Immediate Next Page after sequence A→B

[JakeRi]

I’m struggling to understand how to create a segment that will allow me to generate a list of page URLs that users visit after following a specific sequence.

For example, a user follows this path:

• Page A
• Page B
• Page C

I want to create a hit-based segment that focuses only on the Page C hits (step 3 in the sequence).

Currently, I have a segment set up as follows:

• A sequence where Page URL = A, then within 1 hit Page URL = B, then within 1 hit Nested Page URL exists.
• The overall container is set to Only After Sequence
• .

Can someone help me understand how to make this segment return only the Page C hits?

Additionally, could someone explain why you create a hit-based segment using a Visit or Visitor-level container that includes a sequence?

 

[JakeRi]

clearer image of segment

[Jennifer_Dungan · Accepted solution]

This segment, I believe will pull back ALL pages following the sequence, not just the immediate page after Page B.

If you look at the sequence built by the "Next Item" panel, they have additional logic to exclude the remaining pages.

HITS [

    HIT Container [

        Page (non-repeating instance) exists

    ]

    AND

    VISIT Container (Only After Sequence) [

        HIT [

            Page (non-repeating instance) equals Page A

        ]

        THEN Within 1 Page View

        HIT [

            Page (non-repeating instance) exists

        ]

        THEN

        HIT Exclusion [

            Page (non-repeating instance) exists

        ]

    ]

]

This segment looks for the next page after Page A... in your case, you want Page A > Page B > Page ? 

I would try to add Page B into the sequence here and see if that gets you closer to what you need, as opposed to all pages after the sequence.

Good Luck! Let us know if you are successful.

[bjoern__koth]

Wonder if the Entry value of the dimension is actually affected by the segmentation. If so, "Entry Page" might work, too after the sequence?

[JakeRi]

Hello Bjoern!

Can you expand on this comment? I'm not quite sure how Entry value plays a part in this segmentation.

Cheers from Canada!

[JakeRi]

Thank you for the suggestion!

Do you have any ideas on the best way to validate if the results are correct?

[Jennifer_Dungan]

I will usually test several sequences using new incognito sessions in my QA environment.

I will take note of all my ECIDs (which I track in an eVar), or if you don't have that, use something like a campaign code ("test1", "test2", "test3", etc) so that you can isolate each test.

Then I will test sequences that should be included, and should not be included, and also try to test sequences that have back and forth logic, or that hit the sequence multiple times, etc... basically I try to catch as many scenarios as possible.

I keep track of what I did in each test, then wait an hour for the data to appear in my suite, and test what comes back in my workspace (making an overall segment to include all my test runs - confirming they all there), then testing the sequence segment and see if I get the tests back that I expect... if not, then I would try making tweaks (or even additional segments for comparison with minor variations) and see if I can get what I am looking for.

[JakeRi]

Hi Jennifer,

One quick follow-up question: Would the final HIT Exclusion condition — Page (non-repeating instance) exists — be interpreted as excluding all hits that occur after Page C, or does it exclude entire visits where any page is viewed after Page C?

[Jennifer_Dungan]

It should exclude all the hits after the sequence, not the visit.. but this is where testing comes in, testing sequences that start with and then have other pages, hitting pages then ending with the sequence, having the sequence in the middle, and even having the sequence but with pages in-between (which shouldn't be included if you are looking specifically at A > B > C)