MFOSTER | 2021-04-26 11:36:24 UTC | #1
Hi, Can you let me know what latency we should expect on the response for the analytics API? We are seeing certain metrics for the email media type only appear in API results several hours after the segment start date meaning that the counts like nConnected are changing for a previously polled interval.
And other cases where the raw metrics are coming back from the conversations don't agree with the aggregation results for a particular interval (often the raw data appears to show 1 extra count compared to the aggregation) Yet, 1 day later when polling the same period they then appear to match. It's fairly clear that the aggregation is being recalculated and changed, many hours after the interval finished.
Is there any information on when the conversation aggregations are re-calculated, and what latency we should expect for metrics (particularly email metrics).
thanks!
anon11147534 | 2021-04-28 16:39:48 UTC | #2
Hi,
Analytics jobs data can be delayed by up to 24 hours as it is populated nightly. If you are experiencing delayed data with query endpoints please open a ticket with Care to investigate.
MFOSTER | 2021-04-29 17:11:42 UTC | #3
HI Ronan, thanks, will raise a ticket for the latency. For the aggregation recalculations, do you know what time of day they are performed? Is it nightly for each region? so for london region it would be after 2am local time or something?
anon11147534 | 2021-04-29 13:54:32 UTC | #4
Hi,
It is nightly for each region. The data is processed sequentially for each org so the specific time is dependent on where your org is in the queue. Typically it would not be during business hours though. For more information see Analytics Jobs Data Availability
MFOSTER | 2021-04-29 17:10:34 UTC | #5
hi, the link you sent is for analytics jobs, which we don't use. We are using the below endpoints
/api/v2/analytics/conversations/details/query and /api/v2/analytics/conversations/aggregates/query
Can you confirm whether these endpoints also have their data recalculated and changing over 1-2 days? Since this is what we seem to be noticing.
tim.smith | 2021-04-29 17:24:59 UTC | #6
Analytics data in Genesys Cloud is recalculated nightly as part of a data reconciliation process. This is part of the Data as a Service (DaaS) approach and is one of the reasons that analytics data should be retrieved from the API on-demand; data fetched on-demand will always be the most accurate representation available at the time it is retrieved. The nightly recalculation allows for fixes and changes in the analytics service to be applied retroactively to data that was previously collected, thereby increasing the accuracy of historical data. However, take note that this process is entirely separate from the jobs data availability Ronan mentioned above.
If you believe the data you're seeing is erroneous or need help understanding the aggregate values in context of your detail data, please open a case with Genesys Cloud Care to investigate further; specific customer data cannot be investigated in a public setting such as this forum.
MFOSTER | 2021-04-29 19:34:22 UTC | #7
I see, thanks for the explanation Tim - I'll raise a case on the latency issues we are seeing with emails.
thanks!
system | 2021-05-29 19:41:26 UTC | #8
This topic was automatically closed 30 days after the last reply. New replies are no longer allowed.
This post was migrated from the old Developer Forum.
ref: 10727