You may find your corporate bean counters getting a bit more focused on demonstrating the value of every investment given the current economic conditions. When we do a consulting analysis for a customer, we typically present a section on what the benefits are -- for example, how much agent time could be saved if a given task used speech automation.  Then the customers typically use these estimates to "prove-in" the investment required to deploy that automation.  Everyone thinks, "This is great."

But I rarely see our customers dive deeper into planning on how those benefits will be realized, especially related to the calendar.  For some customers with high call volumes, the savings from full deployment can be significant, sometimes over $1M per year.  But typically once the analysis is accepted by the customer, the actual deployment just plods on.

Here is an innovative approach I recently saw a large customer use -- they drove a development project based on "in-year savings."  This large client was serious -- 3 out of 6 call centers were being closed because of business conditions. They had to go so far as to give agents notice before deployment was complete -- talk about pressure! 

But the good part of this was the focus the customer brought to pushing the best parts of the application earlier in the schedule. Typically, a large application is made up of many pieces, especially when you are planning on automating maybe 10-20 transactions currently done by agents.  But you don't need to take 12 months and deliver the whole thing; it's worth the analysis to look at the transactions individually.  What actual savings does each bring? And what dependencies are there? In many cases, you can take a large project and break it up into several pieces, so several high-benefit transactions can get deployed months before the whole project is done. 

By focusing on "how much can I actually save in this fiscal year" you're forced to look at creative ways to find early savings.