diff --git a/milli/src/search/new/ranking_rule_graph/position/mod.rs b/milli/src/search/new/ranking_rule_graph/position/mod.rs
index ba4743d99..c3d33174b 100644
--- a/milli/src/search/new/ranking_rule_graph/position/mod.rs
+++ b/milli/src/search/new/ranking_rule_graph/position/mod.rs
@@ -91,13 +91,15 @@ impl RankingRuleGraphTrait for PositionGraph {
             };
             positions_for_costs.entry(cost).or_default().push(position);
         }
+        println!(
+            "positions for cost {} : {positions_for_costs:?}",
+            term.term_subset.description(ctx)
+        );
         let mut edges = vec![];
 
         for (cost, positions) in positions_for_costs {
             // TODO: We can improve performances and relevancy by storing
             //       the term subsets associated to each position fetched.
-            //
-            // TODO: group conditions by their cost
             edges.push((
                 cost,
                 conditions_interner.insert(PositionCondition {
diff --git a/milli/tests/assets/test_set.ndjson b/milli/tests/assets/test_set.ndjson
index 4c83cbe14..1a8a23c3e 100644
--- a/milli/tests/assets/test_set.ndjson
+++ b/milli/tests/assets/test_set.ndjson
@@ -47,7 +47,7 @@
     "word_rank": 0,
     "typo_rank": 1,
     "proximity_rank": 10,
-    "attribute_rank": 111,
+    "attribute_rank": 108,
     "exact_rank": 6,
     "asc_desc_rank": 2,
     "sort_by_rank": 0,