There's no dispute that there is a lot of retailer data available to  all parties to a transaction. "There's certainly enough data," Byrne says. "Some would say perhaps there's too much." But if that's so, then why isn't that information being utilized?

Byrne says there's has been "turmoil" from an application standpoint, that companies are just starting to bring to market what he calls structured ways to deal with the data. Clearly, companies have done a pretty good job of collecting oceans of information. The breakdown has been in figuring out what to do with it now that they have access to it.

Up to now, CPG companies have not excelled at getting their product supply lead-time inside their order lead-time, Byrne says. So while they will never be champions of the make-to-order model, they can do a lot better.

Increased downstream data should have led to improved supply chain performance by now, but often companies have been foggy on what to do next. Simply extracting data from last week's sales doesn't necessarily tell you much about this week's, Byrne says. You need to sift "meaningful" information from the data. Models such as "scan one, make one" - referring to making replacements based solely on point-of-sale information - aren't particularly helpful or realistic when promotions spike sales.

On the demand side, companies need to figure out what their demand signals mean. That's especially crucial given the economy. Eighteen to 24 months ago, companies looked at growth as a big driver, but when that didn't happen, they had to attack inventory to take costs out. Something better is needed to become truly demand-driven.

On the supply side, companies must revisit their lead-times and determine how they can shorten them.

To view this video interview in its entirety, Click Here.