How to Teach Your Dog to Stay

A dog that holds any position — sit, down, stand — until you say they're done, built without ever needing the word "stay."

You may have tried teaching "stay" as its own command and ended up repeating "stay... stay... stay..." while your dog creeps toward you or pops up the second you turn around. That frustration is real, and you're not doing it wrong — the approach itself is the problem. The method below builds duration into every position command your dog already knows. No new word needed. When something goes wrong, we tell you exactly what happened and how to fix it.

Before you start

Your dog should already sit or lie down on a verbal cue. If they can't do that yet, start here — it's fast and this page will be waiting for you.

Find your usual quiet training spot. Have 15-20 small treats ready. Ask your dog to sit. If they sit and look at you calmly — relaxed body, no whining, no fidgeting — you're in the right spot. If they sit but immediately pop up, paw at you, or spin in circles, the environment may be too stimulating or your dog may need a walk first. That reaction is useful information, not a reason to push through.

Why some dogs can't settle in certain environments →

Building Duration

Step 1: Mark the hold, not the position

Ask your dog to sit. They sit. Instead of saying "yes" immediately (which would release them), wait one second. Then say "good" in a calm, low voice — and give a treat while they're still sitting.

"Good" is different from "yes." "Good" means that's correct, keep doing it. "Yes" means you're done, you can move. This distinction is the entire foundation of duration work.

If that worked: Repeat 5-6 times. Ask for a sit, pause one second, say "good," treat in position. After the last rep, say "yes" — this is the release. The dog is now free to move. Pay that one generously.

Dog pops up the instant you give the treat?

The dog is treating "good" + treat as the end of the exercise, the same way they treated "yes" + treat during sit training. That's a reasonable assumption on their part — until now, food always meant the behavior was done.

Try this: Deliver the treat to the dog's mouth while they're still in position. Don't toss it, don't hold it in front of them — bring it directly to them so they don't need to shift their body. If they stand to take it, reset. Ask for the sit again, say "good," and feed more carefully. After 2-3 treats delivered in position, say "yes" to release. The dog will learn that "good" means stay put, and "yes" means you're free.

Dog seems confused by "good" — stares at you or starts offering other behaviors?

If your dog hasn't heard a continuation marker before, "good" is just a new sound with no meaning yet. The dog may be trying to figure out what you want by offering sits, downs, paws, or barks.

Try this: Simplify. Ask for the sit. The instant they sit, say "good" calmly and immediately feed a treat while they hold position. Don't wait even a second yet. Repeat 8-10 times: sit, "good," treat in position. You're conditioning "good" to mean "you're on the right track and food is coming — don't move." Once the dog visibly relaxes when they hear "good," start adding the one-second pause.

Step 2: Stretch the pause

Same as Step 1, but now wait two to three seconds before saying "good" and treating. The dog sits. You stand there calmly. Two seconds. "Good." Treat in position. Then release with "yes."

Build this gradually over a few sessions:

Don't increase every rep. Mix it up. Two seconds, then five, then three, then eight. Unpredictability keeps the dog engaged because they can't predict when the reward is coming.

What success looks like: The dog sits, watches you calmly, and doesn't shift or fidget for the full pause. Their body is relaxed, not tense.

Dog holds for a few seconds but then breaks before you mark?

You pushed the time too far too fast. The dog's ability to hold is shorter than what you're asking for, so they give up and move.

Try this: Go back to the last duration that was reliable. If the dog was solid at three seconds but broke at eight, work at three to five seconds for a full session. End on a success. Next session, try six. Build in half-second or one-second increments, not leaps. Duration is patience work — for both of you.

Dog is trembling, lip-licking, or yawning during the hold?

These are stress signals, not defiance. Your dog is uncomfortable with the exercise. This can happen if the dog doesn't understand what's expected, if the duration is too long for their current ability, or if something in the environment is creating anxiety.

Try this: Shorten the duration dramatically — back to one or two seconds. Mark "good," treat, release with "yes," and let the dog move around. Keep the session very short (5-6 reps). If the stress signs continue even at one second, take a break entirely. Play with your dog. Come back to it later or the next day. The goal is a dog that holds position because they're relaxed and confident, not because they're frozen in worry.

Quick check: Is your dog settling into the hold or bracing through it? There's a difference. A settled dog has a loose body, soft eyes, and may even shift their weight to get comfortable. A bracing dog is stiff, staring, and looks like they're about to explode out of position. If you're seeing bracing, the duration is too long or the exercise is too new. Shorten the time and let them succeed easily for a while.

Step 3: Add movement — yours, not theirs

This is where most people run into trouble. Until now, you've been standing still in front of your dog. Now you're going to shift your weight, take a half step to the side, and come right back.

Ask for a sit. Wait two seconds. "Good." Treat in position. Now shift your weight to one foot. "Good." Treat. Step one foot to the side and bring it back. "Good." Treat. Release with "yes."

The key principle: When you change what your body is doing, reduce the duration. You are now asking for two things at once — holding position AND ignoring your movement. Drop the time back to what's easy while you layer in the new variable.

