Go to Top

Revenue per Visitor Up 30%

Revenue per Visitor Increased Over 30%

The Site:

Dolls Kill is an online retailer of women’s clothes. They have built an incredible site and brand by focusing on a specific niche. With very little marketing, they have attracted over 260,000 Facebook followers in just a couple years. They obviously have an amazing site with solid conversion rates. Our challenge was to improve the site even further.

Identifying Areas to Test:

We do not just test random things like colors of buttons or trust badges. Our main strategy to improve a site is to analyze website data to find bottlenecks.

What are bottlenecks?
These are areas of a site that have high abandonment, low engagement, low conversion rates, low page value, etc. Basically, a part of the site that is underperforming

A Bottleneck Discovered at Dolls Kill:
When analyzing dolls kill, we noticed the bounce rate was higher than we thought it should be for landing page visitors to the product pages. Typically, with an ecommerce site, a high bounce rate for a product page usually means a visitor did not see the product they are interested in and left. This is very common, but we had some ideas to improve it.

Here is the original product page…


Our Solution:

To fix this bottleneck, we needed to find a way to get visitors to engage in the product pages even if the main product on that page does not interest them.

We designed a mockup and presented it to Dolls Kill. We wanted to give visitors more product options to browse from the product pages. We added a section at the bottom of the product page that features new products…


Dolls Kill has an amazing design team and they are very particular. They know their site and customer very well and recommended the following design…


Our First Split Test

We setup an A/B/C test to see which version will perform the best – our design, their design and the original.

The results:

After a significant amount of traffic and sales, the conversion rates for each variation were identical.

We were tracking custom variables in analytics for this test so we can have more detailed data. This data revealed that their design increased the Average Order Value by 35% and the Average Revenue per Visitor by 30%.

 Revenue per Visitor

Average Order Value

The new design did not get more customers. Although, by giving more options to visitors who were already prepared to buy, we drastically increased the Average Order Value and Revenue Per Visitor.

The Real Test

I am sure you have seen a page variation do very well in a split test, but fail to live up to expectations when you make that change to your site. There are many reasons for this. The most common is insufficient data.

For Dolls Kill, we ran a TON of data through this test. Even though the testing platform showed no winner as far as conversions, I knew we had enough data to see the improvement in Average Order Value and Revenue per Visitor in Analytics.

So, how did the new design hold up after it went live on Dolls Kill?


The new variation is holding up with an Average Order Value increase of over 20% in the last 3 weeks. The first week of the switch was a little low and now it is starting to do better. It will be interesting to see the AOV in a month or two. It should be closer to the 35% shown in testing, but I will be happy if it holds around 20%. That is a significant increase for a site like Dolls Kill.

Definitive… Measurable… Bankable Results!

Contact Us Now

Fill out the form to the right and we will contact you as soon as possible.
If you are interested in what we can do for your business, you should contact us as soon as possible. We take on a limited number of clients so we have plenty of time to focus on your business. You will work directly with me, not an employee following a script.

Contact me now and let’s talk.

Joe Vlcek

Attach RFP:


Visit Site

Value per Visitor Increased Over 30%

Services Provided:
Managed Conversion Rate Optimization