TTL drop out and come back

[akwankl]

Hi all,

My first post and I am a complete AAM newbie. We have a rule-based trait (and a segment that uses this single trait) that qualifies a customer if they fall into specific incentive programs. We discovered recently that data hasn't been pushing to Adobe for the rule-based trait to determine the qualification and so we have been seeing 0 traits realization. The trait has 15 days TTL. Assuming it worked before, let's say 6 months ago (March 2020), and broke down a month after (April 2020). 

Is it correct to assume that:

1) Those who qualified in March have dropped out of the trait

2) For those who dropped out of the trait, they can no longer qualify back this trait since they are passed their TTL?

Even if we fixed the data issue, the traits will only realize for those who were never realized back in March? And so the best bet is to recreate the trait and segment?

Thanks,

Alan

[allanwitts · Accepted solution]

Hi Alan

The trait TTL specifies how long the trait should remain on a profile after the last date the trait was realized. So with a TTL of 15 days, the trait will only stay on the user profile for 15 days after the day the user last qualified for the trait. However, it does not mean that the user cannot qualify for the trait again. If they perform the action to qualify for the trait again, then the trait would once again be added to their profile, and would stay there for 15 days (or longer if they realize the trait again within that 15 days).

Looking at an example:

• User A qualifies for the trait on day 1 (lets say the trait is to capture visits to the "Contact Us" page on the website)
• The Contact Us trait is added to the profile of User A
• User A doesn't visit the Contact Us page again within the next 15 days
• On day 16, the Contact Us trait is removed from the profile of User A
• On day 35, User A once again visits the Contact Us page, so the trait is once again added to their profile
• On day 40, User A once again visits the Contact Us page, this means they "realize" the trait again, which essentially resets the 15 day countdown.
• On day 50, the trait remains on their profile
• On day 55, the trait is removed from their profile.

So to answer your specific points:

1) Those who qualified in March have dropped out of the trait -> Correct

2) For those who dropped out of the trait, they can no longer qualify back this trait since they are passed their TTL? -> Incorrect (see example above).

Hope this is clear now!

[akwankl]

Amazing, thank you allanwitts

[akwankl]

Hey @allanwitts ,

I have additional questions while I was pondering on your answer. What if this the data source is an onboarded data source that is a manual list upload. I couldn't find any documentation, but say if the list is:

UUID	Trait
12345	"Person_Bought_X" = "Y"

and the TTL is set to 15 days.

For an uploaded list & trait like this, can I assume the workflow is:

1) I uploaded a list and these are the people that qualify for it

2) The moment they log into my property and realized this trait to the profile, the 15 days TTL starts OR is it the moment this list is uploaded the TTL starts?

3) Because it's an uploaded list, there is no such thing as requalifying since it's not a web trait (ex. Person x visited this page) so the person only has 1 TTL cycle

4) If I reupload the list with a .overwrite (is this any different with .sync), everyone on that list is reset and the trait is realized again. With the .sync approach, people who were in the previous list AND in the new list will not realize the trait again.

Thanks.

[allanwitts]

@akwankl- The situation is different for onboarded traits. In that case the "realization" happens when the file containing the key-value pair to qualify for the trait is processed. So with a TTL of 15 days, if the information is not received again within 15 days of the file being sent, the trait would be removed from the user profile.

The times that a user comes to site does not affect the onboarded trait being removed from a profile due to TTL - a user could come to site every day, but the onboarded trait would still be removed from their profile if the file is not uploaded again within the TTL period.

For this reason, it is common to have long TTLs for onboarded traits, or to set the TTL value to 0 such that the trait is never removed from the profile.