diff --git a/pipeline/download/broadband.py b/pipeline/download/broadband.py
new file mode 100644
index 0000000..eab0e44
--- /dev/null
+++ b/pipeline/download/broadband.py
@@ -0,0 +1,80 @@
+import argparse
+import zipfile
+import httpx
+import polars as pl
+from pathlib import Path
+import tempfile
+from tqdm import tqdm
+
+# Ofcom Connected Nations 2025 - Fixed broadband performance (output area & local authority level)
+# Source: https://www.ofcom.org.uk/phones-and-broadband/coverage-and-speeds/connected-nations-20252/data-downloads-2025
+PERFORMANCE_URL = "https://www.ofcom.org.uk/siteassets/resources/documents/research-and-data/multi-sector/infrastructure-research/connected-nations-2025/202507_fixed_broadband_performance_r01.zip"
+
+
+def download_with_progress(url: str, output_path: Path) -> None:
+    with httpx.stream("GET", url, follow_redirects=True, timeout=120) as response:
+        response.raise_for_status()  # pyright: ignore[reportUnusedCallResult]
+        total = int(response.headers.get("content-length", 0))
+        with (
+            open(output_path, "wb") as f,
+            tqdm(total=total, unit="B", unit_scale=True, desc="Downloading") as pbar,
+        ):
+            for chunk in response.iter_bytes(chunk_size=8192):
+                f.write(chunk)
+                pbar.update(len(chunk))
+
+
+def extract_zip(zip_path: Path, extract_dir: Path) -> None:
+    with zipfile.ZipFile(zip_path, "r") as z:
+        z.extractall(extract_dir)
+
+
+def convert_to_parquet(extract_dir: Path, parquet_path: Path) -> None:
+    # Find CSV files in the extracted directory
+    csv_files = list(extract_dir.rglob("*.csv"))
+    if not csv_files:
+        raise FileNotFoundError(f"No CSV files found in {extract_dir}")
+
+    print(f"Found CSV files: {[f.name for f in csv_files]}")
+
+    # Read and concatenate all CSVs (typically split by geography level)
+    frames = []
+    for csv_file in sorted(csv_files):
+        print(f"Reading {csv_file.name}...")
+        df = pl.read_csv(csv_file, infer_schema_length=10000, encoding="utf8-lossy")
+        print(f"  Shape: {df.shape}, Columns: {df.columns}")
+        frames.append((csv_file.stem, df))
+
+    # Save each CSV as a separate parquet file in the output directory
+    parquet_path.mkdir(parents=True, exist_ok=True)
+    for name, df in frames:
+        out = parquet_path / f"{name}.parquet"
+        df.write_parquet(out, compression="zstd")
+        print(f"Saved {out} ({df.shape[0]} rows)")
+
+
+def main() -> None:
+    parser = argparse.ArgumentParser(
+        description="Download Ofcom broadband performance data"
+    )
+    parser.add_argument(
+        "--output",
+        type=Path,
+        required=True,
+        help="Output directory for parquet files",
+    )
+    args = parser.parse_args()
+
+    with tempfile.TemporaryDirectory() as cache_dir:
+        cache = Path(cache_dir)
+        zip_path = cache / "broadband_performance.zip"
+        extract_dir = cache / "extracted"
+        extract_dir.mkdir()
+
+        download_with_progress(PERFORMANCE_URL, zip_path)
+        extract_zip(zip_path, extract_dir)
+        convert_to_parquet(extract_dir, args.output)
+
+
+if __name__ == "__main__":
+    main()
diff --git a/pipeline/download/naptan.py b/pipeline/download/naptan.py
new file mode 100644
index 0000000..52f823a
--- /dev/null
+++ b/pipeline/download/naptan.py
@@ -0,0 +1,57 @@
+"""Download NaPTAN data and extract railway/metro station POIs."""
