I wrote an article on finding the best levers for your business so I encourage you to start there to make sure you’re focusing your resources where they matter.
The Right Mindset
If you focus only on conversion rates and you forget to do your homework, CRO can tremendously harm your business.
We’ve all heard about CRO but few people have the mindset required to really drive results without harming the businesses.
Being Customer Centric
As digital marketers, we tend to focus on numbers. When we do so, we often forget that there are people behind these numbers.
These people are your prospects. They’re the lifeblood of your business. You need to think about them and why they purchase your solution (or don’t).
This is true in business and also in CRO. If you want to increase the success of your tests, you need to understand your customers’ burning desires.
You can’t focus on your numbers. Instead, focus on presenting value to your visitors and you’ll be able to generate much better results.
Thinking about people will also help you to avoid testing things that are bad for your customer experience – like testing a modal popup that you can’t close (true story).
On the other hand, you need to approach CRO with a learner mindset. You have no idea what will work and what won’t.
You’ll need to prioritize experiments (e.g. ICE Scoring) but you’ll need to test plenty of things to see if they work
You need to approach CRO as if you were a newborn. Start by knowing nothing and as you fail (and win), you’ll learn more and more about your customers.
These validated learnings (through data) are what will enable you to increase the rate at which you launch successful tests.
Size of changes
When you’re on a website, do you respond better to a green button or to a red one? You probably don’t care. Guess what? Your visitors don’t care either.
Rather than testing things that have no impact, you need to test things that are big enough to have significant improvements in conversion rate.
Unless you’re Amazon and you have millions of visitors, doing small changes will have small impact. If you want to move the needle, you need to test big, bold changes.
Moz worked on a new homepage a while back which resulted in a $1M increase in revenue. Here is how the homepage looked before and after:
Please note that it’s okay to test something small as long as you have a hypothesis. However, you’ll need to accumulate all these “little wins” to really show results.
I see many teams starting something, investing for a few weeks and stop. That’s a terrible mistake.
If you didn’t think the investment was worth it, why would you even start in the first place?
The thing is – with CRO – you can’t generate results in one or two tests. You need to fail, and fail hard before you can move the needle.
Don’t let these early failures get in the way of your long-term plan. Do the work and it will pay off.
Consider the entire funnel
The best out there don’t care about a conversion rate. They care about growth. Let’s consider the following example:
You increased conversion rate on the website by 20%. The results of your experiments decreased the conversion rate from lead to paid by 20%.
You have done amazing work on the website but if leads don’t convert down the road, you haven’t grown the business.
Always consider your entire funnel. This will ensure that you are actually growing the business and that you’re focusing on the best high-impact area.
- Try to find leading indicators to other funnel stages so that you can avoid waiting months to see the impact of your experiments.
- One interesting way to do that is to factor in “Lead Scoring” within your experiments. You’ll be able to know if you’re optimizing for the best possible leads.
- You can use software like MadKudu or Clearbit to give you more data to score your leads and then feed that back into your CRO strategy.
Ability to take A or B Decision
This is one that I see very often. Someone will say “We’ve been working on a new website for the past 6 months, we plan on A/B testing it really soon”.
To which I always answer, “What will you do if the test fails?”.
It’s hard to A/B test extremely big projects because you’ve invested tons of resources on them and you’re probably happy with the results.
If the results are bad and the test fails, you’ll have no idea as to why the test failed – which ultimately results in lack of validated data.
If the test fails, you’ll end up keeping the new version. Which is probably okay. However, you’ll need to start the CRO process all over again.
On the other hand, if you’re making big changes to your business that you won’t be able to undo, there is no reason to A/B test these changes.
Pro tip: Instead of considering “big projects”, you should use Growth-Driven Design and CRO to make small and iterative improvements.
By doing so, you’ll avoid wasting tons of resources and you’ll gather feedback as early as possible, thus reducing the risk of the investment.
The Right Process
If you want to generate substantial results from CRO, you need a repeatable process. You can’t launch one experiment and expect huge uplifts.
You need to fail. You need to learn before you can generate any significant results, and that is why having a process is at the heart of CRO.
You need to include tons of research within your CRO process. You can’t rely solely on your hunches. You need to analyze and gather data to avoid shooting in the dark.
You need to go deep into understanding why people do what they do, so that you can build interesting hypothesis and test these.
Doing all that research and turning all of it in a cohesive CRO strategy is at the foundation of any successful testing.
The Right Resources
When doing CRO, you need 3 different skills:
- Research to conduct analysis and create hypothesis
- Designer to create the new variations
- Developer to implement these variations
These are the 3 skills required to launch tests and generate results. Whether you need 1 or 3 persons is irrelevant.
Don’t worry though, it’s okay to start small and take it from there (probably even better). It’s ideal if you can have someone driving the process and then rely on your in-house designers and developers to launch the first few tests.
As results begin to show, you can then decide on a better structure and potentially full-time people.
If you decide to rely on in-house resources, you should agree on the number of days allocated to this new project. Doing this will ensure that these activities aren’t getting sidetracked with other projects within the organization.
The Right Tools
Tools really aren’t that important. However, you need to make sure that you have data to inform your decisions and that you have the right measurement tools.
You need an analytics stack that can report numbers and give you some insights as to what you should be testing.
It’s also vital that you use an analytics software that can segment visitors in a way that makes sense (Segmentation is the secret to success).
Pro tip: Using something like Amplitude can tell you more about what happens down the road (after someone signed up). For example: You may realize that a specific Value Proposition is great for conversions, but bad for long-term usage.
You also need a measurement software that can do A/B testing and multivariate testing. This is what will tell you if you’ve improved your conversions or not.
You have tons of measurement software available to you. Compare them and make sure you find one that suits your needs.
If you’ve come this far, you’re likely ready to start doing CRO on your website. You’ll be able to improve your bottom line without increasing your costs.
Don’t overthink all of this. If you’re ready to invest in CRO, it’s better to start now than never. Start doing some research and launch your first tests.
Please make sure that you’re running worthwhile tests. Don’t try the color of a button. Formulate hypothesis and test them relentlessly.
Always remember that Mindset + Process + Strategy are the most important tools of your CRO toolkit t. That’s where you’ll have real impact.
Hope this helps you to generate tons of news leads for your business!
I’m Pierre Lechelle and I’ve been working in the tech ecosystem building businesses for about 10 years. I’m now focused on helping MadKudu to reach $100M in ARR. Connect with me on LinkedIn.