diff --git a/docs/reference/ilm/actions/ilm-rollover.asciidoc b/docs/reference/ilm/actions/ilm-rollover.asciidoc
index 9c4489bac1a94..4731986bd2559 100644
--- a/docs/reference/ilm/actions/ilm-rollover.asciidoc
+++ b/docs/reference/ilm/actions/ilm-rollover.asciidoc
@@ -129,10 +129,10 @@ opt in to rolling over empty indices, by adding a `"min_docs": 0` condition. Thi
 disabled on a cluster-wide basis by setting `indices.lifecycle.rollover.only_if_has_documents` to
 `false`.
 
-NOTE: The rollover action implicitly always rolls over a data stream or alias if one or more shards contain
-      200000000 or more documents. Normally a shard will reach 50GB long before it reaches 200M documents,
-      but this isn't the case for space efficient data sets. Search performance will very likely suffer
-      if a shard contains more than 200M documents. This is the reason of the builtin limit.
+IMPORTANT: The rollover action implicitly always rolls over a data stream or alias if one or more shards contain
+200000000 or more documents. Normally a shard will reach 50GB long before it reaches 200M documents,
+but this isn't the case for space efficient data sets. Search performance will very likely suffer
+if a shard contains more than 200M documents. This is the reason of the builtin limit.
 
 [[ilm-rollover-ex]]
 ==== Example
diff --git a/docs/reference/ilm/index-rollover.asciidoc b/docs/reference/ilm/index-rollover.asciidoc
index a1616807c9ea6..5e6c4b89ba99f 100644
--- a/docs/reference/ilm/index-rollover.asciidoc
+++ b/docs/reference/ilm/index-rollover.asciidoc
@@ -51,3 +51,15 @@ TIP: Rolling over to a new index based on size, document count, or age is prefer
 to time-based rollovers. Rolling over at an arbitrary time often results in
 many small indices, which can have a negative impact on performance and
 resource usage.
+
+IMPORTANT: Empty indices will not be rolled over, even if they have an associated `max_age` that
+would otherwise result in a roll over occurring. A policy can override this behavior, and explicitly
+opt in to rolling over empty indices, by adding a `"min_docs": 0` condition. This can also be
+disabled on a cluster-wide basis by setting `indices.lifecycle.rollover.only_if_has_documents` to
+`false`.
+
+IMPORTANT: The rollover action implicitly always rolls over a data stream or alias if one or more shards contain
+200000000 or more documents. Normally a shard will reach 50GB long before it reaches 200M documents,
+but this isn't the case for space efficient data sets. Search performance will very likely suffer
+if a shard contains more than 200M documents. This is the reason of the builtin limit.
+
diff --git a/docs/reference/ilm/set-up-lifecycle-policy.asciidoc b/docs/reference/ilm/set-up-lifecycle-policy.asciidoc
index cd256bfd966f5..79be6205a8c88 100644
--- a/docs/reference/ilm/set-up-lifecycle-policy.asciidoc
+++ b/docs/reference/ilm/set-up-lifecycle-policy.asciidoc
@@ -68,6 +68,11 @@ PUT _ilm/policy/my_policy
 <2> Delete the index 30 days after rollover
 ====
 
+IMPORTANT: The rollover action implicitly always rolls over a data stream or alias if one or more shards contain
+200000000 or more documents. Normally a shard will reach 25GB long before it reaches 200M documents,
+but this isn't the case for space efficient data sets. Search performance will very likely suffer
+if a shard contains more than 200M documents. This is the reason of the builtin limit.
+
 [discrete]
 [[apply-policy-template]]
 === Apply lifecycle policy with an index template