My thought is that the best practice here would be if a product is viewed after applying a filter as a Y/N, as counting total products viewed might indicate that the customer was not able to filter deep enough to find the right product.
We currently do not have an event implemented to track this, and I'm hoping that I can use a combination of product list filter clicks, product list tile clicks and a product view in a sequential segment and/or a custom metric. However, as I build this out, I'm getting stuck with a visit segment that includes product views that occurred from this specific click path AND product views that happen from another finding method within the visit, rather than the simple Y/N.
Am I thinking about this the right way? Has anyone solved this already?
Topics help categorize Community content and increase your ability to discover relevant content.
Views
Replies
Total Likes
I am not sure I am following your exact need, but I think you are trying to find a
"product viewed" immediately after the use of "product filter" being used?
Did you try to add within logic to your sequential segment?
Or, maybe the problem is that once you are looking at a Visit level segment, you are getting ALL Product Views from the visit?
If this is the problem, then I think what you might need to do is to create a HIT scoped segment, with a VISIT container inside it. But there are still complications, because the biggest problem here is that there is no "only this sequence" in the segment logic... so you get stuff outside the sequence.
Now, I think I can get something close but it still probably isn't perfect....
Here is what I came up with while tinkering:
HIT [
Product View exists
AND
VISIT Container (only after sequence) [
Product List Filter Click exists
THEN within 1 hit
Product View exists
]
AND
EXCLUDE HIT Container (only after sequence) [
Product View exists
AND
VISIT Container (only after sequence) [
Product List Filter Click does not exist
THEN within 1 hit
Product View exists
]
]
]
So what I am thinking here, is:
This still may need tweaking... and it may never be perfect... Adobe's segments really aren't optimized for pulling back data within a small specific sequence....
Views
Replies
Total Likes
Hi @dkleinschmidt
You are thinking about it in the right direction, the key is to measure whether a product view occurred as a direct result of applying a filter, rather than across the whole visit. Counting all product views can muddy the insight, since users may explore products in multiple ways.
A clean approach is -
Capture a filter application as an event (e.g., “Filter Applied”).
Attribute the next product view (within the same hit or a short sequence) to that filter.
Create a binary metric (Y/N) - “Product viewed after filter = Yes” : to simplify success measurement.
This avoids inflating numbers and gives you a direct view of whether filters are helping users find relevant products. If you don’t yet have a dedicated event, you can approximate it using sequential segments (Filter > Product View), but a true event implementation is the best long-term solution.
Views
Replies
Total Likes