diff --git a/docs/python-api.rst b/docs/python-api.rst
index 68ef7e2..fb060b4 100644
--- a/docs/python-api.rst
+++ b/docs/python-api.rst
@@ -182,6 +182,40 @@ Note that the ``pk`` and ``column_order`` parameters here are optional if you ar
 
 An ``upsert_all()`` method is also available, which behaves like ``insert_all()`` but performs upserts instead.
 
+
+Adding columns
+==============
+
+You can add a new column to a table using the ``.add_column(col_name, col_type)`` method:
+
+.. code-block:: python
+
+    db["dogs"].add_column("dob", datetime.date)
+
+You can pass any of the following Python types as the second argument - ``sqlite-utils`` will create the column using the appropriate SQLite type from this list::
+
+    float: "FLOAT"
+    int: "INTEGER"
+    bool: "INTEGER"
+    str: "TEXT"
+    bytes: "BLOB"
+    datetime.datetime: "TEXT"
+    datetime.date: "TEXT"
+    datetime.time: "TEXT"
+
+    # If numpy is installed
+    np.int8: "INTEGER"
+    np.int16: "INTEGER"
+    np.int32: "INTEGER"
+    np.int64: "INTEGER"
+    np.uint8: "INTEGER"
+    np.uint16: "INTEGER"
+    np.uint32: "INTEGER"
+    np.uint64: "INTEGER"
+    np.float16: "FLOAT"
+    np.float32: "FLOAT"
+    np.float64: "FLOAT"
+
 .. _python_api_hash:
 
 Setting an ID based on the hash of the row contents
diff --git a/sqlite_utils/db.py b/sqlite_utils/db.py
index 996a501..3a41434 100644
--- a/sqlite_utils/db.py
+++ b/sqlite_utils/db.py
@@ -19,6 +19,36 @@ ForeignKey = namedtuple(
 )
 Index = namedtuple("Index", ("seq", "name", "unique", "origin", "partial", "columns"))
 
+COLUMN_TYPE_MAPPING = {
+    float: "FLOAT",
+    int: "INTEGER",
+    bool: "INTEGER",
+    str: "TEXT",
+    bytes.__class__: "BLOB",
+    bytes: "BLOB",
+    datetime.datetime: "TEXT",
+    datetime.date: "TEXT",
+    datetime.time: "TEXT",
+    None.__class__: "TEXT",
+}
+# If numpy is available, add more types
+if np:
+    COLUMN_TYPE_MAPPING.update(
+        {
+            np.int8: "INTEGER",
+            np.int16: "INTEGER",
+            np.int32: "INTEGER",
+            np.int64: "INTEGER",
+            np.uint8: "INTEGER",
+            np.uint16: "INTEGER",
+            np.uint32: "INTEGER",
+            np.uint64: "INTEGER",
+            np.float16: "FLOAT",
+            np.float32: "FLOAT",
+            np.float64: "FLOAT",
+        }
+    )
+
 
 class Database:
     def __init__(self, filename_or_conn):
@@ -67,39 +97,10 @@ class Database:
             column_items.insert(0, (hash_id, str))
             pk = hash_id
         extra = ""
-        col_type_mapping = {
-            float: "FLOAT",
-            int: "INTEGER",
-            bool: "INTEGER",
-            str: "TEXT",
-            bytes.__class__: "BLOB",
-            bytes: "BLOB",
-            datetime.datetime: "TEXT",
-            datetime.date: "TEXT",
-            datetime.time: "TEXT",
-            None.__class__: "TEXT",
-        }
-        # If numpy is available, add more types
-        if np:
-            col_type_mapping.update(
-                {
-                    np.int8: "INTEGER",
-                    np.int16: "INTEGER",
-                    np.int32: "INTEGER",
-                    np.int64: "INTEGER",
-                    np.uint8: "INTEGER",
-                    np.uint16: "INTEGER",
-                    np.uint32: "INTEGER",
-                    np.uint64: "INTEGER",
-                    np.float16: "FLOAT",
-                    np.float32: "FLOAT",
-                    np.float64: "FLOAT",
-                }
-            )
         columns_sql = ",\n".join(
             "   [{col_name}] {col_type}{primary_key}{references}".format(
                 col_name=col_name,
-                col_type=col_type_mapping[col_type],
+                col_type=COLUMN_TYPE_MAPPING[col_type],
                 primary_key=" PRIMARY KEY" if (pk == col_name) else "",
                 references=(
                     " REFERENCES [{other_table}]([{other_column}])".format(
@@ -254,6 +255,15 @@ class Table:
         self.db.conn.execute(sql)
         return self
 
+    def add_column(self, col_name, col_type):
+        sql = "ALTER TABLE [{table}] ADD COLUMN [{col_name}] {col_type};".format(
+            table=self.name,
+            col_name=col_name,
+            col_type=COLUMN_TYPE_MAPPING[col_type],
+        )
+        self.db.conn.execute(sql)
+        return self
+
     def drop(self):
         return self.db.conn.execute("DROP TABLE {}".format(self.name))
 
diff --git a/tests/test_create.py b/tests/test_create.py
index 4c7b550..6528a23 100644
--- a/tests/test_create.py
+++ b/tests/test_create.py
@@ -4,6 +4,7 @@ import datetime
 import json
 import pathlib
 import pytest
+import re
 import sqlite3
 
 try:
@@ -122,6 +123,32 @@ def test_create_table_works_for_m2m_with_only_foreign_keys(fresh_db):
     )
 
 
+@pytest.mark.parametrize(
+    "col_name,col_type,expected_schema",
+    (
+        (
+            "nickname", str, "CREATE TABLE [dogs] ( [name] TEXT , [nickname] TEXT)"
+        ),
+        (
+            "dob", datetime.date, "CREATE TABLE [dogs] ( [name] TEXT , [dob] TEXT)"
+        ),
+        (
+            "age", int, "CREATE TABLE [dogs] ( [name] TEXT , [age] INTEGER)"
+        ),
+        (
+            "weight", float, "CREATE TABLE [dogs] ( [name] TEXT , [weight] FLOAT)"
+        ),
+    )
+)
+def test_add_column(fresh_db, col_name, col_type, expected_schema):
+    fresh_db.create_table("dogs", {"name": str})
+    assert "CREATE TABLE [dogs] ( [name] TEXT )" == collapse_whitespace(
+        fresh_db["dogs"].schema
+    )
+    fresh_db["dogs"].add_column(col_name, col_type)
+    assert expected_schema == collapse_whitespace(fresh_db["dogs"].schema)
+
+
 @pytest.mark.parametrize(
     "columns,index_name,expected_index",
     (
@@ -337,3 +364,10 @@ def test_create_table_numpy(fresh_db):
             "np.uint8": 8,
         }
     ] == list(fresh_db["types"].rows)
+
+
+COLLAPSE_RE = re.compile("\s+")
+
+
+def collapse_whitespace(s):
+    return COLLAPSE_RE.sub(" ", s)