In training scent specific work (cadaver, water, narcotics etc.), the first step is to imprint. During that process, some natural behavior comes out in the dog when he starts understanding the scent. 

The second step is to build the natural behavior the dog exhibits into the dog's alert.  In the case of land cadaver, I like to build a passive alert. This stage of training is the most important part of the process. Before moving on, into search patterns and search scenarios, a consistent, reliable alert should be built into the dog. Now this is not to say that you bore the dog by staying on blocks or minnow buckets. You can and should vary your dog's training area. But before you go on to search patterns and search scenarios, the alert should be rock solid.  It should be consistent and offered without input from the handler. This requires repetition before moving on. A reward system eliminates boredom and changing the location of the hide provides variety and allows the dog to "think". If you move forward too soon and add search patterns, lengthen your search time, or tax your dog's search stamina, before you have a consistent, reliable alert, the "fluidity" of your dog's alert is going to go to hell in a hand basket. 

In water work, the number one complaint students have is the inability to "read" their dog. That's because the alert building step was skipped in the process. Time spent at this stage creates a team that is confident in each others abilities and more successful in real world applications.

Jonni Joyce