diff --git a/cpp/ycm/Result.cpp b/cpp/ycm/Result.cpp
index eb7347a4..26b18812 100644
--- a/cpp/ycm/Result.cpp
+++ b/cpp/ycm/Result.cpp
@@ -19,6 +19,7 @@
 #include "standard.h"
 #include "Utils.h"
 #include <boost/algorithm/string.hpp>
+#include <boost/function.hpp>
 #include <algorithm>
 
 using boost::algorithm::istarts_with;
@@ -27,6 +28,31 @@ namespace YouCompleteMe {
 
 namespace {
 
+char ChangeCharCase( char c ) {
+  if ( std::isupper( c ) )
+    return std::tolower( c );
+  return std::toupper( c );
+}
+
+
+bool CharLessThanWithLowercasePriority(const char &first,
+                                       const char &second) {
+  char swap_first = ChangeCharCase( first );
+  char swap_second = ChangeCharCase( second );
+  return swap_first < swap_second;
+}
+
+
+bool StringLessThanWithLowercasePriority(const std::string &first,
+                                         const std::string &second) {
+  return std::lexicographical_compare(
+      first.begin(), first.end(),
+      second.begin(), second.end(),
+      boost::function< bool( const char&, const char& ) >(
+          &CharLessThanWithLowercasePriority ) );
+}
+
+
 int LongestCommonSubsequenceLength( const std::string &first,
                                     const std::string &second ) {
   const std::string &longer  = first.size() > second.size() ? first  : second;
@@ -168,8 +194,9 @@ bool Result::operator< ( const Result &other ) const {
       return text_is_lowercase_;
   }
 
-  // Lexicographic comparison
-  return *text_ < *other.text_;
+  // Lexicographic comparison, but we prioritize lowercase letters over
+  // uppercase ones. So "foo" < "Foo".
+  return StringLessThanWithLowercasePriority( *text_, *other.text_ );
 }
 
 
diff --git a/cpp/ycm/tests/IdentifierCompleter_test.cpp b/cpp/ycm/tests/IdentifierCompleter_test.cpp
index d3c4d0dc..7b17c1ab 100644
--- a/cpp/ycm/tests/IdentifierCompleter_test.cpp
+++ b/cpp/ycm/tests/IdentifierCompleter_test.cpp
@@ -206,6 +206,24 @@ TEST( IdentifierCompleterTest, SameLowercaseCandidateWins ) {
                    "Foobar" ) ).CandidatesForQuery( "foo" ),
                ElementsAre( "foobar",
                             "Foobar" ) );
+
+}
+
+TEST( IdentifierCompleterTest, PreferLowercaseCandidate ) {
+  EXPECT_THAT( IdentifierCompleter(
+                 StringVector(
+                   "chatContentExtension",
+                   "ChatContentExtension" ) ).CandidatesForQuery(
+                       "chatContent" ),
+               ElementsAre( "chatContentExtension",
+                            "ChatContentExtension" ) );
+
+  EXPECT_THAT( IdentifierCompleter(
+                 StringVector(
+                   "fooBar",
+                   "FooBar" ) ).CandidatesForQuery( "oba" ),
+               ElementsAre( "fooBar",
+                            "FooBar" ) );
 }
 
 TEST( IdentifierCompleterTest, ShorterAndLowercaseWins ) {