diff --git a/suite/suite.go b/suite/suite.go
index d62c743a2..790eaa3e4 100644
--- a/suite/suite.go
+++ b/suite/suite.go
@@ -63,6 +63,19 @@ func failOnPanic(t *testing.T) {
 	}
 }
 
+// Run provides suite functionality around golang subtests.  It should be
+// called in place of t.Run(name, func(t *testing.T)) in test suite code.
+// The passed-in func will be executed as a subtest with a fresh instance of t.
+// Provides compatibility with go test pkg -run TestSuite/TestName/SubTestName.
+func (suite *Suite) Run(name string, subtest func()) {
+	oldT := suite.T()
+	defer suite.SetT(oldT)
+	oldT.Run(name, func(t *testing.T) {
+		suite.SetT(t)
+		subtest()
+	})
+}
+
 // Run takes a testing suite and runs all of the tests attached
 // to it.
 func Run(t *testing.T, suite TestingSuite) {
diff --git a/suite/suite_test.go b/suite/suite_test.go
index 1c69d5472..6dfba08cd 100644
--- a/suite/suite_test.go
+++ b/suite/suite_test.go
@@ -152,6 +152,7 @@ type SuiteTester struct {
 	TearDownTestRunCount  int
 	TestOneRunCount       int
 	TestTwoRunCount       int
+	TestSubtestRunCount   int
 	NonTestMethodRunCount int
 
 	SuiteNameBefore []string
@@ -246,6 +247,27 @@ func (suite *SuiteTester) NonTestMethod() {
 	suite.NonTestMethodRunCount++
 }
 
+func (suite *SuiteTester) TestSubtest() {
+	suite.TestSubtestRunCount++
+
+	for _, t := range []struct {
+		testName string
+	}{
+		{"first"},
+		{"second"},
+	} {
+		suiteT := suite.T()
+		suite.Run(t.testName, func() {
+			// We should get a different *testing.T for subtests, so that
+			// go test recognizes them as proper subtests for output formatting
+			// and running individual subtests
+			subTestT := suite.T()
+			suite.NotEqual(subTestT, suiteT)
+		})
+		suite.Equal(suiteT, suite.T())
+	}
+}
+
 // TestRunSuite will be run by the 'go test' command, so within it, we
 // can run our suite using the Run(*testing.T, TestingSuite) function.
 func TestRunSuite(t *testing.T) {
@@ -261,18 +283,20 @@ func TestRunSuite(t *testing.T) {
 	assert.Equal(t, suiteTester.SetupSuiteRunCount, 1)
 	assert.Equal(t, suiteTester.TearDownSuiteRunCount, 1)
 
-	assert.Equal(t, len(suiteTester.SuiteNameAfter), 3)
-	assert.Equal(t, len(suiteTester.SuiteNameBefore), 3)
-	assert.Equal(t, len(suiteTester.TestNameAfter), 3)
-	assert.Equal(t, len(suiteTester.TestNameBefore), 3)
+	assert.Equal(t, len(suiteTester.SuiteNameAfter), 4)
+	assert.Equal(t, len(suiteTester.SuiteNameBefore), 4)
+	assert.Equal(t, len(suiteTester.TestNameAfter), 4)
+	assert.Equal(t, len(suiteTester.TestNameBefore), 4)
 
 	assert.Contains(t, suiteTester.TestNameAfter, "TestOne")
 	assert.Contains(t, suiteTester.TestNameAfter, "TestTwo")
 	assert.Contains(t, suiteTester.TestNameAfter, "TestSkip")
+	assert.Contains(t, suiteTester.TestNameAfter, "TestSubtest")
 
 	assert.Contains(t, suiteTester.TestNameBefore, "TestOne")
 	assert.Contains(t, suiteTester.TestNameBefore, "TestTwo")
 	assert.Contains(t, suiteTester.TestNameBefore, "TestSkip")
+	assert.Contains(t, suiteTester.TestNameBefore, "TestSubtest")
 
 	for _, suiteName := range suiteTester.SuiteNameAfter {
 		assert.Equal(t, "SuiteTester", suiteName)
@@ -290,15 +314,16 @@ func TestRunSuite(t *testing.T) {
 		assert.False(t, when.IsZero())
 	}
 
-	// There are three test methods (TestOne, TestTwo, and TestSkip), so
+	// There are four test methods (TestOne, TestTwo, TestSkip, and TestSubtest), so
 	// the SetupTest and TearDownTest methods (which should be run once for
-	// each test) should have been run three times.
-	assert.Equal(t, suiteTester.SetupTestRunCount, 3)
-	assert.Equal(t, suiteTester.TearDownTestRunCount, 3)
+	// each test) should have been run four times.
+	assert.Equal(t, suiteTester.SetupTestRunCount, 4)
+	assert.Equal(t, suiteTester.TearDownTestRunCount, 4)
 
 	// Each test should have been run once.
 	assert.Equal(t, suiteTester.TestOneRunCount, 1)
 	assert.Equal(t, suiteTester.TestTwoRunCount, 1)
+	assert.Equal(t, suiteTester.TestSubtestRunCount, 1)
 
 	// Methods that don't match the test method identifier shouldn't
 	// have been run at all.