What is the best practice for data structure and aggregations in ES?

Elastic Stack Elasticsearch
1
  1. How to structure the data in elastic search: A) We are pulling data from multiple database tables.
    B) Those tables have elements that connect them to together
    C) We need to update the records multiple times (e.g. opportunity as it transitions through sales stages, confidence in closing, etc.) - in addition to wanting to update some of the data based on other calculations
    What's the best way to structure the data given those characteristics?
  2. Aggregations - What is the best practice for aggregations?
    A) Is it better to run a cached filtered query and then run aggregations against it, or is it better to run multiple nested aggregations (potentially with scripts).
    B) How much are nested aggregations used and supported?
  3. Connection between ES and graphic tools
    A) What is the most common graphic tool that people use with ES?
    B) Are there standard connectors that people use?