addressing reviews

Author: Rishab87

README.md

@@ -307,9 +307,9 @@ Other metrics can be sharded via [Horizontal sharding](#horizontal-sharding).
 
 ### High Availability
 
-For high availability, run multiple kube-state-metrics replicas with anti-affinity rules to prevent single points of failure. Configure 2 replicas, anti-affinity rules on hostname, rolling update strategy with `maxUnavailable: 1`, and a PodDisruptionBudget with `minAvailable: 1`.
+For high availability, run multiple kube-state-metrics replicas to prevent a single point of failure. A standard setup uses at least 2 replicas, pod anti-affinity rules to ensure they run on different nodes, and a PodDisruptionBudget (PDB) with `minAvailable: 1` to protect against voluntary disruptions.
 
-When using multiple replicas, Prometheus will scrape all instances resulting in duplicate metrics with different instance labels. Handle deduplication in queries using `avg without(instance) (metric_name)`. Brief inconsistencies may occur during state transitions but resolve quickly as replicas sync with the API server.
+When scraping the individual pods directly in an HA setup, Prometheus will ingest duplicate metrics distinguished only by the instance label. This requires you to deduplicate the data in your queries, for example, by using `max without(instance) (your_metric)`. The correct aggregation function (max, sum, avg, etc.) is important and depends on the metric type, as using the wrong one can produce incorrect values for timestamps or during brief state transitions.
 
 ### Setup
 

README.md.tpl

@@ -308,9 +308,9 @@ Other metrics can be sharded via [Horizontal sharding](#horizontal-sharding).
 
 ### High Availability
 
-For high availability, run multiple kube-state-metrics replicas with anti-affinity rules to prevent single points of failure. Configure 2 replicas, anti-affinity rules on hostname, rolling update strategy with `maxUnavailable: 1`, and a PodDisruptionBudget with `minAvailable: 1`.
+For high availability, run multiple kube-state-metrics replicas to prevent a single point of failure. A standard setup uses at least 2 replicas, pod anti-affinity rules to ensure they run on different nodes, and a PodDisruptionBudget (PDB) with `minAvailable: 1` to protect against voluntary disruptions.
 
-When using multiple replicas, Prometheus will scrape all instances resulting in duplicate metrics with different instance labels. Handle deduplication in queries using `avg without(instance) (metric_name)`. Brief inconsistencies may occur during state transitions but resolve quickly as replicas sync with the API server.
+When scraping the individual pods directly in an HA setup, Prometheus will ingest duplicate metrics distinguished only by the instance label. This requires you to deduplicate the data in your queries, for example, by using `max without(instance) (your_metric)`. The correct aggregation function (max, sum, avg, etc.) is important and depends on the metric type, as using the wrong one can produce incorrect values for timestamps or during brief state transitions.
 
 ### Setup