Performance hit fetching conversation history via /analytics vs /conversations

We’re refactoring the interaction log in our custom agent desktop. Previously, we just hit /api/v2/conversations/{id} to get the details. Now we need a list of recent interactions for the last 24 hours.

I noticed that there are analytics query endpoints. It seems like they should be the right place to go for bulk data, but the documentation is vague on when to use them over the standard conversation API.

I tried a quick POST to the analytics endpoint with this JSON:

{
 "groupBy": ["conversationId"],
 "dateRange": {
 "startDate": "2023-10-25T00:00:00.000Z",
 "endDate": "2023-10-26T00:00:00.000Z"
 }
}

The response takes about 2 seconds. If I just loop through /api/v2/conversations for known IDs, it’s faster for small batches, but obviously doesn’t scale. Is there a hard rule here? Do I only use analytics for reporting dashboards, or is it safe to use for real-time UI updates? The latency is killing the UX.