diff --git a/CHANGELOG.md b/CHANGELOG.md
index d1df44a4a63..caf6747dc71 100644
--- a/CHANGELOG.md
+++ b/CHANGELOG.md
@@ -6,6 +6,7 @@
 * [ENHANCEMENT] Improve readability of cpu and memory metrics on operational dashboard [#764](https://github.com/grafana/tempo/pull/764) (@bboreham)
 * [ENHANCEMENT] Add `azure_request_duration_seconds` metric. [#767](https://github.com/grafana/tempo/pull/767) (@JosephWoodward)
 * [ENHANCEMENT] Add `s3_request_duration_seconds` metric. [#776](https://github.com/grafana/tempo/pull/776) (@JosephWoodward)
+* [ENHANCEMENT] Add `tempo_ingester_flush_size_bytes` metric. [#777](https://github.com/grafana/tempo/pull/777) (@bboreham)
 
 ## v1.0.1
 
diff --git a/modules/ingester/flush.go b/modules/ingester/flush.go
index bdbd291eb36..bf05a9f7ca1 100644
--- a/modules/ingester/flush.go
+++ b/modules/ingester/flush.go
@@ -36,6 +36,12 @@ var (
 		Help:      "Records the amount of time to flush a complete block.",
 		Buckets:   prometheus.ExponentialBuckets(1, 2, 10),
 	})
+	metricFlushSize = promauto.NewHistogram(prometheus.HistogramOpts{
+		Namespace: "tempo",
+		Name:      "ingester_flush_size_bytes",
+		Help:      "Size in bytes of blocks flushed.",
+		Buckets:   prometheus.ExponentialBuckets(1024*1024, 2, 10), // from 1MB up to 1GB
+	})
 )
 
 const (
@@ -275,6 +281,7 @@ func (i *Ingester) handleFlush(userID string, blockID uuid.UUID) (retry bool, er
 		start := time.Now()
 		err = i.store.WriteBlock(ctx, block)
 		metricFlushDuration.Observe(time.Since(start).Seconds())
+		metricFlushSize.Observe(float64(block.BlockMeta().Size))
 		if err != nil {
 			return true, err
 		}