diff --git a/apps/sim/app/workspace/[workspaceId]/w/components/sidebar/components/search-modal/utils.test.ts b/apps/sim/app/workspace/[workspaceId]/w/components/sidebar/components/search-modal/utils.test.ts
index a82b79fd5f9..9705e158620 100644
--- a/apps/sim/app/workspace/[workspaceId]/w/components/sidebar/components/search-modal/utils.test.ts
+++ b/apps/sim/app/workspace/[workspaceId]/w/components/sidebar/components/search-modal/utils.test.ts
@@ -315,3 +315,34 @@ describe('filterAndCap', () => {
     expect(capped).toEqual(filterAndSort(items, id, 'item').slice(0, MAX_RESULTS_PER_GROUP))
   })
 })
+
+describe('filterAndSort — name vs. secondary text', () => {
+  const rows = [
+    { name: 'Leads', folder: 'Sales / EMEA' },
+    { name: 'Contacts', folder: 'Sales / AMER' },
+    { name: 'EMEA', folder: 'Archive' },
+  ]
+  const run = (search: string) =>
+    filterAndSort(
+      rows,
+      (r) => r.name,
+      search,
+      (r) => r.folder
+    ).map((r) => r.name)
+
+  it('matches a query spanning the name and the secondary text', () => {
+    expect(run('leads emea')).toEqual(['Leads'])
+  })
+
+  it('ranks a name match above a secondary-text match', () => {
+    expect(run('emea')).toEqual(['EMEA', 'Leads'])
+  })
+
+  it('still matches on secondary text alone', () => {
+    expect(run('amer')).toEqual(['Contacts'])
+  })
+
+  it('drops rows that match neither field nor the two joined', () => {
+    expect(run('zzz')).toEqual([])
+  })
+})
diff --git a/apps/sim/app/workspace/[workspaceId]/w/components/sidebar/components/search-modal/utils.ts b/apps/sim/app/workspace/[workspaceId]/w/components/sidebar/components/search-modal/utils.ts
index 72265a6740e..153dac2512a 100644
--- a/apps/sim/app/workspace/[workspaceId]/w/components/sidebar/components/search-modal/utils.ts
+++ b/apps/sim/app/workspace/[workspaceId]/w/components/sidebar/components/search-modal/utils.ts
@@ -244,9 +244,16 @@ const NAME_MATCH_TIER = 1_000_000
 
 /**
  * Ranks an item by its name first, falling back to secondary text (ids, aliases,
- * option labels) only when the name doesn't match — a name match always wins, so
- * an exact name hit isn't diluted by a long secondary string ("Agent" beats
- * "Pi Coding Agent" for the query "agent").
+ * option labels, folder paths) only when the name doesn't match — a name match
+ * always wins, so an exact name hit isn't diluted by a long secondary string
+ * ("Agent" beats "Pi Coding Agent" for the query "agent").
+ *
+ * When neither field matches alone the two are matched joined, so a query that
+ * spans both ("leads emea" for a `Leads` table in `EMEA`) still finds the row.
+ * That is last on purpose: matching the joined text would otherwise let a
+ * secondary hit masquerade as a name hit. Positions from the joined match do not
+ * index into either field, which is safe only because `filterAndSort` scores
+ * with them and discards them — never pass them to a highlighter.
  */
 function scoreItem(name: string, extra: string | undefined, search: string): FuzzyResult {
   const byName = fuzzyMatch(name, search)
@@ -255,7 +262,8 @@ function scoreItem(name: string, extra: string | undefined, search: string): Fuz
     return { matched: true, score: byName.score + NAME_MATCH_TIER, positions: byName.positions }
   }
   const byExtra = fuzzyMatch(extra, search)
-  return byExtra.matched ? byExtra : NO_MATCH
+  if (byExtra.matched) return byExtra
+  return fuzzyMatch(`${name} ${extra}`, search)
 }
 
 /**