I'd like to start by saying that neither I nor Mr. Linklater was there at the time, so I will be answering based on my knowledge, which I gathered either through discussions, through reading, or through colleagues. Before I do, I just want to point out that we often forget that this pay transformation was really two projects. There was the modernization—the IT project—but there was also the consolidation, which was the pay advisers. Those two happened at the same time.
In terms of contingency, to go directly to your question, you're correct from what I gathered also that the pilot project was not done. The decision not to do the pilot project was made considering that there was going to be more testing done closer to wave one. The contingency plan was for the transfer of the data. My understanding is that the biggest risk felt at the time was that when you actually turned the switch on, the data would not transfer from one system to the other, and the paycheques would not go out every two weeks.
There was a contingency plan until after the first wave. If there had been any fatal errors in that transfer, they would not have moved to the second wave. There were no fatal errors, so, by default, the second wave was basically going.
In terms of a contingency plan for what we're facing now, in terms of the compensation advisers and being able to process all of the transactions, it's clear that there was no contingency plan at that level. Frankly, I think part of what we're learning is that the complexity of this was never estimated. It was underestimated. There was no contingency plan for that aspect, so we had to react in real time to increase the capacity of the people to be able to do the transactions.