ALTER TABLE <table name> ADD <columnName> ROWVERSION;

Nothing special

Evaluate how often that table is accessed by the ID column and how often it's accessed by the date column. If it's evaluated by the date far more often, consider...

That's probably something you need to address...

CASE wHEN (DATEDIFF(day, A.StartDate, A.EndDate)) = 0 THEN 1 ELSE Sum(A.budgetedTotalWorkload)/ (DATEDIFF(day, A.StartDate, A.EndDate)) END

Don't, it'll have no effect on performance. Stats aren't created or updated until needed.

Do an update with fullscan afterwards, but no need to disable any options.

Same answer.

Well of course it'll go for a clustered index scan if there's only a clustered index on the table. There's no other possible method of getting the rows required.

Do you...

Index usage is not tracked to the column level.

Start by taking out the nolock (unless incorrect results are acceptable) and replacing the * with the specific columns you need. Then post the other indexes the table has (because...

You could capture and analyse every single query that executes against that table and test them all out with the narrower index.

Why are you worried about a single column?

I'm 6 years older than you are and I'm on the younger end of the SQL people that I know. Definitely on the younger end of the people at the...

Edit: Not sure anymore...

A nonclustered index will always have 0 lookups. Lookups only occur to the clustered index. DMVs track the index usage, nothing tracks usage to the column level.

Not required, waste of time at best.

Fire that vendor.