In order to reinforce the desired behavior, you must first get the behavior. Some behaviors occur naturally, such as sitting, lying down or barking. Not only do these behaviors occur naturally, but they are apt to occur at predictable times as well. You can place the animal in the situation where the behavior is most likely to happen and then wait patiently.
An alternative for those who have no patience, and for those behaviors, such as shaking hands, playing dead or begging that do not occur naturally, there is another method. These behaviors can be elicited. An easy way of eliciting a behavior is to lure it with food. Teaching the "sit" is an easy way to demonstrate this method.
Touch a treat to the dog's nose and then lift it up slightly and move it toward the dog's forehead. If you don't lift the treat too high, the dog will follow the treat with his nose and his rear end will just naturally hit the ground. You "click" as his rear end hits the ground (timing the click is critical!) and give him the treat. You have "lured" him into a sit. Repeat this several times and then simply "tease" him by touching a treat to his nosing briefly and then removing it. Chances are he will offer the sit to you. Once you have him offering you the sit, you can begin attaching the cue to the behavior.
Cues are attached to a behavior only after you have the dog offering the behavior. This is because you can only teach one thing at a time, and since a cue without a behavior is useless, you might as well teach the behavior first. Remember that your dog does not understand English ( regardless of what they say about Lassie ). A word only becomes meaningful when it has been paired with an action over many repetitions. So first get the dog to offer you the behavior on a regular basis; then put a name to it; fade the reinforcement as he becomes proficient....and then start teaching another behavior!!!