diff --git a/prdoc/pr_10802.prdoc b/prdoc/pr_10802.prdoc
new file mode 100644
index 0000000000000..1e79b476a404e
--- /dev/null
+++ b/prdoc/pr_10802.prdoc
@@ -0,0 +1,13 @@
+title: 'benchmarking: fix timing leak from bulk setup operations'
+doc:
+- audience: Runtime Dev
+  description: |-
+    Fixes timing leaks in benchmarks with large setup operations (e.g., clearing
+    27k staking entries). After bulk deletions are committed, the first new allocation
+    can trigger memory allocator overhead that leaks into benchmark timing.
+
+    The fix adds a memory allocator warmup step in `commit_db()` that performs a
+    dummy write/clear cycle to absorb this overhead before timing starts.
+crates:
+- name: frame-benchmarking
+  bump: patch
diff --git a/substrate/frame/benchmarking/src/utils.rs b/substrate/frame/benchmarking/src/utils.rs
index 0f0cc3558a1ff..c41c6a99854cd 100644
--- a/substrate/frame/benchmarking/src/utils.rs
+++ b/substrate/frame/benchmarking/src/utils.rs
@@ -289,7 +289,17 @@ pub trait Benchmarking {
 
 	/// Commit pending storage changes to the trie database and clear the database cache.
 	fn commit_db(&mut self) {
-		self.commit()
+		self.commit();
+
+		// Warmup the memory allocator after bulk deallocation.
+		// After draining the overlay with many entries, the first new allocation can trigger memory
+		// defragmentation.
+		const WARMUP_KEY: &[u8] = b":benchmark_warmup:";
+		self.place_storage(WARMUP_KEY.to_vec(), Some(vec![0u8; 32]));
+		self.place_storage(WARMUP_KEY.to_vec(), None);
+
+		// Reset tracking so warmup operations don't appear in benchmark results.
+		self.reset_read_write_count();
 	}
 
 	/// Get the read/write count.