Dog breaks the instant you shift your weight or step sideways?

Your dog has been reading your body this entire time. When you shift your weight, the dog reads that as "we're about to do something" — because up until now, your body movement has always signaled the end of an exercise or the start of something new. The dog isn't disobeying. They're responding to your body language the way dogs always do.

Try this: Make your first movement almost invisible. Don't step — just shift your weight slightly to one side while staying planted. Say "good" and treat before the dog has a chance to react. Repeat until the dog doesn't flinch at your weight shift. Then take a small step with one foot and immediately bring it back. "Good," treat. You're teaching the dog that your movement doesn't mean anything has changed for them. Build to a full step to the side, then two steps, over multiple sessions.

Important: Don't increase your movement AND the duration at the same time. If you're adding a step to the side, keep the hold time short. One new variable at a time.

Dog follows you when you step away?

The dog thinks stepping away is an invitation to come with you. From their perspective, you've always been right there, and when you leave, the logical response is to follow. This is especially common in dogs with strong engagement — they want to be near you.

Try this: Take your step and immediately return before the dog decides to follow. Say "good" the instant you're back in position, treat. If the dog has already followed, don't scold — ask for the sit again from where you both are, succeed at a shorter distance, and build from there. You can also practice with the dog on a leash attached to a doorknob or heavy furniture. The leash gently prevents the follow so the dog has a chance to discover that staying in place is what pays. Just make sure you return and reward before the dog hits the end of the leash and gets frustrated.

Step 4: Build real distance

Once your dog can handle you taking a step or two without breaking, you're ready to build actual distance. This follows the same pattern: increase one thing, hold everything else steady.

  1. Start close. Take one step away, return immediately. "Good." Treat in position.
  2. Take two steps away, return. "Good." Treat.
  3. Take three steps, pause one second at the far point, return. "Good." Treat.
  4. Gradually increase the number of steps and the pause at the far point.

Always return to the dog to deliver the reward. Don't call them to you — that teaches them that the hold ends when you call, and they'll start anticipating.

Dog holds when you walk away but breaks when you turn your back?

Turning your back removes the eye contact and body language the dog has been using to read you. For the dog, your face disappearing is a significant change — it's like the conversation suddenly ended.

Try this: Turn partially — a quarter turn — and immediately turn back. "Good." Treat. Build up to a half turn, then a full back-turn, over several sessions. Each time, reduce the distance and duration to make the turn itself the only challenge. Once the dog can handle you turning away at three feet, combine it with the distance you've already built.

Dog holds at home but breaks at the front door before walks?

This is a threshold and generalization issue disguised as a training problem. The front door before a walk is one of the most exciting moments in your dog's day. Arousal is sky-high, and the behavior your dog learned in the calm living room hasn't been generalized to this context yet — their brain is buzzing with anticipation and the skill simply doesn't exist here.

Try this: Don't start duration work at the front door. That's the final exam, not the first lesson. Instead, practice sits with brief holds in gradually more exciting contexts: in the living room with your shoes on, then near the hallway, then facing the door (but not opening it), then with your hand on the doorknob. Each stage is its own training session. When you finally reach the door, the hold time should be very short — one to two seconds — and the reward for holding should be the walk itself. Sit, hold, "yes," door opens. Over time, stretch the hold at the door. But getting there too fast will only teach the dog that the front door is a place where rules don't apply.

Quick check: How is your dog's body language during the holds at distance? A dog who is comfortable will look relaxed — maybe even bored. A dog who is stressed will be stiff, panting, or fixated on you with hard eyes. If you're seeing stress at a particular distance, you've gone too far. Cut the distance in half and rebuild. The emotional quality of the hold matters as much as the duration.

Step 5: Add duration at distance

Now you're combining what you've built: the dog holds position while you're several steps away, for a meaningful stretch of time.

Build this the way you'd stack blocks:

  1. Three steps away, two-second pause. Return. "Good." Treat.
  2. Three steps away, five-second pause. Return. "Good." Treat.
  3. Five steps away, three-second pause. Return. "Good." Treat.
  4. Five steps away, eight-second pause. Return. "Good." Treat.

Mix it up. Don't always make it harder. Throw in easy reps — one step, one second — so the dog keeps winning. The pattern they should learn is: hold position, the person comes back, good things happen.

When you're ready to finish: Walk back to the dog, say "good" and treat one final time, then say "yes." That's the release. The exercise is over and they're free.

Dog seems to have a time limit — holds for 15 seconds but always breaks around 20?

Every dog has a current ceiling, and it's not fixed. The dog consistently breaks at the same point because that's where their confidence or focus runs out.

Try this: Don't keep testing the ceiling. Instead, work well below it and make those reps deeply rewarding. If the dog breaks at 20 seconds, spend a week doing rock-solid 10-12 second holds with multiple "good" + treat deliveries during the hold. Then push to 15. Then 18. The ceiling lifts when the dog's confidence in the exercise grows — and confidence grows from repeated success, not repeated failure.

