Sounds like a typical job for SAS/Add-in for MS Office (AMO).
If your business users a familiar with pivot tables, using OLAP could work well (if the data is suite for multi dimensional analysis).
A part from that, AMO has pretty much the same task (wizards) as EG, an let's the user browse, filter and query data.
But when you say large, how large?
It could be performance issues with users that are not database aware, query very large tables. You may want to create smaller specialized data/information marts for some users.
You also may want to combine this with the use of information maps, which can cover database logic (such as star schema joins, row level security and caclulated columns.
You can search support.sas.com to find lot of documentation and papers on this topic.
/Linus
Data never sleeps