Last updated: May 23, 2026
In this article
What the 3D Stay Framework Is
The 3D framework — Duration, Distance, Distraction — is the foundational training structure that force-free trainers use to build a reliable stay. Pat Miller (Peaceable Paws) and Karen Pryor (the modern clicker-training tradition) both teach the 3D model, and most credentialed CCPDT, IAABC, and KPA-CTP trainers operationalize it the same way. The framework names the three independent dimensions that make a stay difficult: how long, how far away, and how distracting.
The single most important rule in the 3D framework is that the three dimensions are built one at a time, never combined on the same repetition. If you raise duration on a rep, you ease (or hold) distance and distraction. If you raise distance, you ease duration and distraction. If you raise distraction, you ease duration and distance. Owners who try to push all three at once consistently produce broken stays, frustrated dogs, and stalled training.
The framework applies to sit-stay, down-stay, stand-stay, place / mat duration, and any other duration-and-position behavior. It is also the foundation for more advanced obedience and sport behaviors — competition-level stays, recall reliability, and off-leash work all build on the same 3D framework.
The Foundational Position
Before any stay work, the position cue itself must be solid. The dog must reliably sit (or down, or stand) on cue, in low-distraction environments, with food reward following. The step-by-step sit training and step-by-step down training protocols cover the foundation work that stay-training builds on.
For most pet dogs, down-stay is easier to maintain for long durations than sit-stay. The down is a more relaxed position physically and the dog is less likely to drift. For shorter stays in public settings, sit-stay is often more practical because it is visually clearer and the dog can re-orient faster. Both have their place.
Add the stay cue (“stay” or a hand signal — open palm facing the dog) once the dog reliably holds position for 2 to 3 seconds after the position cue. The stay cue should be its own discrete signal, not an extension of the position cue. The stimulus-control and cue-transfer framework covers the broader theory for cue isolation.
Building Duration First
Duration is the first D. Start at 1 to 3 seconds. Cue stay, wait 1 to 3 seconds, then deliver the reward to the dog while the dog is still in position. Do not call the dog to you for the reward — the reinforcement happens AT the position. This teaches the dog that staying-in-place produces food, not that returning to the handler produces food.
Build duration in tiny increments. From 3 seconds, go to 5 seconds. From 5, to 8. From 8, to 12. From 12, to 20. From 20, to 30. Reward at random durations within the build range so the dog cannot predict when the reward arrives — variable reinforcement schedules create more durable behavior than fixed schedules.
Add a release cue (“OK,” “free,” “all done”) that explicitly ends the stay. The dog learns that the stay continues until the release cue is given. Many broken stays come from dogs who self-release because the handler never built a clear release signal. The release cue and the stay cue together define the duration boundary.
Adding Distance Second
Once duration is solid at 30 to 60 seconds in low-distraction environments, add distance. Cue stay at 1 foot, take one step back, return immediately, reward. Then 2 feet. Then 3 feet. Then 5 feet. Then across the room. Then to the doorway. Then around the corner (out of sight briefly).
When you raise distance, drop duration. A dog who reliably holds a 60-second stay at 1 foot may not hold a 60-second stay at 10 feet — the distance itself is a new challenge. Start the new distance at 5 to 10 seconds and rebuild duration at the new distance. This is the classical Pat Miller / Karen Pryor rule applied operationally.
Out-of-sight stays are an advanced extension and not required for most pet contexts. AKC CGC tests up to 20 feet of distance with the handler in view. AKC obedience competition tests longer durations and out-of-sight stays. For most family dogs, 30-second stays at 10 to 15 feet are the practical ceiling.
Adding Distraction Third
Distraction is the most variable D and the one most owners underestimate. Distractions come in many categories: visual (people walking, dogs passing), auditory (doorbells, doors closing, household appliances), olfactory (food on the ground, scent of another animal), and motion (a thrown toy, a rolling ball).
Build distraction by category, lowest intensity first. For visual distractions, start with a single person walking calmly past at 10 feet, then 5 feet, then directly in front of the dog. Reward heavily for sustained stay. For auditory, start with a single quiet sound, then add volume. For food on the ground, start with low-value food at a distance, then closer, then more valuable food.
