SQL Server - Detecting non-indexed columns but used in WHERE clause

How to detect a column included in WHERE clause but used in indexed?

Little Background:

Until the time the table has few 开发者_JAVA百科number of records things will be okay, once it started having millions of records then index should be created for a column which is used in WHERE clauses in stored procs, inline queries etc.,

Since we have hundreds of stored procs and queries that often gets changed by the devs I wanted to have a automated way of identifying those columns that are used in WHERE clauses but not an index is created. How to do that in SQL Server 2008?

Use the "Missing indexes" dmv stuff to spot what could be added?

• Example (this is weighted so you have an idea of cost too)
• MSDN

Edit: I'll qualify... this allows you to consider indexes based on cost/use/benefit. An index may only be used at 04:00 on a Sunday, so from a cost perspective it's probably not worth it

You should first identify your expensive queries: How Can I Log and Find the Most Expensive Queries?