diff --git a/work_trace/wizards/wt_import_timeline_wizard.py b/work_trace/wizards/wt_import_timeline_wizard.py
index 49d9a3f5..01016674 100644
--- a/work_trace/wizards/wt_import_timeline_wizard.py
+++ b/work_trace/wizards/wt_import_timeline_wizard.py
@@ -230,6 +230,19 @@ class WtImportTimelineWizard(models.TransientModel):
                 stop['distance_from_previous'] = 0.0
                 stop['travel_time_from_previous'] = 0.0
 
+        # Post-processing: reclassify likely flights using stop-to-stop speed.
+        # Many flights have no activity segment or timelinePath data — Google just
+        # records a large gap between two distant visits. We catch those here.
+        for stop in stops[1:]:
+            dist = stop.get('distance_from_previous', 0.0)
+            tt   = stop.get('travel_time_from_previous', 0.0)
+            if (dist >= MIN_FLIGHT_DISTANCE_MILES
+                    and tt >= MIN_FLIGHT_DURATION_HOURS
+                    and tt > 0):
+                implied_speed = dist / tt
+                if implied_speed > 150:
+                    stop['travel_mode'] = 'flying'
+
         LocationLog = self.env['wt.location.log']
         existing = set(
             r.arrived_at.strftime('%Y-%m-%d %H:%M:%S')