Stay is one of the more genuinely useful obedience commands a dog can learn, and also one of the most commonly taught incorrectly. Most owners try to build a long, rock-solid stay too fast, in too distracting an environment, and without a clear signal for when the stay actually ends. This guide breaks stay down into its real building blocks: why the command needs an explicit release word, how to build duration, distance, and distraction as three separate skills rather than one big leap, and a distinction most competing guides skip entirely, the difference between "stay" and "wait." For the bigger picture on where stay fits among other commands and how long training realistically takes, see teaching a dog basic obedience commands.
Why stay needs its own release word
A stay is not finished until the handler says it is finished, and that requires a clear, consistent release word or phrase, something like "okay" or "free," used only to end a stay and nothing else. Without a dedicated release word, a dog has no reliable way to know whether standing up was correct or a mistake, which makes the entire command inconsistent from the dog's point of view. Many owners accidentally release a dog from a stay just by relaxing their posture, breaking eye contact, or walking toward the dog, all of which a dog reads as cues that the exercise is over even when the handler did not intend to end it. Picking one specific word, using it every single time a stay ends, and never using that same word for anything else in daily life, is a small detail that makes an enormous difference in how quickly and reliably a dog learns the command.
The three Ds: duration, distance, and distraction
The most common mistake in stay training is trying to build all three of the meaningful variables (how long the dog holds the position, how far the handler moves away, and how many distractions are present) at the same time. Increasing more than one of these at once is usually what causes a dog to break a stay that seemed reliable a session earlier. Instead, work through them roughly in this order:
- Duration first. With the dog in a sit or down right next to you, ask for a stay and reward it after just one or two seconds, gradually working up to ten, then thirty seconds, then longer, all without moving away from the dog at all.
- Distance second. Once a dog can hold a stay next to you for at least thirty seconds, start adding small amounts of distance, one step back, then two, returning to reward the dog for holding position rather than calling the dog to you (calling ends the stay and rewards movement, not stillness).
- Distraction last. Only once duration and distance are both solid in a quiet room should real-world distractions get added: another person walking by, a dropped object, a door opening. Distraction is consistently the hardest of the three for most dogs and deserves the most patience.
Trying to combine all three too early, for example walking ten feet away for a full minute during a first attempt at a park, is a near-guaranteed way to see the stay fall apart. Go back to whichever variable is causing the break, reduce it, and rebuild from there rather than repeating the same failing attempt.
The Cornell distinction most guides miss: stay versus wait
Most obedience guides treat "stay" as the only hold-still command a dog needs, but Cornell University's Riney Canine Health Center draws a genuinely useful distinction between two separate cues that most competing content collapses into one. "Wait" means holding back temporarily before being released, useful for moments like carrying groceries through a door or opening a crate door without the dog bursting out; it does not matter what position the dog is in, and the hold does not need to last long. "Stay," by contrast, means holding a specific position exactly where the dog was left until the handler returns or gives the release word, useful for situations like sitting calmly while guests arrive or staying put while an owner crosses the street to check the mail. Teaching these as two distinct cues, rather than using "stay" loosely for both situations, gives a dog a clearer and more specific understanding of what is actually being asked in each case, which in turn makes both commands more reliable. This batch's companion post on teaching a dog to wait at the door covers the wait side of this distinction in full.
Step-by-step teaching method
Start with your dog in a sit (stay also works from a down, which many dogs find more comfortable to hold for longer periods). With the dog sitting, say the cue word "stay" once, along with an open-palm hand signal held toward the dog, then wait one or two seconds before marking with a clicker or verbal "yes" and delivering a treat while the dog is still in position. Critically, deliver the reward to the dog while it is still holding the stay; do not call the dog toward you to give the treat, since that teaches the dog that moving toward the handler is what earns the reward, undermining the entire point of the exercise. Repeat several short stays like this before adding any distance or duration increase, then use the release word to formally end the exercise every time, followed by praise or a second reward for the release itself.
Building real duration without setting the dog up to fail
A useful rule when increasing duration is to vary the length of each individual stay rather than always making each one longer than the last. If every single repetition is longer than the one before it, a dog quickly learns that holding still gets harder and harder every time, which can cause anticipatory breaking. Instead, mix in easy short stays between the longer ones, so the dog cannot predict exactly how long any given stay will last and stays engaged rather than bracing for an ever-increasing challenge. Return to reward the dog periodically during a longer stay, not only at the very end, especially early in training, so the dog is reminded that staying still is actively paying off throughout the hold, not just at the finish.
