As Kimberly Tripp - the Queen of Indexing - explains quite nicely in her blog post The Clustered Index Debate continues..., having a clustering key on a database table pretty much speeds up all operations - not just SELECT.
SELECT are generally slower on a heap as compared to a clustered table, as long as you pick a good clustering key - something like an INT IDENTITY. If you use a really really bad clustering key, like a GUID or a compound key with lots of variable length components, then, but only then, a heap might be faster. But in that case, you really need to clean up your database design in the first place...
So in general, I don't think there's any point in a heap - pick a good, useful clustering key and you should benefit in all respects.
