diff --git a/clients/bucket_schema/csv.go b/clients/bucket_schema/csv.go
index 92c27aa..93c8ac7 100644
--- a/clients/bucket_schema/csv.go
+++ b/clients/bucket_schema/csv.go
@@ -11,7 +11,7 @@ import (
 type csvColumn struct {
 	Name     string                 `csv:"name"`
 	Type     api.ColumnSemanticType `csv:"type"`
-	DataType *api.ColumnDataType    `csv:"data_type,omitempty"`
+	DataType *api.ColumnDataType    `csv:"dataType,omitempty"`
 }
 
 func init() {
diff --git a/clients/bucket_schema/csv_test.go b/clients/bucket_schema/csv_test.go
index f7d5d76..039a9e8 100644
--- a/clients/bucket_schema/csv_test.go
+++ b/clients/bucket_schema/csv_test.go
@@ -20,7 +20,7 @@ func TestDecodeCSV(t *testing.T) {
 		{
 			name: "valid",
 			data: heredoc.Doc(`
-				name,type,data_type
+				name,type,dataType
 				time,timestamp,
 				host,tag,
 				usage_user,field,float
@@ -30,7 +30,7 @@ func TestDecodeCSV(t *testing.T) {
 		{
 			name: "valid with alternate order",
 			data: heredoc.Doc(`
-				type,data_type,name
+				type,dataType,name
 				timestamp,,time
 				tag,,host
 				field,float,usage_user
@@ -40,7 +40,7 @@ func TestDecodeCSV(t *testing.T) {
 		{
 			name: "invalid column type",
 			data: heredoc.Doc(`
-				name,type,data_type
+				name,type,dataType
 				time,foo,
 			`),
 			expErr: `failed to decode CSV: record on line 0; parse error on line 2, column 2: "foo" is not a valid column type. Valid values are [timestamp, tag, field]`,
@@ -48,23 +48,32 @@ func TestDecodeCSV(t *testing.T) {
 		{
 			name: "invalid column data type",
 			data: heredoc.Doc(`
-				name,type,data_type
+				name,type,dataType
 				time,field,floaty
 			`),
 			expErr: `failed to decode CSV: record on line 0; parse error on line 2, column 3: "floaty" is not a valid column data type. Valid values are [integer, float, boolean, string, unsigned]`,
 		},
+		{
+			name: "invalid dataType header",
+			data: heredoc.Doc(`
+				name,type,data_type
+				time,field,float
+				time2,field,
+			`),
+			expErr: `failed to decode CSV: found unmatched struct field with tags [dataType]`,
+		},
 		{
 			name: "invalid headers",
 			data: heredoc.Doc(`
-				name,foo
+				name,foo,
 				time,field
 			`),
-			expErr: `failed to decode CSV: found unmatched struct field with tags [type data_type]`,
+			expErr: `failed to decode CSV: record on line 2: wrong number of fields`,
 		},
 		{
 			name: "invalid CSV",
 			data: heredoc.Doc(`
-				type,type,data_type
+				type,type,dataType
 				time,timestamp
 			`),
 			expErr: `failed to decode CSV: record on line 2: wrong number of fields`,
diff --git a/clients/bucket_schema/testdata/columns.csv b/clients/bucket_schema/testdata/columns.csv
index 23b21c3..2f17763 100644
--- a/clients/bucket_schema/testdata/columns.csv
+++ b/clients/bucket_schema/testdata/columns.csv
@@ -1,4 +1,4 @@
-name,type,data_type
+name,type,dataType
 time,timestamp,
 host,tag,
 usage_user,field,float
\ No newline at end of file