FAQs (Audience)

My customer is trying to pull together segments using lists, but due to the limitation of nested segments, they are encountering an error. They need to add 24 more lists to this segment, but this issue is preventing them from doing so. Can you please investigate and address this?

The error occurs when a Customer List is based on a Segment that contains more than three filter conditions. If you're attempting to add a Segment with over three filter conditions, it exceeds the allowable number of operators within a single segment, leading to this issue.

I have several customer lists that, when combined, should encompass a large audience. However, when I create an audience using these lists, the customer count is noticeably lower than expected, with a significant portion categorized as prospects rather than customers. What factors could contribute to this difference, and what steps can help identify and address the issue?

Customer Count Discrepancy in Matched List:

• Main Count: Represents the unique individuals we estimate the segment includes, reflecting the total audience.

• Email: Indicates the number of identifiers in the segment that are permissioned for activation in the email channel.

• Display (Cookies): The number of segment identifiers that are available for activation in the display channel using cookie-based targeting (e.g., Sizmek).

• CTV (Connected TV): The count of segment identifiers that can be activated in the display channel for connected TV targeting, using IFA (Identifier for Advertisers).

• Mobile (MAID): Refers to the number of mobile identifiers in the segment that can be activated in the display channel, using mobile advertising IDs (MAIDs).

• Social (Email SHA-256): Represents the number of hashed email identifiers available in the segment for activation in social channels.

How long should I expect to wait for the counts to appear after enabling the required settings?

Please note that it could take up to 48 hours for the counts to reflect after the required settings are enabled.

In the client’s account, we have two segments with identical filters except for the lookback window: one with a 30-day lookback and the other with a 90-day lookback. Surprisingly, the segment with a 30-day lookback window shows a much larger count (142,037) than the 90-day lookback (399,495), even though I would expect the 90-day count to be at least as large.

To troubleshoot, I created two new segments with the same filters, adjusting only the event filter from “contains one of” to “is equal to.” This change resulted in counts that aligned with my expectations (the 90-day count was larger than the 30-day), but both counts were significantly lower than those in the original segments.

Could you help me understand why changing the event filter from “contains one of” to “is equal to” would result in such different counts?

The count discrepancies you're seeing are due to how the conditions "contains one of" and "is equal to" function differently.

With the "contains one of" condition, the system includes any record where the specified text appears within the field, even as part of a larger string. For instance, if the filter is set to "contains AB123," the system will count any record with "AB123" in it, even if it’s part of a longer sequence, such as "AB1234567." This broad match results in a larger count since partial matches are allowed.

On the other hand, with "is equal to," only records that exactly match "AB123" will be included. No additional characters or variations are allowed, making this filter much stricter. This difference is why you’re seeing lower counts when you use "is equal to," as only exact matches are counted, excluding partial matches that "contains one of" would capture.

Why is there a discrepancy in my matchlist count in the UI before and after export?

The number after export can be larger than the match size since the same input email might match more than one email address in the Zeta universe.

Why am I getting a validation error when running a query-based segment using %?

You need to add two % instead of one when used inside a query base segment.

Example:
and e.properties:campaign_tags::string like '%%BPS%%'
and e.properties:campaign_tags::string like '%%warming%%'

Why am I observing fluctuations in the segment's count with each campaign sent?

last time - The last action was performed within the specified range. This will exclude users who performed the operation today

any time - Any action was performed within specified range

For instance, if you performed an action on 5/24, the segment would not be included in a last time: 1-365 days query run on 5/25, as the action occurred outside the exclusion of 'today.' However, they will be included in any time: 1-365 days query run on 5/25.

Using within last captures the behavior, including the 0-day case.

Using the 'data flows' feature in the citizens-main-dev instance, I am able to move files with a .csv extension between FTP directories, but it doesn’t work for files with a .zip extension. Could you please confirm if it’s possible to unzip a file using 'data flows' and transfer it from one FTP directory to another?

The toggle ‘Unzip file’ is only applicable to gzip files. For regular zip files, you will need to add a separate node: "Input Compression Type." There is also a corresponding "Output Compression Type" node if the output file needs to be zipped.

• If the file is not a gzip, the ‘Unzip file’ toggle does not need to be turned ON.

• If the file is a gzip and needs to be decompressed, you must toggle ‘Unzip file’ ON. No additional node is required in the flow.

• You’ll only need the additional node if working with a regular zip file, though this node can also be used in place of the toggle for gzip files.

Why doesn’t the updated count for a segment immediately show in the audience UI after I make changes, run a count, and save the segment? It takes some time for the updated count to be reflected.

The counts are not persisted if they are calculated before the segment is saved, as they may be used for 'exploratory purposes' and might not be relevant if the segment isn’t saved. However, when you save the segment or if the segment definition remains unchanged, the count is then persisted. After saving, it may take a few seconds for the count to refresh and update. If you refresh the page after a short wait, you should see the updated count.

Why did my lookalike audience processing fail with the message that the audience segment must make a unique set?

If a Lookalike Audience (LAL) or Matched List is archived, and you attempt to create a new one with the same name, the process will fail. To avoid this, ensure that the new LAL or Matched List is given a unique name that hasn't been used before.

While building a prospect segment to run a customer pulse, why do all of the segment counts appear as zero?

The Audience Sampling feature is activated daily from the backend. Therefore, it is advisable to allow 24 hours for the Prospect Count to be generated. If the count continues to display as zero, this action should be performed from the backend.

