diff --git a/index.js b/index.js
index 24e7db0..773ae6a 100644
--- a/index.js
+++ b/index.js
@@ -83,8 +83,6 @@ module.exports = function (encodedURI) {
 	}
 
 	try {
-		encodedURI = encodedURI.replace(/\+/g, ' ');
-
 		// Try the built in decoder first
 		return decodeURIComponent(encodedURI);
 	} catch (err) {
diff --git a/readme.md b/readme.md
index 3ed99b0..bd71b4b 100644
--- a/readme.md
+++ b/readme.md
@@ -7,7 +7,6 @@
 
 ## Why?
 
-- Decodes `+` to a space.
 - Converts the [BOM](https://en.wikipedia.org/wiki/Byte_order_mark) to a [replacement character](https://en.wikipedia.org/wiki/Specials_(Unicode_block)#Replacement_character) `�`.
 - Does not throw with invalid encoded input.
 - Decodes as much of the string as possible.
diff --git a/test.js b/test.js
index 86fabd5..0424493 100644
--- a/test.js
+++ b/test.js
@@ -3,8 +3,8 @@ import m from './';
 
 const tests = {
 	'test': 'test',
-	'a+b': 'a b',
-	'a+b+c+d': 'a b c d',
+	'a+b': 'a+b',
+	'a+b+c+d': 'a+b+c+d',
 	'=a': '=a',
 	'%': '%',
 	'%25': '%',
@@ -20,7 +20,7 @@ const tests = {
 	'%7 B%ab%%7C%de%%7 D': '%7 B%ab%|%de%%7 D',
 	'%ab': '%ab',
 	'%ab%ab%ab': '%ab%ab%ab',
-	'%61+%4d%4D': 'a MM',
+	'%61+%4d%4D': 'a+MM',
 	'\uFEFFtest': '\uFEFFtest',
 	'\uFEFF': '\uFEFF',
 	'%EF%BB%BFtest': '\uFEFFtest',