Creating segments based on belonging to traits in sequence | Community
Skip to main content
edwardb89593165
edwardb89593165
November 21, 2017
Solved

Creating segments based on belonging to traits in sequence

  • November 21, 2017
  • 3 replies
  • 3169 views

I work in destination marketing and am trying to use AAM to assist in working out the effectiveness of our marketing. I think a possible solution would be to generate segments based on sequential trait rules but am not sure how to do this.

For example:

Trait 1 - User is not located in X

Trait 2 - User is located in X

The segment logic I'd like to build is that the user originally belonged to trait 1, then qualified for trait 2 (i.e. person not located in destination views our website from their home market, then comes to the destination and views our website while on holiday).

Is there a way to build a segment which captures all people in trait 2 who were previously a member of trait 1?

Any assistance would be much appreciated.

    This post is no longer active and is closed to new replies. Need help? Start a new post to ask your question.
    Best answer by costi_muraru

    Hi edwardb89593165​,

    That is correct. Today you can do a sequential segmentation using Adobe Analytics and export that segment to Adobe Audience Manager.

    There is also a plan to support this natively in Adobe Audience Manager, though I personally don't have an ETA for it.

    There is also a minimal way of achieving something remotely to what you are describing and that is by using a recency frequency rule in AAM.

    You can create a segment to group this audience: "All the users that have been located in location X in the last 7 days and were previously marked as not being located in location X."

      frequency([1000T] >= 7D) >= 1  AND  frequency([2000T] <= 7D) >= 1

    • 1000T ---> User is not located in X
    • frequency([1000T] >= 7D) >= 1 ---> User was not located in X in the time interval (profile creation time, NOW - 7 Days)

    • 2000T ---> User is located in X
    • frequency([2000T] <= 7D) >= 1  ---> User was located in X in the last 7 days

    You can also play with the number of qualifications for a specific trait. (Eg. Never looked over the "Ski Equipment section" till 7 days ago and now the user looked at it for 10 times in the last 7 days).

       frequency([3000T] >= 7D) == 0  AND  frequency([3000T] <= 7D) >= 10

    Hope it helps.

    Cheers,

    Constantin

    Computer Scientist | Audience Manager | Adobe

    3 replies

    KCTDOT17
    KCTDOT17
    Level 2
    November 21, 2017

    Hi Edward - When I was at the Summit this year, this was shown as one of the new features - sequential segmentation - at the time we were told it should be launching later this year.  I did a quick search in AAM help but didn't see anything come up.

      edwardb89593165
      edwardb89593165Author
      November 21, 2017

      Thanks. They might have been talking about Analytics, as there seems to be some information about sequential segmentation via AA (i.e. someone visits website from country X followed by country Y), which could be used to create an Analytics segment that can be pushed into AAM, however in many cases I'm basing a segment on a pixel trait on a 3rd party site, which rules out the AA solution.

      I've searched through the AAM documentation and help to see if it can be done via AAM only and can't seem to find anything.

      costi_muraru
      Adobe Employee
      costi_muraruAdobe EmployeeAccepted solution
      Adobe Employee
      November 24, 2017

      Hi edwardb89593165​,

      That is correct. Today you can do a sequential segmentation using Adobe Analytics and export that segment to Adobe Audience Manager.

      There is also a plan to support this natively in Adobe Audience Manager, though I personally don't have an ETA for it.

      There is also a minimal way of achieving something remotely to what you are describing and that is by using a recency frequency rule in AAM.

      You can create a segment to group this audience: "All the users that have been located in location X in the last 7 days and were previously marked as not being located in location X."

        frequency([1000T] >= 7D) >= 1  AND  frequency([2000T] <= 7D) >= 1

      • 1000T ---> User is not located in X
      • frequency([1000T] >= 7D) >= 1 ---> User was not located in X in the time interval (profile creation time, NOW - 7 Days)

      • 2000T ---> User is located in X
      • frequency([2000T] <= 7D) >= 1  ---> User was located in X in the last 7 days

      You can also play with the number of qualifications for a specific trait. (Eg. Never looked over the "Ski Equipment section" till 7 days ago and now the user looked at it for 10 times in the last 7 days).

         frequency([3000T] >= 7D) == 0  AND  frequency([3000T] <= 7D) >= 10

      Hope it helps.

      Cheers,

      Constantin

      Computer Scientist | Audience Manager | Adobe

        Terms & ConditionsAccessibility statement

        Loading footer...