Hello!
I am looking to enhance personalization a bit on our site, customers are able to do multiple types of transactions on our site, they can pay their bill, update service, or make ecommerce based purchases. From funnel analysis we can look at the volume of unique visitors that view products vs services but what I am more interested in is a cumulative view.
For instance, if I look at the fallout to determine that 10% of my traffic views ecommerce and I look back 12 months, the top level number grows too to create a trend. However, from a segmentation perspective I would prefer to capture the total cumulative volume of traffic that has ever gone there. If I know, for example you viewed product A in June, even if you never viewed it again, I would prefer to show you that product even if you've dropped off from a unique visitor perspective in August to that portion of the flow. So if I roughly have 10% of my traffic every month that has viewed ecommerce every month over the last 12 months, what percentage of my traffic that regularly logs in has ever viewed ecommerce? It's probably not the same 10% every month.
I believe the solution would be to create a segment at the Visitor level, add the condition that the user has viewed the ecommerce section, apply the segment to a report that shows traffic to the home page since majority traffic is routed through that page. Then, I would make a calculated metric from that table (unique visitors home page commerce viewers)/(total unique visitors home page no filter)*100 to get a percentage. I think this would work because it should make the home page my control, but would love to hear feedback!
Edit: A cohort analysis is telling me that based on the inclusion action of "viewed of a product" I have X percentage of customers who perform the same interaction month over month in the given time frame, so I could also simplify and extrapolate from there I assume?
