diff --git a/setup.py b/setup.py
index aff7834d7b..99db831484 100755
--- a/setup.py
+++ b/setup.py
@@ -69,7 +69,7 @@ kwargs = {
         'pandas', 'lxml', 'uncertainties'
     ],
     'extras_require': {
-        'test': ['pytest', 'pytest-cov'],
+        'test': ['pytest', 'pytest-cov', 'colorama'],
         'vtk': ['vtk'],
     },
 }
diff --git a/tests/testing_harness.py b/tests/testing_harness.py
index 0d3c4bec00..2266189c05 100644
--- a/tests/testing_harness.py
+++ b/tests/testing_harness.py
@@ -10,9 +10,25 @@ import sys
 import numpy as np
 import openmc
 from openmc.examples import pwr_core
+from colorama import Fore, init
 
 from tests.regression_tests import config
 
+init()
+
+
+def colorize(diff):
+    """Produce colored diff for test results"""
+    for line in diff:
+        if line.startswith('+'):
+            yield Fore.RED + line + Fore.RESET
+        elif line.startswith('-'):
+            yield Fore.GREEN + line + Fore.RESET
+        elif line.startswith('^'):
+            yield Fore.BLUE + line + Fore.RESET
+        else:
+            yield line
+
 
 class TestHarness(object):
     """General class for running OpenMC regression tests."""
@@ -108,11 +124,17 @@ class TestHarness(object):
         shutil.copyfile('results_test.dat', 'results_true.dat')
 
     def _compare_results(self):
-        """Make sure the current results agree with the _true standard."""
+        """Make sure the current results agree with the reference."""
         compare = filecmp.cmp('results_test.dat', 'results_true.dat')
         if not compare:
+            expected = open('results_true.dat').readlines()
+            actual = open('results_test.dat').readlines()
+            diff = unified_diff(expected, actual, 'results_true.dat',
+                                'results_test.dat')
+            print('Result differences:')
+            print(''.join(colorize(diff)))
             os.rename('results_test.dat', 'results_error.dat')
-        assert compare, 'Results do not agree.'
+        assert compare, 'Results do not agree'
 
     def _cleanup(self):
         """Delete statepoints, tally, and test files."""
@@ -325,11 +347,13 @@ class PyAPITestHarness(TestHarness):
         """Make sure the current inputs agree with the _true standard."""
         compare = filecmp.cmp('inputs_test.dat', 'inputs_true.dat')
         if not compare:
+            expected = open('inputs_true.dat', 'r').readlines()
+            actual = open('inputs_error.dat', 'r').readlines()
+            diff = unified_diff(expected, actual, 'inputs_true.dat',
+                                'inputs_error.dat')
+            print('Input differences:')
+            print(''.join(colorize(diff)))
             os.rename('inputs_test.dat', 'inputs_error.dat')
-            for line in unified_diff(open('inputs_true.dat', 'r').readlines(),
-                                     open('inputs_error.dat', 'r').readlines(),
-                                     'inputs_true.dat', 'inputs_error.dat'):
-                print(line, end='')
         assert compare, 'Input files are broken.'
 
     def _cleanup(self):