You keep repeating "stay stay stay" and the dog gets up anyway?

This is one of the most natural things owners do, and it works against you. Every time you say "stay" (or the dog's name, or "wait, wait"), you are producing a new sound that the dog has to process. Each new sound is a small event, and events cue the dog to do something — usually to look at you, shift their body, or move. You are inadvertently giving the dog a reason to break position by talking.

Try this: Say nothing. The original command ("sit" or "down") is the only instruction the dog needs. Silence is your friend during the hold. Use "good" sparingly — once or twice to mark that they're on the right track, followed by a treat in position. Resist the urge to narrate. If the dog breaks, simply reset: ask for the position again and start a shorter hold. No scolding, no "I said stay." Just reset and make it easier.

Step 6: Add mild distractions

Once the dog can hold position while you move around at a reasonable distance for 20-30 seconds in your quiet training spot, introduce a small distraction. Not another dog, not the doorbell — something minor. Drop a pen on the floor. Shuffle your feet. Cough.

Ask for the sit, build to a short hold, then introduce the distraction. If the dog holds, "good" and treat. If they break, the distraction was too much. Go back to a simpler distraction or shorter distance.

Dog holds through sounds but breaks the moment something moves?

Visual movement triggers a different response than sound. Movement activates drive — the dog's natural urge to chase, investigate, or follow things that move. Sounds are easier to ignore because they don't engage the same chase instinct.

Try this: Start with slow movement. Slowly raise and lower your arm. Slowly shift a chair. If the dog holds, mark and reward. Build to faster movements over time. If the dog breaks at any point, make the movement slower or more distant. You're teaching the dog that movement in the environment doesn't change their job.

Dog holds when you're quiet but breaks when someone else enters the room?

A new person is a major environmental change — new smells, new sounds, new social opportunity. This is a big jump in difficulty, and it's normal for the dog to lose focus.

Try this: Have the other person enter the room calmly while you're standing close to the dog (not at a distance). Keep the hold short — two to three seconds — and mark and reward immediately. If the dog can't hold even that, have the person stand in the doorway without entering. Build from there. The other person's presence is the variable — keep distance and duration short while the dog adjusts to it.

Quick check: How many variables are in play right now? If you're working on distance AND duration AND a new distraction, that's three things at once — and that's too many. If the dog is struggling, ask yourself which variable you can remove. Drop back to just one new thing at a time.

If this isn't clicking yet

Some dogs learn to hold position in a few sessions. Some need a couple of weeks of daily practice. Both are normal. A dog that learns duration slowly is not a stubborn dog — they're a dog that needs more clarity and more repetition at each stage.

If you've tried the adjustments above and you're still stuck, the issue is almost always one of these three things:

  1. The environment is too hard. Even your living room may be too much if other pets are around, kids are playing, or the TV is on. Try a bedroom with the door closed and nothing happening. Duration requires the dog to be calm — and calm requires a calm environment.
  2. The reward isn't competing. If what's in the environment is more interesting than what you're offering, the dog has no reason to stay put. A higher-value treat — real meat, cheese, something rare — can shift the equation. The dog decides what's valuable, not you.
  3. Your dog needs a different approach. Some dogs experiment and offer behaviors — they try things until something works. Others wait and need to be guided. If your dog breaks position and just stares at you, luring them back into the sit with a treat and marking the hold works well. If your dog is fidgety and keeps offering random behaviors (pawing, barking, spinning), be patient — shorten the hold, mark the moment they're still, and build from there.

These aren't character flaws in your dog. They're variables. Adjust the variables.

Making it harder (gradually)

Once your dog holds position reliably in your quiet training spot with you moving around, it's time for generalization — teaching the dog that this skill works everywhere, not just in your training spot. Change one variable at a time:

The single biggest mistake in this phase is changing too many things at once. If you're practicing in a new location for the first time, that IS the variable. Keep your distance close, your duration short, and your distractions minimal. Let the dog succeed in the new place before layering anything on top of it.

Real-world goals you can build toward, one at a time:

Each of these is an endpoint, not a starting point. Get there in stages.

If the dog fails, you moved too fast. Go back to the last point where they succeeded and stay there longer. That's not a setback — it's the dog telling you where their current limit is.

Why this works

Duration holds when the handler's markers are steady, calm, and consistent. If your "good" sounds different every time — rushed one moment, bored the next, anxious the next — the dog loses the information that keeps them in place. Your consistency is what makes the continuation marker work.

The continuation marker "good" gives the dog real-time information: you're doing the right thing, keep doing it. Without that feedback, the dog is guessing. With it, they know exactly what's working — holding still — and they know more reward is coming if they continue. This is why the method doesn't need "stay" as a separate command — "sit" means get into position, "good" means you're on track, "yes" means you're done. Fewer signals, clearer communication.

How Learning Works → — the full picture of how dogs form these connections The Tools → — markers, luring, and the release word explained in depth

Go deeper