In this series, we will cover different ways to maximize your conversion. In part 1 of this guide we will talk about the basics of conversion optimization. The next parts will go more into detail and we will cover topics like design, structure and features, which can boost your conversion. So let’s go into the basics!
Know your target group
This is probably the most basic part about conversion optimization, but also one of the most underrated ones. If you take a closer look at all the guides and studies about conversion optimization, you probably get the feeling, that it is all about coloring a CTA button. Presumably, the next thing you will be doing is coloring your buttons red – and then you wait. Some days later you can’t believe that nothing happens as this one case study told you to receive a 1200% conversion boost just by coloring your buttons red. Yeah, we have been through this, too!
Although every hint and study about conversion optimization is useful and definitely worth reading, you should keep in mind, that it might not be applicable to your own business case. One of the reasons for this is, that you are addressing a different target group. Let’s asume you are selling expensive luxury watches and you want to tune your conversion. Then there is a study, which implies that changing your product titles to something more action-oriented might increase your conversion. Basically this is a good thought, but it might work best for low priced goods, ebooks and similar stuff. Luxury goods on the other side are sold by emotions. So, using an emotional approach for your titles might work best for your case. This example shows you, that you have to read studies with caution and think about ways to transfer the results to your own case.
How to identify your target group
You would be surprised, how less people know about their customers. One of the main mistakes is not being enough specific about your target group. To remain with the example above: watches are for everyone, but those watches you are selling most likely aren’t. So the first step of getting to know your clients is to look at the characteristics of the product you are selling. Those characteristics will give you important hints about the target group and where to find them.
- Product Facts
- Type: digital vs. physical
- Price: expensive vs. cheap
- Design: creative vs. classical
- Product Usage
- Advantage: What is your USP?
- Restraints: Who can use your product, who cannot? (e.g. gender, complementary vs. used alone)
The knowledge about your product should then be transfered to your target group. Sticking to the following scheme, when defining your target group, will help you to be most specific. Try to think of your average customer and assign the following attributes:
- Media competence
- Place of residence
- Country, State, City
- Active in a circuit of …
- Price sensitivity
- Media usage
- Payment performance
- Brand confidence
There are much more attributes you could think of, but for me those have been the most important regarding conversion optimization.
Does your traffic match with your target group?
Okay, now you have a perfect definition of your target group. Next question to ask is: where does your traffic come from? It often happens that your traffic and your target group differ. This is another reason, why implementing a new feature might not bring you the conversion boost you expected. If you don’t have targeted traffic, your conversion will not increase no matter what you do. You basically have two options here:
- Change your traffic strategy
- Change your „conversion strategy“ / Rethink your target group
The first option should be rather clear and it is the one, which will work in most cases. You could change your traffic strategy by placing more targeted advertising. Or write blog articles, which might be more interesting for your target group.
The second option is the much more interesting one. You probably wonder, why it might be a good advice to adjust your optimizing efforts to your traffic.
Let’s look at another example:
You have a startup, which is about to produce and sell a new luxury watch (yes, I like luxury watches). In the first step you have defined your target group as rather old, financially situated and interested in sports cars and golf. Therefore you optimized your landing page to target this audience best. Now you have done a lot of social media marketing, you have launched a blog about startup life and you have done a lot of business networking in the startup scene. After a while you managed to reach a decent amount of regular traffic, but there are no pre-orders. Then you should definitely have a closer look at your traffic. Your traffic sources will probably indicate, that your traffic mainly consists of young people and early adopters, who are most active in social media. Instead of changing your traffic sources it can be much more effective to change the way you target your audience and kind of rebrand your product.
Option 2 is only an option, when you have targeted traffic, but it differs from your target group. This way chances are good to convert the traffic you already have.
Watch your competitors
This one is another basic thing, but I will at least mention it as it belongs to the topic. You should steadily observe your competitors and scan their pages for changes. This helps you to see what works for them and thus to identify mistakes you make. It also helps you to identify your target group. When it is about identifying your competitors google can be your best friend.
Try, Test, Optimize
Conversion optimization can be a tough road. Usually things aren’t working out the way you expected. Unless you are very lucky, optimization is not a one-shot business. It is more like:
Optimize -> Wait -> Evaluate -> Repeat
But how do you test the right way? And how do you draw the right conclusions? Basically there are two ways of doing tests – A/B-Tests and multivariate tests.
A/B-Testing is the most basic kind of testing. You deliver two different variants (A and B) of the same page and split your traffic to see rather variant A or variant B. You will then get results on which of the two variants perfoms best. An A/B-Test is pretty simple to implement and it doesn’t require a large amount of traffic to work. Both variants of your page should only differ in one specific element – e.g. the color of your CTA button. That way you can be sure that your results are based on this one change. So before you start your test, you need to have a hypothesis, which you can accept or reject after your testing period. For best results with an A/B-Test your hypothesis could look something like this:
Action-oriented titles lead to more sales
Now you can create two variants of your page with two different product titles and see, if the action-orientated title outperforms the other one. If it does, the hypothesis will be accepted. If it doesn’t you, you will reject it.
The problem with A/B-Testing is, that you often won’t see significant changes. More often it is a combination of different changes, which will lead to better results. This is where multivariate testing can help.
When doing multivariate tests you create as many versions as needed for your website by combining different changes. Again there should be a hypothesis defined before starting the test. As you probably have a large number of combinations of different elements you should be more generic, when defining a hypothesis for a multivariate test. You will be most successful when you choose one topic to test. Your hypothesis could be something like:
Action-orientation leads to more sales
As you can see the topic is action-orientation, but the hypothesis still is rather generic. Instead of only testing a specific element like the title, you are testing a whole topic. You could now create different variants by using action-orientated titles, buttons and descriptions and position them in different ways. The more variants you choose to test, the more traffic you will need to validate your hypothesis.
I hope, you liked part 1 of our guide. The next part will be about design and how you can use it to change the user’s behavior.