diff --git a/doc/source/tutorial/coordinator.rst b/doc/source/tutorial/coordinator.rst
index 38d925ec..faf9ab2b 100644
--- a/doc/source/tutorial/coordinator.rst
+++ b/doc/source/tutorial/coordinator.rst
@@ -26,16 +26,16 @@ identifying the running program.
 After the coordinator is created, it can be used to use the various features
 provided.
 
-In order to keep the connection to the coordination server active, you must
-call regularly the :meth:`~tooz.coordination.CoordinationDriver.heartbeat`
-method. This will ensure that the coordinator is not considered dead by
-other program participating in the coordination.
+In order to keep the connection to the coordination server active, the method
+:meth:`~tooz.coordination.CoordinationDriver.heartbeat` method must be called
+regularly. This will ensure that the coordinator is not considered dead by
+other program participating in the coordination. Unless you want to call it
+manually, you can use tooz builtin heartbeat manager by passing the
+`start_heart` argument.
 
 .. literalinclude:: ../../../examples/coordinator_heartbeat.py
    :language: python
 
-We use a pretty simple mechanism in this example to send a heartbeat every
-once in a while, but depending on your application, you may want to send the
 heartbeat at different moment or intervals.
 
 Note that certain drivers, such as `memcached` are heavily based on timeout,
diff --git a/examples/coordinator_heartbeat.py b/examples/coordinator_heartbeat.py
index ddbea137..2a6534a6 100644
--- a/examples/coordinator_heartbeat.py
+++ b/examples/coordinator_heartbeat.py
@@ -1,15 +1,5 @@
-import time
-
 from tooz import coordination
 
-ALIVE_TIME = 5
-
 coordinator = coordination.get_coordinator('zake://', b'host-1')
-coordinator.start()
-
-start = time.time()
-while time.time() - start < ALIVE_TIME:
-    coordinator.heartbeat()
-    time.sleep(0.1)
-
+coordinator.start(start_heart=True)
 coordinator.stop()