I have attached one of the best videos on how to train a dog to come every time.

Video train a dog to come everytime

More indepth info on training

How a change in thinking that will greatly increase your dog training success. 

A common way of thinking about how dog’s interact with us is based on a human viewpoint.  We wish that dog’s behave  because we tell them what to do and  motivated by love  & desire to please us they do what they are told. A good dog behaves well and bad dog’s don’t.   Hence the name obedience training, which has been used interchangeably for dog training for generations.

But, science into how learning happens shows us otherwise.  So how do dog’s learn “to come” when called?

I will break down into parts how learning  works.

First,  the word, “come’.  Dogs and most animals respond to words as just sounds. Human speech is hard for dogs to distinguish and sounds like noise to dogs, much as barking does to us.   A specific word only becomes a sound signal after the dog figures out the importance and usage of the sound.  We decide that certain words are important for training as cues for dogs. The dog does not know about that.  Not until the dog learns the use of a word in a training sequence does the word  becomes a sound signal.  We must also teach the dog, which behaviors go with the signal, the order to act out the behavior and which behaviors do not go with signal.  Next, the dog must also learn which sounds are not the one that pairs with “come when called”.

A. Make the signal easier:

1. Use short crisp, distinct sounds. Repeat the word pairing lesson 20 times in each training as one set. Only add difficultly once dog has mastered “come” in the quiet indoors and many rooms. Then practice  in different locations, start 6 inches away and increase distance slowly, different peoples voices, with background noises, distractions (fun and scary).

2. Start with an easier signal than sound.  Hand or arm signals are easier  for dogs because so much of their communication is made up of complex body language.  Dogs watch you closely every movement you make in an attempt to pick up on a signal.  Make it easier by using hand/arm signals. Keep the rest of your body still. Use one hand/arm. Add sound after the dog is beginning to do it how you want it.

B. The sequence of learning is backwards from the order of  doing the final behavior.

Let me explain further….

Dogs learn to use a set of behaviors by trial and error. So if a behavior is followed by something the dog likes, they attempt the behavior again. That is why treats work so quickly with training, especially in puppies. Timing of the reward is very important because dogs are doing behavior all the time. The closer the timing of the dog’s behavior are to the arrival of the treat the easier it is for the dog to understand what they did that resulted in treat delivery. The dog will experiment with body movements to figure out which sequence or posture or behavior gets the reward.

The first part of the video above suggests pairing the sound “come” prior to something the dog likes. He is taking advantage of a dogs natural tendency to  approach when we make any sound.  It seems simple but the thinking is a 3 step process. The trainer is ready for the dog who will approach, he will give something the dog likes, plus he adds the word “come” before the dog gets to him. This is a great way to introduce the idea of “come”  when you first get a new puppy.  Something the dog likes can be: praise, ear rubs, and treats whenever the puppy comes to you for any reason. Change up the rewards, so eventually food will not be needed. Make “come” part of everyday life.

C. Make the signal a signal.

The first time you hear a  siren it has no meaning. It is just an annoying noise.We watch our parents pull to the side of the road as the noisy vehicle goes by. Later on we learn as someone explains. The signal does not make us drive to the shoulder of the road it is the order that we have learned to do when we hear the sound signal. So it is with dogs. A signal is a good one if it clearly helps the dog predict what to do next to get the final outcome. The sequence is how mammals learn by trial and error.  Called the ABC”s of learning. Take advantage of this way of thinking to speed up your training and solve the stumbles that happen along the way to well trained dog.

The sequence is how a behavior happens in real time.

A stands  for antecedent. Which simply means whatever comes before the behavior.

B stands for the observed or goal behavior. These are not always the same.

C stands for consequence. Here in is the biggest secret to good trainers. They all know it is “C”, the consequence that determines future behavior. If the consequence is a good one, the dog will try out the behavior again to see if the same consequence will follow. The behavior is a guess at how to get the consequence again.

The A is not always a signal as we wish it to be. It can be any set of circumstances, the dog recognizes, that happened before a they behavior worked to get the big C.

But training happens backwards from execution of a already trained behavior. The sequence of thought is: planning the big C and delivering it close to the behavior.  The dog will try out behaviors, until they suceed  at getting the big C.   The dog also observes what precedes the sequence, B->C trying to figure out when doing a certain behavior gets the big C. The dog will bet on your signal,  do the behavior, in the hope of getting the big C.

So training involves manipulating these three things. We add clarity and reliability to the Antecedent which helps the dog recognize it as a signal. We are clear about which behaviors get the rewards and when. We add good quality rewards for the behaviors that we want the dog to do. The dog becomes reliable the more reliably the consequence happens when the dog gives us the asked for behavior.