diff --git a/README.md b/README.md
index 03a5d963..2bb6d1bb 100644
--- a/README.md
+++ b/README.md
@@ -19,8 +19,9 @@
 
     ```java
     import com.google.cloud.pubsublite.kafka.ProducerSettings;
-    import org.apache.kafka.clients.producer.*;
     import com.google.cloud.pubsublite.*;
+
+    import org.apache.kafka.clients.producer.*;
     
     ...
     
@@ -51,17 +52,22 @@
 1. Read some messages using:
 
     ```java
+    import com.google.cloud.pubsublite.cloudpubsub.FlowControlSettings;
     import com.google.cloud.pubsublite.kafka.ConsumerSettings;
-    import org.apache.kafka.clients.consumer.*;
     import com.google.cloud.pubsublite.*;
-    import com.google.cloud.pubsublite.cloudpubsub.FlowControlSettings;
-    
+ 
+    import org.apache.kafka.clients.consumer.*;
     ...
     
     private final static String ZONE = "us-central1-b";
     private final static Long PROJECT_NUM = 123L;
     
     ...
+    TopicPath topic = TopicPath.newBuilder()
+       .setLocation(CloudZone.parse(ZONE))
+       .setProject(ProjectNumber.of(PROJECT_NUM))
+       .setName(TopicName.of("my-topic"))
+       .build();
    
     SubscriptionPath subscription = SubscriptionPath.newBuilder()
         .setLocation(CloudZone.parse(ZONE))
@@ -75,11 +81,13 @@
             .setBytesOutstanding(10_000_000)  // 10 MB
             .setMessagesOutstanding(Long.MAX_VALUE)
             .build())
-        .setAutocommit(true);
+        .setAutocommit(true)
+        .build();
    
     try (Consumer<byte[], byte[]> consumer = settings.instantiate()) {
+       consumer.subscribe(Arrays.asList(topic.toString()));
        while (true) {
-         ConsumerRecords<byte[], byte[]> records = consumer.poll(Long.MAX_VALUE);
+         ConsumerRecords<byte[], byte[]> records = consumer.poll(Duration.ofSeconds(30));
          for (ConsumerRecord<byte[], byte[]> record : records) {
            System.out.println(record.offset() + “: ” + record.value());
          }