Client: Top Condoms Canada
Split-Testing Platform: Optimizely
Site Type: Hosted E-commerce (Multi-Product)
Test Approach: Radical Redesign A/B Split on entire conversion funnel (desktop only traffic)
Objective: Increase in sales and revenue
Commentary and Critique on Original Site (Control)
The conversion rate of Top Condoms Canada has historically been healthy, so the co-owners believed the conversion rate could not be improved by much. In addition, the site is selling a range of commoditized products which are widely available.
We identified some possible areas of improvement for conversion testing, based on 4 main areas: initial experience, value proposition, friction and anxiety. Below we highlight some of the major issues we looked at.
While a visitor to the website can probably tell that Top Condoms Canada is an online e-commerce shop retailing condoms in Canada from the name, we wanted to make it even more obvious. You usually only have the first 3–7 seconds to catch the attention of the visitor, and in those short few seconds we want to ensure that we communicate what the website is about and what they can do here.
There was no clear value proposition identified and communicated on the website. Having a clear value proposition is important for all businesses. This was even more important for Top Condoms Canada as they are mainly selling condoms, which are widely available.
There needs to be a good reason why a customer should buy from this site as opposed to buying from a competitor’s e-commerce site, or a regular brick and mortar shop.
There wasn’t a clear path to the main conversion goal of a purchase. A good number of the visitors land on the homepage, and it isn’t particularly clear what action should be taken.
On the product page, the add to cart button is the same color as the heading. The checkout and view cart features after the potential customer has added a product to the cart just consists of 2 simple text links which are the same color as the rest of the sidebar navigation.
The lack of an obvious navigation path to the conversion goal is a further cause of friction.
Based on customer feedback, it was determined that having discreet shipping for all the products on the site was a feature they liked, as it protected their privacy. This is a feature that has been standard for all items shipped since Top Condoms Canada was started.
However, the fact that discreet shipping is standard across all orders is not immediately obvious to the potential customer.
In addition, there is potential for some anxiety for the potential customer when purchasing from a smaller niche-shop as compared to purchasing from a bigger e-retail shop such as Amazon. First time customers might be thinking that this relatively unknown shop might be a fly-by-night operation.
Based on the above analysis, we came up with a hypothesis that the conversion rate of the site (as defined by the number of unique visitors who purchased a product), and also the revenue-per-visitor would increase if we addressed the above issues.
Note that these are not the only things we changed. As the radical redesign consists of many changes throughout the conversion funnel and would be too lengthy to list, we just highlight the main points here for Conversions Guaranteed readers.
Here are the specifics:
We decided to test a core value proposition of having a strong money-back guarantee. This guarantee policy has always been in place at Top Condoms Canada, however it was not immediately obvious.
Our new treatment variation expressed this core value proposition at all stages of the conversion funnel as well as on each individual page on the sidebar for increased congruence and continuity.
Significant traffic was going directly to the homepage, so we replaced the default banner graphic which didn’t express much value with a newly designed banner with a call to action button to the most popular condom products.
Supporting the core value proposition, we also made it clear that products from the shop are premium quality brands. The process level value proposition (on call to action buttons for example), was also reinforced where possible.
We redesigned the header to be even more explicit about what the site is about: providing premium quality condoms in Canada online, and included the main value proposition statement.
We created a default clear path for users landing on the homepage in the form of a visually appealing call-to-action button to quickly to go to a page where they can see a list of the most popular products on the site. Visitors who already know what they want have the option of shopping by brand or type of product via the navigation on the sidebars.
On this popular products, the copy on the control page was replaced with copy that further reinforced the value proposition and reduced the anxiety and friction for the visitor.
Having this new conversion flow reduces friction and helps guide the visitor towards the intended conversion goal (a purchase) instead of leaving the visitor to wonder “what should I do now?”.
The buttons representing the core conversion actions were also made bigger and more obvious by using a different color (orange). The order on the sidebar cart preview was also switched around to emphasize our preferred conversion action (proceed to check out).
Running the Test
We ran the test solely on desktop (non-mobile) traffic, as the BigCommerce checkout process differed for mobile browsers.
We used Optimizely as the split test platform for this test. This test was done before Optimizely launched their multi-page testing feature, so testing one funnel against another was more challenging as we had to use their custom variation code.
We now use their multi-page experiments feature for our funnel test and this has cut down our implementation time significantly.
Another technical challenge was the BigCommerce hosted e-commerce platform used by Top Condoms Canada. The site used AJAX extensively in their checkout process, so parts of the experiment had to be custom coded by using the Optimizely API.
The test was initially started in end-December. We found that there were wild fluctuations in the conversion patterns in the final week of December, around the New Year holiday in Canada.
Although the sales pattern of the site historically does not show a seasonal effect on conversions, we decided to restart the test after the holiday season to try to reduce the probability of a validity threat to the data.
The radical redesign resulted in a 53.7% increase in number of sales and a 63.7% increase in revenue, both at 95% statistical confidence.
Note: Parts of the screenshots have been redacted to protect private client information.
Addressing the combination of initial visitor experience, value proposition, friction and anxiety was very effective in increasing the conversion rate of the website significantly. This was in spite of condoms being a product which is commonly available rather than being completely unique.
This good conversion rate increase was possible even though the pre-optimization conversion rate was already pretty high.