diff --git a/packages/domain/src/domain/sap/worksheet/tests/test_e2e_elmhurst_sap_score.py b/packages/domain/src/domain/sap/worksheet/tests/test_e2e_elmhurst_sap_score.py
index 9748a66f..6a9782f9 100644
--- a/packages/domain/src/domain/sap/worksheet/tests/test_e2e_elmhurst_sap_score.py
+++ b/packages/domain/src/domain/sap/worksheet/tests/test_e2e_elmhurst_sap_score.py
@@ -74,15 +74,15 @@ _ELMHURST_000474_EXPECTED: Final[ElmhurstExpectedSap] = ElmhurstExpectedSap(
 
 @pytest.mark.xfail(
     reason=(
-        "Table 3c two-profile combi-loss override not yet implemented. PCDB "
-        "18118 (Vaillant ecoTEC sustain 24) lodges separate_dhw_tests=2 → "
-        "spec routes to Table 3c, which uses both Profile M (F1, R1) and "
-        "Profile L (F2, R2) loss factors. Our override gate (`_pcdb_table_"
-        "3b_combi_loss_override`) only accepts separate_dhw_tests==1 (Table "
-        "3b row 1, single-profile) → falls back to Table 3a keep-hot time-"
-        "clock 600 kWh/yr default = 25x overshoot on combi loss → +712 HW "
-        "kWh → continuous SAP +0.83 over PDF (66 vs 65). Re-enable when "
-        "Table 3c lands per the next ticket (see project memory)."
+        "Useful space-heating undershoot. Slice 6+7 landed Table 3c so "
+        "Σ(61) (combi loss) closes — HW kWh = 2119 vs PDF 2116 (Δ<3). "
+        "But useful_space_heating_kwh_per_yr = 9156 vs PDF 10111 = ~9.4% "
+        "undershoot, dominating an unmasked +£cost gain that pushes SAP "
+        "67 vs PDF 65 (Δ=+2, was Δ=+1 under the pre-Table-3c Table 3a "
+        "default which masked the residual with a +575 kWh HW overshoot). "
+        "The residual sits in the §9/§10 cascade (internal gains / mean "
+        "internal temp / HLC / responsiveness), not Appendix J. Tracked "
+        "separately under the cohort residual roadmap."
     ),
     strict=True,
 )