Yes. That's a fair point. In fact, there have been two rounds of implementation already. One of them was importer-facing and the other one was not. We have seen some pieces of it.
I think the opportunity or lesson learned, as we talked about earlier, is why did we have to do a big bang approach? Why didn't we take an agile approach? Really, that's the lost opportunity. Most people around the table have done systems implementation, whether in government or in the private sector. Typically, you go through the design phase. You start the testing. You have your lessons learned. You rejig. Then you implement. Then you do your process improvement. That's not what we're experiencing here. That's what we've experienced on past projects, so that, I guess, is a lesson reinforced in this case.
We hope that once we move forward and get it in and stabilized, we will have those iterative processes. We're just very concerned at what cost that will be, at this point, since we're going to go big bang on May 13.