T-SQL: Fast concat table rows

concat.md

create table #rows (id uniqueidentifier, value varchar(10))

insert into #rows
values	('11111111-1111-1111-1111-111111111111', 'a'),
	('22222222-2222-2222-2222-222222222222', 'b'),
	('11111111-1111-1111-1111-111111111111', 'c'),
	('22222222-2222-2222-2222-222222222222', 'd'),
	('11111111-1111-1111-1111-111111111111', 'e'),
	('33333333-3333-3333-3333-333333333333', 'f'),
	('33333333-3333-3333-3333-333333333333', 'g')

create table #result (id uniqueidentifier, value varchar(10))
create clustered index idx_result on #result(id) -- for ordered

insert into #result
select * from #rows order by id -- Important: result must be ordered by id

declare	 @value varchar(max),
	 @id uniqueidentifier = '00000000-0000-0000-0000-000000000000'

update #result 
set  
	@value = value = case 
				when @id <> id then isnull(value,'') 
				else @value + isnull(value,'') 
			 end,
	@id = id

select id, max(value) from #result group by id