Null values should be ingested into AEP?

[ADLSSt200]

Hi, my data has some of the null value for the optional field in AEP, should I ingest these null records? How will it affect profile and identity stitching?

[Shubham_AEP]

Based on my Best understanding,

In Adobe Experience Platform (AEP), you can ingest records with null values for optional fields, but it’s best to omit those fields if no value is available.

For identity fields (like email or CRM ID), never send nulls, this breaks identity stitching and prevents proper profile merging.

Null values might also affect the audience segmentation, as profiles with missing data may not qualify for certain segments. ---- If you send null values in identity fields (like email, ECID, or CRM ID), AEP will not stitch the data to an existing profile, because it relies on valid identity values. This means the data becomes an orphan fragment, not connected to any unified profile. Over time, this can lead to incomplete profiles, failed segmentation, and broken personalization.

To ensure clean data and accurate results, only send meaningful values and exclude fields when data is not available.

Hope this will help you.

[brekrut · Accepted solution]

A null value is a relational database term which is used to represent a value which does not exist in the column of the database table.

If data is ingested into the Adobe Experience Platform profile store the data will be represented as the text of null or an empty string depending on the transformation used to ingest data. 

Segmentation has conditions of exists or does not exist which represent if a value is present within the attribute.  If you wish to have a null value present in the datalake and promote this upon profile do not populate the attribute with any string value to ensure the segmentation criteria does not exist picks up the condition correctly.

[KevinFosterDX]

Null values are fine and expected for fields that aren't marked as required.

But while zero length strings are also allowed, these should be prevented to avoid complicating audience rule definitions. Best practice is to set a minimum length value on the schema definitions of string fields.

But be careful... this is captured at the field group level. So it can affect multiple schemas and multiple datasets including that field group, so would need monitoring of all related dataset ingestion to discover if anything is sending zero length strings. Behavior should be that the record is accepted and the field is saved as a null, but would definitely confirm that via dev sandbox testing before configuring in a production sandbox.

[RiteshY18]

@ADLSSt200 it depends on the use case and your schema design. It's good to have a null value in the schema if planning to use it later for your marketing needs.

- You can use a null value during segmentation or for audit purposes to check missing values

- If the field is optional and you don't need to track or need it in segmentation, then it's better to omit rather than increase the data size with a large number of null values. 

- If you want to clear out an existing value, you send/use null values.

[KevinFosterDX]

Best practice is to set a minimum length greater than zero on string fields in your schemas. Always. There's never a time where a zero length field value is a better choice than a null. 

The AEP UI now adds "minimum = 1" by default when you add a new string field to any schema for just this reason, and while you can override that you should not.

[kautuk_sahni]

@ADLSSt200 Just checking in — were you able to resolve your issue?
We’d love to hear how things worked out. If the suggestions above helped, marking a response as correct can guide others with similar questions. And if you found another solution, feel free to share it — your insights could really benefit the community. Thanks again for being part of the conversation!