IBM made a big splash with its introduction of Dynamic Data Warehousing -- and with good reason. Whatever the underlying technology may be -- continuous trickle-feeding of data is part of it -- the use case it provided was the most compelling I've seen to date.
Picture this: You're a claims agent for a major insurance company, dealing with complaints all day long. Someone calls up to complain about a rejected claim, and your system provides an instant history of that individual's prior calls. You see a bar graph that tells you he has had several rejected claims in the past, and all because the forms were filled out improperly. So, you can be pretty sure what went wrong this time, and fix it faster because you know where to look.
Of course, the case for real-time ETL has been made before. In an interview with IT Business Edge last August, Jennifer St. Louis of DataMirror described a warehouse customer that needed software to plan personnel needs for shipments due to go out that day. Obviously, day-old data wouldn't work in a situation like that.
As the real-time data applications gain steam, you may even end up seeing them operating in your daily life. Example: At the supermarket, you may get free coupons at check-out based on what you just bought. The bottom line, it seems, is this: Where there's instant data, there's instant gratification.