Common reasons a stay breaks down
A dog that repeatedly breaks a stay is usually not being stubborn; something about the criteria has moved too fast. The most frequent culprits are increasing distance and duration at the same time, rewarding inconsistently so the dog is not sure staying is worth it, using body language that accidentally signals release (turning away, walking toward the dog, relaxing posture), or simply asking for more than the dog is currently capable of holding given the distraction level in that specific environment. A dog that is anxious, fearful, or highly aroused by something in the environment (another dog, a loud noise, excitement about a person approaching) will also struggle to hold a stay regardless of how well it has been trained in calmer settings, since impulse control is a finite resource that gets used up faster under stress. If a stay that was previously reliable suddenly falls apart across multiple sessions, it is worth stepping back to an easier version of the exercise rather than repeating the failing attempt, and considering whether pain, anxiety, or a genuinely new distraction is the real cause.
Duration, distance, and distraction progression
| Stage | Duration | Distance from dog | Distraction level |
|---|---|---|---|
| 1 | 1 to 5 seconds | 0 feet (standing next to dog) | Quiet room, no distractions |
| 2 | 10 to 30 seconds | 0 feet | Quiet room |
| 3 | 10 to 30 seconds | 2 to 5 feet | Quiet room |
| 4 | 30 to 60 seconds | 5 to 10 feet | Familiar room with minor activity |
| 5 | 1 to 2 minutes | 10 feet or more, out of sight briefly | Yard or quiet outdoor space |
| 6 | Varies, real-world use | Varies | Busier settings: front door, public spaces |
Each stage should feel easy before moving to the next one; if a dog is failing more than roughly one in five attempts at a given stage, that is a signal to stay at the current stage longer rather than advance.
Using stay safely and realistically
Stay is genuinely useful for managing a dog around an open door, a nervous guest, a toddler, or a moment where the handler needs both hands free, but it should never be treated as a substitute for supervision or a physical barrier in a truly high-risk situation, like a dog near a busy street or an aggressive dog near another animal. A trained stay is a strong management tool, not a guarantee, and even a well-trained dog can break a stay if startled or sufficiently provoked. Combining a solid stay with sensible management, like a leash, a gate, or a closed door, gives the best real-world safety margin rather than relying on the command alone. It is also worth building the release word into daily routines outside of formal training sessions, so the dog hears and responds to it in contexts beyond a dedicated practice block, which strengthens the reliability of the whole stay-and-release pattern over time.
When to bring in professional help
Most dogs can build a genuinely useful stay through consistent, reward-based home practice over a few weeks. If a dog is not progressing at all after several weeks of correctly sequenced practice, seems distressed or panicked rather than simply excited during stay attempts, or the household needs a stay specifically to manage a reactive or fearful dog around triggers, a trainer certified through the Certification Council for Professional Dog Trainers can usually diagnose what is going wrong faster than continued solo troubleshooting, and for dogs with significant anxiety or reactivity, a consultant listed with the International Association of Animal Behavior Consultants or a veterinary behaviorist is the more appropriate referral. As with every command in this series, the AVSAB's position on humane dog training applies here too: reward-based methods, not corrections or aversive tools, are the approach with real evidence behind it.
Which position to stay from: sit or down
Stay can be taught from either a sit or a down, and most dogs benefit from learning it in both positions eventually, since real-life situations call for different ones. A sit-stay is useful for short holds, like waiting at a curb or holding still for a leash to be clipped on. A down-stay is generally more comfortable for a dog to hold for longer stretches, since lying down uses less muscular effort than sitting upright, which is why many trainers default to teaching a down-stay first for any stay expected to last more than a minute or two. Older dogs and dogs with any joint sensitivity in particular tend to find a down-stay noticeably easier to sustain than a long sit-stay, so it is worth teaching both and choosing whichever position suits the specific situation and the individual dog's comfort.
How stay connects to other commands in this series
Stay is one of the more foundational cues in a well-rounded obedience routine, and several other commands either build on it directly or work alongside it. The place command is, in a sense, an extended and location-specific version of a down-stay, asking a dog to go to a mat or bed and hold a settled position there for a longer period, often while the handler moves around the house rather than staying nearby. A reliable stay also makes leave it easier to layer in later, since both commands rely on the same underlying skill of a dog choosing to hold back rather than act on an impulse, even when something interesting is happening nearby. For a full sense of how stay fits into a realistic month-by-month or command-by-command training timeline, how long it actually takes to train a dog walks through typical timeframes for stay alongside every other command in this series.