diff --git a/poetry.lock b/poetry.lock
index 506c43f..50e62e1 100644
--- a/poetry.lock
+++ b/poetry.lock
@@ -1,4 +1,4 @@
-# This file is automatically @generated by Poetry 1.8.2 and should not be changed by hand.
+# This file is automatically @generated by Poetry 1.8.1 and should not be changed by hand.
 
 [[package]]
 name = "aiohappyeyeballs"
@@ -3172,6 +3172,20 @@ files = [
 [package.dependencies]
 six = ">=1.5"
 
+[[package]]
+name = "python-dotenv"
+version = "1.0.1"
+description = "Read key-value pairs from a .env file and set them as environment variables"
+optional = false
+python-versions = ">=3.8"
+files = [
+    {file = "python-dotenv-1.0.1.tar.gz", hash = "sha256:e324ee90a023d808f1959c46bcbc04446a10ced277783dc6ee09987c37ec10ca"},
+    {file = "python_dotenv-1.0.1-py3-none-any.whl", hash = "sha256:f7b63ef50f1b690dddf550d03497b66d609393b40b564ed0d674909a68ebf16a"},
+]
+
+[package.extras]
+cli = ["click (>=5.0)"]
+
 [[package]]
 name = "python-json-logger"
 version = "2.0.7"
@@ -4353,4 +4367,4 @@ propcache = ">=0.2.0"
 [metadata]
 lock-version = "2.0"
 python-versions = "^3.11"
-content-hash = "b221f6df6159ddf1a012515162d9ee5d217cf8e4902cdce7e7277b378a4b8f5d"
+content-hash = "236f13732306ac65d5f1fba74680b4968888a19340b7335c560a723b36e3b724"
diff --git a/pyproject.toml b/pyproject.toml
index 3da851f..aa28e79 100644
--- a/pyproject.toml
+++ b/pyproject.toml
@@ -15,6 +15,7 @@ seaborn = "^0.13.2"
 geopandas = "^1.0.1"
 geodatasets = "^2024.8.0"
 streamlit = "^1.39.0"
+python-dotenv = "^1.0.1"
 
 [tool.poetry.group.dev.dependencies]
 pre-commit = "^3.8.0"
diff --git a/src/config.py b/src/config.py
index f46d9c7..e5ece3e 100644
--- a/src/config.py
+++ b/src/config.py
@@ -1,7 +1,15 @@
+"""Override any of these settings by adding environment variables with the same name."""
+
 from pathlib import Path
+import os
+from dotenv import load_dotenv, find_dotenv
+
+load_dotenv(find_dotenv())
 
-REPO_NAME = "ClimatePolicyRadar/all-document-text-data"
+REPO_NAME = os.getenv("REPO_NAME", "ClimatePolicyRadar/all-document-text-data")
 REPO_URL = f"https://huggingface.co/datasets/{REPO_NAME}"
-CACHE_DIR = Path(__file__).parent / "../cache"
+CACHE_DIR = os.getenv("CACHE_DIR", Path(__file__).parent / "../cache")
 
-REVISION = "main"  # Use this to set a commit hash. Recommended!
+DATA_REVISION = os.getenv(
+    "DATA_REVISION", "main"
+)  # Use this to set a commit hash. Recommended!
diff --git a/src/notebooks/1-loading-data.ipynb b/src/notebooks/1-loading-data.ipynb
index 4ea902c..cb114f6 100644
--- a/src/notebooks/1-loading-data.ipynb
+++ b/src/notebooks/1-loading-data.ipynb
@@ -65,14 +65,14 @@
    "source": [
     "REPO_NAME = \"ClimatePolicyRadar/all-document-text-data\"\n",
     "REPO_URL = f\"https://huggingface.co/datasets/{REPO_NAME}\"\n",
-    "CACHE_DIR = \"../../cache\"\n",
+    "DATA_CACHE_DIR = \"../../cache\"\n",
     "\n",
     "REVISION = \"main\"  # Use this to set a commit hash. Recommended!\n",
     "\n",
     "snapshot_download(\n",
     "    repo_id=REPO_NAME,\n",
     "    repo_type=\"dataset\",\n",
-    "    local_dir=CACHE_DIR,\n",
+    "    local_dir=DATA_CACHE_DIR,\n",
     "    revision=REVISION,\n",
     "    allow_patterns=[\"*.parquet\"],\n",
     ")"
@@ -109,7 +109,7 @@
     "# Create a view called 'open_data', and count the number of rows and distinct documents\n",
     "# in the view\n",
     "db.execute(\n",
-    "    f\"CREATE VIEW open_data AS SELECT * FROM read_parquet('{CACHE_DIR}/*.parquet')\"\n",
+    "    f\"CREATE VIEW open_data AS SELECT * FROM read_parquet('{DATA_CACHE_DIR}/*.parquet')\"\n",
     ")\n",
     "db.sql(\"SELECT COUNT(*), COUNT(DISTINCT document_id) FROM open_data\")"
    ]
diff --git a/src/streamlit_apps/searchable_world_map.py b/src/streamlit_apps/searchable_world_map.py
index c4d91e4..c0c06a8 100644
--- a/src/streamlit_apps/searchable_world_map.py
+++ b/src/streamlit_apps/searchable_world_map.py
@@ -10,14 +10,14 @@
 
 
 from src.data_helpers import download_data
-from src.config import CACHE_DIR, REVISION
+from src.config import CACHE_DIR, DATA_REVISION
 
 
 @st.cache_resource
 def load_data():
     download_data(
         cache_dir=str(CACHE_DIR),
-        revision=REVISION,
+        revision=DATA_REVISION,
     )
 
     db = duckdb.connect()