diff --git a/README.md b/README.md
index c476ea9..31b8c38 100644
--- a/README.md
+++ b/README.md
@@ -419,3 +419,21 @@ wondering what the tool assumed.
   Those are preserved but will not resolve without the linked file.
 - Description joins can drop a space where Candy's print engine split a word
   (`Fixed-ChargeItems`). Cosmetic.
+
+
+### Where the placement check stands
+
+The check is sound — silent on Blueberry, fires on Central Towers — but it has
+exposed that **the alignment drift is not fixed.** Fifteen lines on Central
+Towers are still placed on the wrong row, fourteen of them carrying a value.
+
+A gap-penalty sweep from -0.20 to -1.50 moves it not at all, so this is not the
+aligner chasing scores. Both documents contain **four separate near-identical
+depth-band groups** in `1200DB` alone, and a single global alignment
+distributes them wrongly. Closing that needs a different strategy — anchor on
+the confidently matched lines and interpolate within each block, rather than
+solving the whole document as one problem.
+
+Until then **Central Towers is blocked and will not produce a file.** That is
+the correct behaviour: the app now knows it is wrong and refuses, rather than
+sending it.
diff --git a/boq_engine/template.py b/boq_engine/template.py
index c690a97..c6f4316 100644
--- a/boq_engine/template.py
+++ b/boq_engine/template.py
@@ -371,6 +371,12 @@ class Slot:
 # a description opening with a lettered sub-item label (a), b), c)) starts a
 # NEW line rather than continuing the one above
 LETTER_ONLY = re.compile(r"^\(?[a-z]\)\s", re.I)
+# a description that OPENS with its own reference - ".01 ", "1.2 ", "c) " -
+# starts a NEW line. Only bare continuation text may be folded into the row
+# above. Excluding only "a)" left ".01 0.5 - 1.0m Deep" being swallowed by the
+# sub-group heading above it, which is how a whole depth band went missing from
+# the pool and its rate drifted onto the next group's rows.
+STARTS_LINE = re.compile(r"^\(?[a-z]\)\s|^\.?\d+(?:\.\d+)*[\s.]", re.I)
 # ".01", "1.2", "a)" - a description that OPENS with a reference starts a new
 # line; only bare continuation text may be folded into the one above
 STARTS_LINE = re.compile(r"^\(?[a-z]\)\s|^\.?\d+(?:\.\d+)*[\s.]", re.I)
@@ -426,7 +432,7 @@ def read_slots(ws, lay: Layout) -> list[Slot]:
         # A description opening "a)" / "b)" / "c)" is a NEW line and must never
         # be absorbed, or sub-items get folded into the heading above them.
         if (out and priced and not raw and desc and out[-1]._bare
-                and not LETTER_ONLY.match(desc)):
+                and not STARTS_LINE.match(desc)):
             prev = out[-1]
             prev.desc = f"{prev.desc} {desc}".strip() if prev.desc else desc
             prev.row = r
diff --git a/boq_engine/verify.py b/boq_engine/verify.py
index 0e7adaf..cd31320 100644
--- a/boq_engine/verify.py
+++ b/boq_engine/verify.py
@@ -180,6 +180,15 @@ def _placement_report(result, out) -> list:
 
         row_desc = _norm_txt(_row_desc(out[sheet], row, lay))
         want = _norm_txt(want)
+        # A cell holding a formula is not the row's identity - these templates
+        # pull a description straight out of a Measurement sheet
+        # ("=Measurement!A18"). The row may be perfectly correct; we simply
+        # cannot read it, so say nothing rather than cry mismatch.
+        # any formula inside the text means this row pulls its description from
+        # elsewhere (these templates use "=Measurement!A18"); it may be the
+        # right row and simply not readable here
+        if "=" in row_desc:
+            continue
         if not row_desc:
             bad.append((sheet, row, want[:44], "that row carries no text at all"))
             continue