diff --git a/causal_testing/data_collection/data_collector.py b/causal_testing/data_collection/data_collector.py
index 9a0a152f..718ed222 100644
--- a/causal_testing/data_collection/data_collector.py
+++ b/causal_testing/data_collection/data_collector.py
@@ -48,6 +48,10 @@ def filter_valid_data(self, data: pd.DataFrame, check_pos: bool = True) -> pd.Da
                 f"Missing columns: missing data for variables {missing_variables}. Should they be marked as hidden?"
             )
 
+        # Quick out if we don't have any constraints
+        if len(self.scenario.constraints) == 0:
+            return data
+
         # For each row, does it satisfy the constraints?
         solver = z3.Solver()
         for c in self.scenario.constraints:
@@ -56,6 +60,7 @@ def filter_valid_data(self, data: pd.DataFrame, check_pos: bool = True) -> pd.Da
         unsat_core = None
         for _, row in data.iterrows():
             solver.push()
+            # Check that the row does not violate any scenario constraints
             # Need to explicitly cast variables to their specified type. Z3 will not take e.g. np.int64 to be an int.
             model = [
                 self.scenario.variables[var].z3