diff --git a/web/core/views.py b/web/core/views.py
index c91aef3..508fa12 100644
--- a/web/core/views.py
+++ b/web/core/views.py
@@ -332,10 +332,35 @@ def import_wizard(request):
         form = ImportForm(request.POST, request.FILES)
         if form.is_valid():
             try:
+                raw = form.cleaned_data["file"].read()
+
+                # --- header check (tolerant: utf-8-sig removes BOM, strip spaces, case-insensitive) ---
+                import io, csv as _csv
+                sio = io.StringIO(raw.decode("utf-8-sig", errors="replace"))
+                rdr = _csv.reader(sio)
+                first_row = next(rdr, [])
+                norm = [c.strip().lower() for c in first_row]
+                expected_norm = [c.lower() for c in EXPECTED_HEADERS]
+
+                # If the first row doesn’t look like our header, allow import to proceed,
+                # but show a gentle warning in the result page.
+                header_ok = (norm == expected_norm)
+
+                # hand off to your existing robust importer
                 report = import_csv_bytes(
-                    form.cleaned_data["file"].read(),
+                    raw,
                     dry_run=form.cleaned_data["dry_run"],
                 )
+
+                # attach header check info (if your template shows report dicts)
+                report = report or {}
+                report["header_ok"] = header_ok
+                if not header_ok:
+                    messages.warning(
+                        request,
+                        "The first row does not match the expected header; assuming the file has no header."
+                    )
+
                 return render(
                     request,
                     "import_result.html",