diff --git a/README.md b/README.md
index 172c0a5..613dec1 100644
--- a/README.md
+++ b/README.md
@@ -124,8 +124,8 @@ data_destinations:
 
 ### Limitations
 
-1) The emulator works well in development and test environments, but does not have the [strict](https://ambar.cloud/blog/provably-correct-data-streaming-our-white-paper) delivery and durability guarantees required of production environments.
-2) The emulator consumes more resources on databases and web servers, because it does not implement features such as change data capture and [adaptive load](https://ambar.cloud/blog/optimal-consumption-with-adaptive-load).
+1) The emulator works well in development and test environments, but does not have the [strict](https://ambar.cloud/blog/provably-correct-data-streaming-our-white-paper) delivery and durability guarantees Ambar gives you in production environments.
+2) The emulator consumes more resources than Ambar does on your databases and web servers, because it does not implement features such as change data capture and [adaptive load](https://ambar.cloud/blog/optimal-consumption-with-adaptive-load).
 3) The emulator can handle thousands of messages per second but does not scale horizontally, because the emulator can only run in a single machine.
 4) The emulator doesn't provide filtering (yet - we will eventually add this).