over $150,000 in additional monthly revenue by increasing conversion rate by 6.9% and revenue per user by 37.8% per test

Client

Ozarke

Founder

Alex

Industry

Home Decor

Short Term Results

6.9% increase CNV & 37.8% Revenue Per User

Long Term Results

$150,000 In additional monthly revenue

The Problem

Ozarke came to us in August 2023 having launched the brand in late 2019 and successfully scaled the brand to mid 7-figures, and were now looking to significantly scale the business over the coming 12 months. 

While there were no major concerns about their website, they appreciated that a CRO program could have a significant impact on their bottom line by increasing both conversion rates and average order values.

However...

What had they done to try to resolve the issue?

How Customers Who Click Helped

Every project starts with research.

We spoke to Ozarke’s customers, mined their reviews, and reviewed their customer service tickets to find out exactly what their customers cared about. What were the positive and negative things they were saying. Why were some customers buying and others weren’t.

The end goal of all this research was to find out exactly what Ozarke’s customers cared about when buying lighting, wall panels, and other items of furniture.

It quickly became apparent that while price and quality was important (as always), customers very much appreciate Ozarke being a US business with its own product line.

Step 1 - Research

What research methods did we employ?

In this case it was the Product Detail Page (PDP).

While these pages got plenty of traffic, the add-to-cart rate was very low. Those who did add to cart though tended to convert quite well, and at decent AOVs, so there was no initial concern over the checkout.

What were Ozarke’s visitors seeing on the page, where were they interacting, and what were they missing?

In this case, the image gallery got a huge amount of attention. This makes sense due to the nature of the product, but if customers weren’t converting was it because they didn’t like what they saw? Or because they were still missing something?

In addition to this, the search function on the website received a lot of engagement, indicating that customers tended to have a good idea of what they were looking for.

Here we were looking for 2 key pieces of information.

These are the two key areas we have to identify in order to improve conversion.

Why did some people buy?

Why did others not?

When we’re analysing the data and the feedback we aim to pick out several key focus areas for testing, based on 3 key customer behaviours.

Usability, Anxiety, and Motivation.

In the case of Ozarke we found 3 key areas that would be the focus of our testing.

Step 2 - Testing

Once we’ve identified our Focus Areas, or bottlenecks, we start to create tests to explore our hypotheses.

Some of these tests will involve adding information or elements to a page, others involve removing, and some others still require changing or restructuring how information is provided to visitors.

For each focus area we come up with 5-10 experiments, each of which has 3 ideas attached to it in order to allow us to speed up the iteration process immediately. 

It can be quite easy to think of an idea for a test, but when forced to come up with 3 options for fixing the problem, we really explore the issue more deeply and come up with better initial solutions. 

In addition when it comes to the iteration phase we already have a much deeper understanding of the problem and our route to solution.

We then prioritize which tests to run based on an 8-point scoring model to determine the impact of the test vs the effort required to build it.

We then follow a set process when rolling out tests:

The main metric we're basing results on? Revenue Per User.

While our aim might be to improve conversion rate, if a test has a negative impact on AOV which negates any conversion increase, we may not want to implement that test live on site.

But we don’t just finish there.

Before we actually resolve a test and move on we assess 3 more things:

Learnings - how did the test perform against our hypothesis?

Recommendations - to implement or not, and whether this should only impact certain segments, channels, or devices

Iterations - regardless of whether a test is successful or not, we don’t just leave things there. There was a reason we tested something, so there’s valid reason to keep testing it.

Step 3 - Iteration

Iterative improvements is the key to CRO and A/B testing.

• If a test is successful and increases our chosen metric, then its likely we can do even better with a further test.
• Likewise if a test is unsuccessful, we need to review, understand why, and test something new. There was a reason we identified this area for testing, so it may just be that the execution was wrong.

Finally, once we have established why our tests won and the impact they had, this insight can be passed on to the wider business for further exploitation.

The learnings we gain from CRO can be hugely beneficial to the acquisition & retention teams, as well as product development, customer service and much more.

How did we work with OZARKE?

At Customers Who Click we believe in transparency and collaboration. 

• With Ozarke we had 1 call every week for the first month.
• We then move to a call every 2 weeks once testing beings to discuss the pipeline, any new ideas, and any updates from OZARKE (E.g. changes to the team, new tech platforms they are introducing, or new commercials and business models they want to explore.)

Not only can we provide insight and advice into these areas, but they may impact testing so its important we’re aware.

We shared our project management tool with OZARKE, allowing for a view on which tests are where in the pipeline, and also it allows us to tie communicate efficiently regarding each individual test, and allows us to gain an understanding of which types of tests work well, which dont, and which research is contributing to success and growth for OZARKE.

What results did we see?

Over the course of the first 3 months, we ran 7 tests for Ozarke, with an impressive 71% win rate (most testing programs achieve 10-20%).

We achieved this high success rate due to the intense research and planning that goes into our testing. We do expect this rate to fall once tests become more complex and nuanced.

Overall, our tests generated over $150,000 in additional monthly revenue for Ozarke.

Example Test - Cart Cost Breakdown

A very impactful test we ran for Ozarke was editing their slidecart and breaking down the costs being shown.

Initially, the cart simply had the full payable amount on the Checkout CTA, and did not mention anything to do with shipping costs.

As you can see in the image below of the winning variant. We removed the pricing from the button itself and placed it above. We also added a line to reinforce the key selling point that shipping was free.

Despite still being on the same screen, the control version linked the price too closely with the act of continuing to checkout. It’s seen as “i am going to continue to checkout to pay $X”. The new version simply confirms the price, reinforces that there is no extra charge for shipping, and then gives the customer the option to continue to checkout.

Example Test 2 - Search Bar Visibility

This is actually a test we have run several times now and almost always seen success. There  will always be customers who want to use search to browse a website and more directly get to where they want to be.

In the case of Ozarke, we knew customers were using search and they were converting. However the control version simply displayed a search icon inbetween account and cart.

Our winning variant simply opened the search bar slightly and added the text “Search Products”. This outperformed another variant which saw the search bar fully expanded in place of the Ozarke logo, which was placed to the side.

Want Results like OZARKE?