diff --git a/milli/src/search/new/db_cache.rs b/milli/src/search/new/db_cache.rs
index fb36c0d9f..6193f4c58 100644
--- a/milli/src/search/new/db_cache.rs
+++ b/milli/src/search/new/db_cache.rs
@@ -138,7 +138,7 @@ impl<'ctx> SearchContext<'ctx> {
     }
 
     /// Retrieve or insert the given value in the `word_prefix_docids` database.
-    pub fn get_db_word_prefix_docids(
+    fn get_db_word_prefix_docids(
         &mut self,
         prefix: Interned<String>,
     ) -> Result<Option<RoaringBitmap>> {
diff --git a/milli/src/search/new/logger/visual.rs b/milli/src/search/new/logger/visual.rs
index 068b5ad68..72e33f339 100644
--- a/milli/src/search/new/logger/visual.rs
+++ b/milli/src/search/new/logger/visual.rs
@@ -434,7 +434,7 @@ fill: \"#B6E2D3\"
                     writeln!(file, "{}: phrase", p.description(ctx))?;
                 }
                 if let Some(w) = term_subset.use_prefix_db(ctx) {
-                    let w = ctx.word_interner.get(w);
+                    let w = ctx.word_interner.get(w.interned());
                     writeln!(file, "{w}: prefix db")?;
                 }
 
diff --git a/milli/src/search/new/query_term/mod.rs b/milli/src/search/new/query_term/mod.rs
index 0a0d1a7eb..d8c2bb0c7 100644
--- a/milli/src/search/new/query_term/mod.rs
+++ b/milli/src/search/new/query_term/mod.rs
@@ -159,12 +159,12 @@ impl QueryTermSubset {
         self.two_typo_subset.intersect(&other.two_typo_subset);
     }
 
-    pub fn use_prefix_db(&self, ctx: &SearchContext) -> Option<Interned<String>> {
+    pub fn use_prefix_db(&self, ctx: &SearchContext) -> Option<Word> {
         let original = ctx.term_interner.get(self.original);
         let Some(use_prefix_db) = original.zero_typo.use_prefix_db else {
             return None
         };
-        match &self.zero_typo_subset {
+        let word = match &self.zero_typo_subset {
             NTypoTermSubset::All => Some(use_prefix_db),
             NTypoTermSubset::Subset { words, phrases: _ } => {
                 // TODO: use a subset of prefix words instead
@@ -175,7 +175,14 @@ impl QueryTermSubset {
                 }
             }
             NTypoTermSubset::Nothing => None,
-        }
+        };
+        word.map(|word| {
+            if original.ngram_words.is_some() {
+                Word::Derived(word)
+            } else {
+                Word::Original(word)
+            }
+        })
     }
     pub fn all_single_words_except_prefix_db(
         &self,
diff --git a/milli/src/search/new/ranking_rule_graph/proximity/compute_docids.rs b/milli/src/search/new/ranking_rule_graph/proximity/compute_docids.rs
index 760c7272c..3e75f948e 100644
--- a/milli/src/search/new/ranking_rule_graph/proximity/compute_docids.rs
+++ b/milli/src/search/new/ranking_rule_graph/proximity/compute_docids.rs
@@ -55,7 +55,7 @@ pub fn compute_docids(
             compute_prefix_edges(
                 ctx,
                 left_word.interned(),
-                right_prefix,
+                right_prefix.interned(),
                 left_phrase,
                 forward_proximity,
                 backward_proximity,
diff --git a/milli/src/search/new/resolve_query_graph.rs b/milli/src/search/new/resolve_query_graph.rs
index c78f0c5ee..f4938ca12 100644
--- a/milli/src/search/new/resolve_query_graph.rs
+++ b/milli/src/search/new/resolve_query_graph.rs
@@ -44,7 +44,7 @@ pub fn compute_query_term_subset_docids(
     }
 
     if let Some(prefix) = term.use_prefix_db(ctx) {
-        if let Some(prefix_docids) = ctx.get_db_word_prefix_docids(prefix)? {
+        if let Some(prefix_docids) = ctx.word_prefix_docids(prefix)? {
             docids |= prefix_docids;
         }
     }