diff --git a/capture-server/tests/common.rs b/capture-server/tests/common.rs
index fa86881..2959d83 100644
--- a/capture-server/tests/common.rs
+++ b/capture-server/tests/common.rs
@@ -28,6 +28,7 @@ pub static DEFAULT_CONFIG: Lazy<Config> = Lazy::new(|| Config {
     print_sink: false,
     address: SocketAddr::from_str("127.0.0.1:0").unwrap(),
     redis_url: "redis://localhost:6379/".to_string(),
+    overflow_enabled: true,
     overflow_burst_limit: NonZeroU32::new(5).unwrap(),
     overflow_per_second_limit: NonZeroU32::new(10).unwrap(),
     overflow_forced_keys: None,
diff --git a/capture/src/config.rs b/capture/src/config.rs
index 0c6ab1c..7f147c7 100644
--- a/capture/src/config.rs
+++ b/capture/src/config.rs
@@ -13,6 +13,9 @@ pub struct Config {
     pub redis_url: String,
     pub otel_url: Option<String>,
 
+    #[envconfig(default = "true")]
+    pub overflow_enabled: bool,
+
     #[envconfig(default = "100")]
     pub overflow_per_second_limit: NonZeroU32,
 
diff --git a/capture/src/server.rs b/capture/src/server.rs
index 22a1f3b..1cfbae0 100644
--- a/capture/src/server.rs
+++ b/capture/src/server.rs
@@ -46,24 +46,31 @@ where
             .register("rdkafka".to_string(), Duration::seconds(30))
             .await;
 
-        let partition = OverflowLimiter::new(
-            config.overflow_per_second_limit,
-            config.overflow_burst_limit,
-            config.overflow_forced_keys,
-        );
-        if config.export_prometheus {
-            let partition = partition.clone();
-            tokio::spawn(async move {
-                partition.report_metrics().await;
-            });
-        }
-        {
-            // Ensure that the rate limiter state does not grow unbounded
-            let partition = partition.clone();
-            tokio::spawn(async move {
-                partition.clean_state().await;
-            });
-        }
+        let partition = match config.overflow_enabled {
+            true => {
+                let partition = OverflowLimiter::new(
+                    config.overflow_per_second_limit,
+                    config.overflow_burst_limit,
+                    config.overflow_forced_keys,
+                );
+                if config.export_prometheus {
+                    let partition = partition.clone();
+                    tokio::spawn(async move {
+                        partition.report_metrics().await;
+                    });
+                }
+                {
+                    // Ensure that the rate limiter state does not grow unbounded
+                    let partition = partition.clone();
+                    tokio::spawn(async move {
+                        partition.clean_state().await;
+                    });
+                }
+                Some(partition)
+            }
+            false => None,
+        };
+
         let sink = KafkaSink::new(config.kafka, sink_liveness, partition)
             .expect("failed to start Kafka sink");
 
diff --git a/capture/src/sinks/kafka.rs b/capture/src/sinks/kafka.rs
index 4a2bd94..262b719 100644
--- a/capture/src/sinks/kafka.rs
+++ b/capture/src/sinks/kafka.rs
@@ -82,14 +82,14 @@ impl rdkafka::ClientContext for KafkaContext {
 pub struct KafkaSink {
     producer: FutureProducer<KafkaContext>,
     topic: String,
-    partition: OverflowLimiter,
+    partition: Option<OverflowLimiter>,
 }
 
 impl KafkaSink {
     pub fn new(
         config: KafkaConfig,
         liveness: HealthHandle,
-        partition: OverflowLimiter,
+        partition: Option<OverflowLimiter>,
     ) -> anyhow::Result<KafkaSink> {
         info!("connecting to Kafka brokers at {}...", config.kafka_hosts);
 
@@ -206,7 +206,10 @@ impl KafkaSink {
 impl Event for KafkaSink {
     #[instrument(skip_all)]
     async fn send(&self, event: ProcessedEvent) -> Result<(), CaptureError> {
-        let limited = self.partition.is_limited(&event.key());
+        let limited = match &self.partition {
+            Some(limiter) => limiter.is_limited(&event.key()),
+            None => false,
+        };
         let ack =
             Self::kafka_send(self.producer.clone(), self.topic.clone(), event, limited).await?;
         histogram!("capture_event_batch_size").record(1.0);
@@ -222,7 +225,10 @@ impl Event for KafkaSink {
         for event in events {
             let producer = self.producer.clone();
             let topic = self.topic.clone();
-            let limited = self.partition.is_limited(&event.key());
+            let limited = match &self.partition {
+                Some(limiter) => limiter.is_limited(&event.key()),
+                None => false,
+            };
 
             // We await kafka_send to get events in the producer queue sequentially
             let ack = Self::kafka_send(producer, topic, event, limited).await?;
@@ -295,7 +301,7 @@ mod tests {
             kafka_topic: "events_plugin_ingestion".to_string(),
             kafka_tls: false,
         };
-        let sink = KafkaSink::new(config, handle, limiter).expect("failed to create sink");
+        let sink = KafkaSink::new(config, handle, Some(limiter)).expect("failed to create sink");
         (cluster, sink)
     }