Why does a user appear to have entered a segment in their journey path, even after meeting the entrance criteria designed to exclude them from the experience?

We strongly recommend using a unique email address for each user profile. Using the same email address across multiple profiles can negatively affect the user's journey and create complications in tracking and personalization.

I was trying to export a segment from my TA Partner instance and ZMP was not allowing the file to be sent via SFTP. Is it because the file is first-party data enriched with one of the public attributes?

This is because the segment might include demographic data that cannot be exported.

Why does the source file name remain unchanged when it's being transferred via FTP?

When replacing a file in the Matched List Segment, it’s essential to update the source file directly through the UI. Simply updating the file in the segment’s SFTP location will not properly update the segment.

• What criteria are used to determine if a user is included in the segment export? Is it any change to the person record or is it a change that is specific to the criteria of the segment?  

• Does it include the record if they are removed from the segment?

• The profile update "last_update" property is the criteria used to determine if a user part of the segment is eligible for the incremental audience export.

• No, it should not include the record if they exited the segment before the scheduled export run.

Upon creating a new ZMP instance for my client, an automatically generated segment named "Active" came into existence. Is there a way to prevent this default segment from being created on new instances?

Removing the default segment "Active" would result in the loss of valuable data related to customer activity and engagement. This segment plays a crucial role as a baseline for tracking user behavior, making it essential for future segmentation analysis and performance comparison with other segments. Therefore, deleting this segment is not recommended due to its significance in monitoring user interactions.

Why is the lookalike segment I created not showing up in the Audience Explorer?

Kindly access the missing lookalike segment in the Audience Explorer list and select 'Rescore Audience'. After the rescoring process is finished, this lookalike segment should then be visible in the Audience Explorer list.

What is the difference between other segment and query-based segments, and are there any expected limitations of the two?

Generating a lookalike or constructing a list isn't practical with a query-based or external database segment because these segments are dynamic and formed in real time according to particular queries or conditions. Conversely, other segments are built using predefined criteria or attributes, such as demographic details (age, gender, location) or behavioral data (purchase history, website engagement), which remain constant once established until they are manually adjusted. Unlike regular segments, which are static, the conditions of query-based segments or external database segments might encompass intricate combinations of data attributes, interactions, or events.

The count estimator for my campaign displayed there were a significant number of records that fell under "Other Reasons". What are 'other reasons' for an unsubscribed user being skipped from a campaign?  

“Other Reasons” would generally be more explicit and implicit unsubscribes not based on deliverability. These would include unsubs from email links, unsubs from preference centers, spam complaints, response unsubs where people have replied to a campaign, or maybe manual unsubs where someone logged into the platform and marked an individual as unsubscribed, imported them, or used the API to unsubscribe the person.

Is there an easy way to unsubscribe a large group of people?

Use the built in "Contact Unsubscriber" tool. Note that this requires the user to have Super User access to the instance in question. To use this option, go to Settings > Audiences > Data Tools > Contact Unsubscriber. From there, drag and drop a CSV file with just a "contact_value" header and a list of contact addresses to be unsubscribed. 

When we configure the Inbound Message Handler (IMH) to NOT process unsubscribes, we have the option to pass the unsub request onto a configured address. How do we record the response in this case?

In case of forward_all_replies being false, we don’t forward the response. Still processing the unsub request, we unsubscribe the user and also generate the unsub event.

What is the concept of Net People?

If there is an audience/segment filter,
Net people = Number of users who entered/exited a specific segment.

But if we removed the audience/segment filter,
Net people = Number of users who entered/exited all the segments on the entire account.

While creating a few segments, I had turned the auto archive option on. Why have those segments not been archived yet and are showing the status as Active?

Your segment(s) might be used in a Draft campaign, hence were not archived on the mentioned date. A segment gets archived only when it's not being used elsewhere within the platform.

Are unsubscribe clicks counted within unique click counts?

Clicks on unsubscribe links are counted as a unique click. This is because when we click on the unsubscribe link, it reflects an unsubscribe event as well as generates a click event. So, both events are reflected in reporting and counted as a click and as an unsubscribe.

Why does my customer segment show the status “sent”, but hasn’t been delivered yet?

After a segment is done processing, the system takes up to 20 minutes for the audience to be updated in the backend.

While exporting some segments, why am I noticing some new columns for “zync_cookie” and “email_md5_id” included in my exports?

New fields being included automatically in exports were part of a platform update, and it was intentional. The new fields are “zync_cookie” and “email_md5_id”.

My customer has tried to unsubscribe from the email list by clicking on the email unsubscribe link found in their campaigns. Why is it not working?

There are two steps for a user opting to unsubscribe through the Unsub link:

• Select the Unsub link in the email, then Confirm the email,

OR

• Submit the step in the subsequent web page

Only then will the user be unsubscribed from the platform.

Is it possible to change the Snowflake Share timezone from UTC to Eastern time?

We are unable to modify the time zone for Snowflake Share.

Can we share segments via the team-based sharing feature from Experiences or Campaigns?

Not currently. In the team-based sharing access, you can’t share or restrict access.

How can you tell if a segment has campaigns linked to it?

You can manually check your campaigns or use a clear naming convention to filter those with segment links. For alternatives, reach out to Support for assistance.

What is the default segment on the dashboard?

The segment name is ‘active’ and it includes all customers who were last seen or whose last contact was within the last 120 days.