I am going to just let these two screenshots of the AB Test do the heavy lifting of today’s post...
Top Line Metrics:
Not a while lot of difference here. As far as AB tests go, this would be called inconclusive. When I was using a different experimentation app, whatever train-of-thought led to this test would be derailed and I’d move on with my life.
But, we added in he engagement tracking… because it sucks to let a good test go to waste.
Engagement Metrics:
Something interesting is starting to peek through here…
(Group A is the group with the change made… a glitch in my own workflow, I set it up backwards… but Group A is the focus here)
Group A: 68% clicked on a product card image
235 visitors
160 clicks
Group B: 47% clicked on a product card image
266 visitors
126 clicks
That is quite a bit more engagement on the collection page, which is good. But look at the 3rd line of the images.. that’s a 50% increase in clicks on the StableToe.
Little context - StableToe is 1 of 3 toe spacers that we sell. It is a great product that people love, but it is harder to explain than the others, because it is so new, and there is less content online about it.
Seeing this increase in attention is quite significant for this product line.
All we added was a blurb under each of these 3 products.
That was the whole test.
Even though there is little difference in the top line metrics, there is a big ‘ol signal here that people are interested… we need to do more to showcase it.
So in conclusion… I need to what every operator should do in this case:
- hide the results
- fix the problem before my boss sees it
- then do the test anew so I come out looking like it was all part of the plan.
Right, shit…geez I gotta jet.
- Shep
Comments
Nothing yet. Say the first thing.
Sign in to join the conversation.