Good point. Only this is just a small part of a much larger dataset that all has to be in the same matrix. Unfortunately, SSRS does not like to have more than one DataSet in a matrix without a bunch of hoops and gotchas; so I use Stored Procedures to do most of the heavy lifting and only return a small amount of rows. This keeps the report processing lean and mean.

The original dataset would be well over 153,000 rows over 100 columns if I let the matrices do all the walking. When I did that the report took over 45 minutes to process. Now, with this final piece I can whittle that monster dataset down to the 24 rows of salespeople and 60 columns. I have found that in SSRS if you have more than just about 40 records and 50-60 columns of data the report is too slow for live viewing; and has to be cached nightly; which will not work for the salespeople who must check the report multiple times per day to see where they are. Also, because we use Dynamics GP (another "brilliant" Microsoft invention) the data is so normalized that it takes no less than 3-4 complex joins just to get a list of what was sold and who sold it for any given point in time.

I know its a long reply, but I wanted to let you know I appreciate the thought and have actually gone down that rabbit hole with no success.

Cheers!