+
+import argparse
+import io
+import urllib.request
+from pathlib import Path
+
+import polars as pl
+
+NAPTAN_CSV_URL = "https://naptan.api.dft.gov.uk/v1/access-nodes?dataFormat=csv"
+
+
+def download_naptan(output: Path) -> None:
+    output.parent.mkdir(parents=True, exist_ok=True)
+
+    print(f"Downloading NaPTAN data from {NAPTAN_CSV_URL}")
+    with urllib.request.urlopen(NAPTAN_CSV_URL) as resp:
+        raw = resp.read()
+
+    print(f"Downloaded {len(raw) / (1024 * 1024):.1f} MB")
+
+    df = (
+        pl.read_csv(io.BytesIO(raw), infer_schema_length=0)
+        .with_columns(
+            pl.col("Latitude").cast(pl.Float64, strict=False),
+            pl.col("Longitude").cast(pl.Float64, strict=False),
+        )
+        .drop_nulls(subset=["Latitude", "Longitude"])
+        .select(
+            pl.col("ATCOCode").alias("id"),
+            pl.col("CommonName").alias("name"),
+            pl.col("StopType").alias("category"),
+            pl.col("Latitude").alias("lat"),
+            pl.col("Longitude").alias("lng"),
+        )
+    )
+
+    df.write_parquet(output)
+    size_mb = output.stat().st_size / (1024 * 1024) 
+    print(f"Wrote {output} ({size_mb:.1f} MB, {len(df):,} stations)")
+
+    counts = df.group_by("category").len().sort("len", descending=True)
+    for row in counts.iter_rows(named=True):
+        print(f"  {row['category']}: {row['len']:,}")
+
+
+def main() -> None:
+    parser = argparse.ArgumentParser(description="Download NaPTAN station data")
+    parser.add_argument(
+        "--output", type=Path, required=True, help="Output parquet file path"
+    )
+    args = parser.parse_args()
+    download_naptan(args.output)
+
+
+if __name__ == "__main__":
+    main()
diff --git a/pipeline/download/ofsted.py b/pipeline/download/ofsted.py
new file mode 100644
index 0000000..b783ae9
--- /dev/null
+++ b/pipeline/download/ofsted.py
@@ -0,0 +1,62 @@
+import argparse
+import httpx
+import polars as pl
+from pathlib import Path
+import tempfile
+
+# Management information - state-funded schools - latest inspections (as at 30 Apr 2025)
+# Source: https://www.gov.uk/government/statistical-data-sets/monthly-management-information-ofsteds-school-inspections-outcomes
+URL = "https://assets.publishing.service.gov.uk/media/681cd390275cb67b18d870fc/Management_information_-_state-funded_schools_-_latest_inspections_as_at_30_Apr_2025.csv"
+
+
+def download_file(url: str, output_path: Path) -> None:
+    with httpx.stream("GET", url, follow_redirects=True, timeout=60) as response:
+        response.raise_for_status()  # pyright: ignore[reportUnusedCallResult]
+        total = int(response.headers.get("content-length", 0))
+        downloaded = 0
+        with open(output_path, "wb") as f:
+            for chunk in response.iter_bytes(chunk_size=8192):
+                f.write(chunk)
+                downloaded += len(chunk)
+                if total:
+                    print(
+                        f"\rDownloaded {downloaded / 1024 / 1024:.1f} MB / {total / 1024 / 1024:.1f} MB",
+                        end="",
+                    )
+    print(f"\nSaved to {output_path}")
+
+
+def convert_to_parquet(csv_path: Path, parquet_path: Path) -> None:
+    print("Reading CSV...")
+
+    df = pl.read_csv(
+        csv_path,
+        infer_schema_length=10000,
+        encoding="utf8-lossy",
+        null_values=["NULL", "Not applicable"],
+    )
+
+    print(f"Shape: {df.shape}")
+    print(f"Columns: {df.columns}")
+
+    df.write_parquet(parquet_path, compression="zstd")
+    print(f"Saved to {parquet_path}")
+
+
+def main() -> None:
+    parser = argparse.ArgumentParser(
+        description="Download Ofsted school inspection outcomes data"
+    )
+    parser.add_argument(
+        "--output", type=Path, required=True, help="Output parquet file path"
+    )
+    args = parser.parse_args()
+
+    with tempfile.TemporaryDirectory() as cache_dir:
+        csv_path = Path(cache_dir) / "ofsted_latest_inspections.csv"
+        download_file(URL, csv_path)
+        convert_to_parquet(csv_path, args.output)
+
+
+if __name__ == "__main__":
+    main()
diff --git a/pipeline/transform/transform_poi.py b/pipeline/transform/transform_poi.py
index de92d24..ac6d98f 100644
--- a/pipeline/transform/transform_poi.py
+++ b/pipeline/transform/transform_poi.py
@@ -1,4 +1,5 @@
 import argparse
+import warnings
 from pathlib import Path
 
 import polars as pl