diff --git a/distributed/tests/test_spans.py b/distributed/tests/test_spans.py
index cf01e11ec7..c345a7dfcd 100644
--- a/distributed/tests/test_spans.py
+++ b/distributed/tests/test_spans.py
@@ -26,25 +26,22 @@
 async def test_spans(c, s, a):
     x = delayed(inc)(1)  # Default span
 
-    @span("p2")
-    def f(i):
-        return i * 2
-
     with span("my workflow") as mywf_id:
         with span("p1") as p1_id:
             y = x + 1
-        z = f(y)
+        with span("p2") as p2_id:
+            z = y * 2
 
     zp = c.persist(z)
     assert await c.compute(zp) == 6
 
     ext = s.extensions["spans"]
 
-    p2_id = s.tasks[z.key].group.span_id
     assert mywf_id
     assert p1_id
     assert p2_id
     assert s.tasks[y.key].group.span_id == p1_id
+    assert s.tasks[z.key].group.span_id == p2_id
     assert mywf_id != p1_id != p2_id
 
     expect_annotations = {
@@ -135,15 +132,12 @@ async def test_repeat_span(c, s, a, b):
     """Opening and closing the same span will result in multiple spans with different
     ids and same name
     """
-
-    @span("foo")
-    def f(x, key):
-        return c.submit(inc, x, key=key)
-
     with span("foo"):
         x = c.submit(inc, 1, key="x")
-    y = f(x, key="y")
-    z = f(y, key="z")
+    with span("foo"):
+        y = c.submit(inc, x, key="y")
+    with span("foo"):
+        z = c.submit(inc, y, key="z")
     assert await z == 4
 
     sbn = s.extensions["spans"].spans_search_by_name["foo",]
diff --git a/docs/source/spans.rst b/docs/source/spans.rst
index dcb5339cb8..235cc4fb91 100644
--- a/docs/source/spans.rst
+++ b/docs/source/spans.rst
@@ -32,27 +32,19 @@ For example:
     dask.config.set({"optimization.fuse.active": False})
     client = Client()
 
-    @span("ML preprocessing")
-    def preprocess(df):
-        ...
-        return df
-
-    @span("Model training")
-    def train(df):
-        ...
-        return model
-
     with span("Alice's workflow"):
         with span("data load"):
             df = dd.read_parquet(...)
 
-        df = preprocess(df)
-        model = train(df)
+        with span("ML preprocessing"):
+            df = preprocess(df)
+
+        with span("Model training"):
+            model = train(df)
 
     model = model.compute()
 
-In the above example, note how the :func:`span` context manager is used as a decorator
-too, and how it can be nested.
+Note how the :func:`span` context manager can be nested.
 The example will create the following spans on the scheduler:
 
 - ``("Alice's workflow", )``