Define a clustered index on the column that appears most frequently in the WHERE clause of SELECT
statements.
But not in JOIN criteria? No, no way. The basic rule for clusters should be the most frequently used access path to the data. This may be primary keys, this may be foreign keys, or it might be simply search criteria, but I wouldn't suggest limiting it to WHERE clauses.
"The credit belongs to the man who is actually in the arena, whose face is marred by dust and sweat and blood"
- Theodore Roosevelt
Author of:
SQL Server Execution Plans
SQL Server Query Performance Tuning