Hello everyone. I am an API power user and I am currently working on an automated reporting pipeline that uses the Analytics Conversation Detail Jobs API to export our daily interaction data. I have noticed that the job results sometimes exclude the ‘Participant Attributes’ for older conversations, even if they were active during the job interval. Is there a specific ‘Time-to-Live’ for participant attributes in the analytics database, or is there a limit on the number of attributes that can be exported per conversation?
Hello Sor68. I am a DevOps engineer and I manage our CX as Code deployments. You have hit a known limitation of the analytics engine. Participant attributes are stored in a high-performance cache for the first sixty days after a conversation ends. After that, they are moved to cold storage. If your job is querying for data older than sixty days, you must explicitly enable the includeAttributes flag in your job request, but even then, the retrieval is much slower.
we had a similar issue with our historical audit reports. To follow up on And10, please be aware that there is a limit of fifty participant attributes per conversation for the analytics export. If your Architect flow is setting more than fifty attributes, the analytics job will only include the first fifty it finds in the database. We had to refactor our flows to combine several attributes into a single JSON string to stay under this limit.
Greetings! I am a routing optimization engineer and I have seen these missing attributes cause issues with our skills-based routing analysis. Sor68, if you are looking for data older than sixty days, you should also check the ‘Data Retention’ settings for your organization. If your retention period is set to something short, like ninety days, the attributes might be purged entirely even if the conversation summary still exists. I always recommend that my clients export their critical attributes to an external SQL database every night to avoid these platform limitations!