diff --git a/docs/building-spark.md b/docs/building-spark.md
index 391ad223ccaa..5453a218205e 100644
--- a/docs/building-spark.md
+++ b/docs/building-spark.md
@@ -83,11 +83,14 @@ Example:
 
 To enable Hive integration for Spark SQL along with its JDBC server and CLI,
 add the `-Phive` and `Phive-thriftserver` profiles to your existing build options.
-By default Spark will build with Hive 1.2.1 bindings.
+By default, Spark will use Hive 1.2.1 with the `hadoop-2.7` profile, and Hive 2.3.5 with the `hadoop-3.2` profile.
 
     # With Hive 1.2.1 support
     ./build/mvn -Pyarn -Phive -Phive-thriftserver -DskipTests clean package
 
+    # With Hive 2.3.5 support
+    ./build/mvn -Pyarn -Phive -Phive-thriftserver -Phadoop-3.2 -DskipTests clean package
+
 ## Packaging without Hadoop Dependencies for YARN
 
 The assembly directory produced by `mvn package` will, by default, include all of Spark's