Hello,
I wanted to understand implications to a segment's population when an underlying trait's qualified profiles are either updated or deleted using d_sid or d_unsid parameters.
For example, if I use "d_unsid" in my ingestion file to remove 5 UUIDS from my trait then -
1. How will the trait population will be affected in terms of Unique Trait Realization and Total Trait Population.
2. Assuming there is a segment built using just the above trait, how it's Real-Time Segment Population and Total Segment Population will be affected after file ingestion.
@shweta_singh @Ankita_Sodhi @Varun_Kalra
Thanks,
Har**bleep**
Solved! Go to Solution.
Topics help categorize Community content and increase your ability to discover relevant content.
Views
Replies
Total Likes
Hello @v-hapan ,
Thank you for writing to us! I hope you are keeping well
Before I start with the explanation, I would like to outline that d_unsid parameter which is responsible for removing any device/CID from a trait can't be used in real-time. However, you can use d_sid to qualify for any trait in real-time. Currently we do not support real-time trait removal.
d_unsid can be used with file ingestion to remove traits from any profile.
Coming back to query, assuming that our segment has only 1 trait, if we are uploading data files then we get our segment to reflect the updated data in 48hours window which will be visible in the total segment population.
If devices are not coming online much or CID are not having any authenticated activities then you might see a 0 real-time population and total segment population is something to rely upon.
For situations where you qualify for a trait in real-time, Edge server calculates the segment qualification based on the Profile Merge Rule and qualify a given device/CID for the segment in real-time. In this case, you will get your real-time population. Please note that if a device/CID is seen online more than once then every online activity will be counted under real-time segment population. For example, if device A which is part of Segment A comes online 5 times, not explicitly qualifies/requalifies for segment, then real-time segment population will be 5.
Additional note, If you wish to remove users/devices from a segment in real-time then you may want to think around frequency/recency settings or ANDNOT logic.
I hope this helps, please let me know if you have any additional question.
Hello @v-hapan ,
Thank you for writing to us! I hope you are keeping well
Before I start with the explanation, I would like to outline that d_unsid parameter which is responsible for removing any device/CID from a trait can't be used in real-time. However, you can use d_sid to qualify for any trait in real-time. Currently we do not support real-time trait removal.
d_unsid can be used with file ingestion to remove traits from any profile.
Coming back to query, assuming that our segment has only 1 trait, if we are uploading data files then we get our segment to reflect the updated data in 48hours window which will be visible in the total segment population.
If devices are not coming online much or CID are not having any authenticated activities then you might see a 0 real-time population and total segment population is something to rely upon.
For situations where you qualify for a trait in real-time, Edge server calculates the segment qualification based on the Profile Merge Rule and qualify a given device/CID for the segment in real-time. In this case, you will get your real-time population. Please note that if a device/CID is seen online more than once then every online activity will be counted under real-time segment population. For example, if device A which is part of Segment A comes online 5 times, not explicitly qualifies/requalifies for segment, then real-time segment population will be 5.
Additional note, If you wish to remove users/devices from a segment in real-time then you may want to think around frequency/recency settings or ANDNOT logic.
I hope this helps, please let me know if you have any additional question.
Views
Replies
Total Likes
Views
Likes
Replies
Views
Likes
Replies