I’m back with the monthly dose of product updates.

What comes next is not for the faint-hearted.

?Update #1: Create experiments based on buying behavior.

There are 99 million bicycles in Beijing. That’s a fact. Like the fact that we have over 40 parameters (ex: weather, location or days since the last visit) which you can use to segment your customers when applying experiments.

We recently added a new one to the list. It’s called Magento groups* and it enables you to target only specific customers with specific buying behaviors.

To stay relevant it’s important to customize your message and treat your customers fairly. We use the power of the RFM segmentation, to shed light over who is your VIP customer, your loyal one, the most active or the newbie.

*The option is available only for Magento but we are working towards extending this.

 

?Update #2: Assisted conversions vs Linear Distribution

It’s easier to work with an example here.

??Jane Doe visits a fashion website. Upon reaching the homepage, she is included in an A/B which tests preferences against the color of the “Search” button.

On the category page, Jane sees a ?, at the price of 30$, she likes and adds the ? to her ?.

Upon reaching the cart pages, she is in a hurry and is on the point of exiting the page, when a pop-up offering Free Shipping appears. After this, she decides to ✔️buy the shirt right then.

Assisted Conversions: 
This reporting model displays the number of conversions on each experiment without taking into consideration the fact that the user made a conversion on the same goal in two different experiments or even more than two experiments.

Following the example above, for Jane, the Sale conversion is added as 1 conversion with the value of 30$ on the A/B test and 1 conversion with the value of 30$ on the popup experiment.

Linear Distribution:
This reporting model displays the number of conversions on each experiment for each user by dividing this number to the number of experiments the user is included in; the value of the conversion is also divided.

Following the example above, Jane was included in 2 experiments, which means that she generated a 0.5 conversion with the value of 15$ on the A/B test and a 0.5 conversion with the value of 15$ on the popup with Free Shipping experiment.

 

?Update #3: Goals filtering

Because we know it is difficult to follow statistics when there are multiple goals set up, we have implemented a goal filtering option in the view section of the experiment.

 

?Update #4: Omniconvert integration with Woopra

We have great news for those of you using Woopra for your customer analytics – you can now integrate it with the Omniconvert platform, as shown in this news item on our help page.

That’s it for now, you, smart converter!
Want a helping hand to make your experiments rise and shine?

Let me know at [email protected]

8 Shares
Share8
Tweet
Share