When you raise distraction, drop duration and distance back to easy levels. The classical 3D rule: never raise more than one D at a time. A dog who holds a 30-second stay at 10 feet in a quiet room will often fail a 30-second stay at 10 feet with a person walking past. Drop to 5 seconds at 3 feet with the distraction, succeed at that level, then build duration back up at the new distraction level.
Generalization Across Environments
The 3D framework operates within a single environment. Generalization extends the framework across environments. A stay that is rock-solid at home rarely transfers cleanly to the backyard, the park, or a friend’s house. Each new environment is a fresh training problem, and the 3D build often has to be partially redone in each context.
The generalization and proofing protocol covers the broader framework for moving cues between environments. The short version: practice in at least 5 to 8 distinct environments before treating a behavior as fully generalized. Dogs are poor generalizers — Pat Miller and Susan Friedman both stress that “the dog isn’t being stubborn; the cue hasn’t been generalized.”
For dogs being prepped for the AKC Canine Good Citizen test, generalization is the single biggest predictor of pass / fail. The stay items (sit-stay, down-stay, stay-while-handler-leaves) are tested in a novel environment with strangers present — exactly the generalization scenario most home-trained dogs fail.
Force-Free Method and Why It Matters
The AVSAB Position Statement on Humane Dog Training and the AVSAB Position Statement on the Use of Punishment for Behavior Modification in Animals both name reward-based methods as the recommended approach. For stay training specifically, force-free methods produce a relaxed, willing stay; aversive correction methods produce a tense, anxious stay that often breaks under unfamiliar pressure.
The LIMA (Least Intrusive, Minimally Aversive) hierarchy places positive reinforcement of the desired behavior as the primary tool for stay-training. CCPDT, IAABC, KPA, and APDT credentialed trainers teach stay through reinforcement of duration, not through correction of broken stays. Owners ultimately choose; we recommend the credentialed-positive-first path.
Marker training accelerates stay-acquisition substantially. A clicker or verbal marker (“yes”) precisely identifies the moment of the held position, which the dog can then connect to the reward. The place and mat duration training protocol applies the same 3D framework to mat-based duration work.
Common Mistakes and Fixes
Mistake: rewarding the dog from your hand when you return. Fix: reward AT the position, before any release. Returning-to-handler-for-reward teaches the dog to break stay to earn the food. Reward placement defines the behavior the dog repeats.
Mistake: skipping the release cue. Fix: always explicitly release the stay. “OK,” “free,” or “all done” — pick one and use it consistently. Without a release cue, the dog cannot tell where the stay ends, and self-releases become the default.
Mistake: pushing duration before the position is solid. Fix: confirm the dog reliably enters position on cue before adding stay duration. A wobbly sit cannot support a 30-second stay; rebuild the sit first.
Practical Applications in Daily Life
Stay is the cue that makes door management, leash-on management, food-preparation management, and vet-waiting-room management possible. The same 3D framework that builds the CGC down-stay also builds the door-opens-but-dog-waits behavior, the dinner-table-stay behavior, and the leash-clip-without-spinning behavior.
For dogs in HDB or condo settings, stay-on-mat in the lift lobby or void deck is a daily management tool. For dogs working through the recall mastery and distraction proofing protocol, stay is the pre-cue that lets you build distance before calling the dog back. The two protocols pair tightly.
For competition or sport-track owners, stay underpins almost every advanced obedience exercise. AKC Obedience trials (Novice, Open, Utility) and AKC Rally all rely on reliable position-and-duration stays as core building blocks. The 3D framework scales from pet-dog basics to top-tier competition.
Frequently Asked Questions
How long should a stay be?
For pet dogs, a 30 to 60 second stay at 10 to 15 feet in moderate distraction is the practical ceiling and covers nearly all daily-life situations. For AKC CGC, 20 feet for the duration of the item (typically 30 to 60 seconds) is the standard. Competition obedience tests longer durations with handler out of sight.
Should I use a verbal cue or a hand signal?
Both work. Most owners pair them initially (verbal cue + open-palm hand signal) and later use either independently. For deaf dogs or in noisy environments, hand signals are essential. For competition obedience, verbal cues alone may be required at higher levels.
My dog breaks stay every time I leave the room. What do I do?
Out-of-sight distance is a major step. Drop duration to 2 to 5 seconds and build up gradually with you just stepping briefly out of sight. The dog needs to learn that out-of-sight does not mean the handler is gone. This often takes weeks of patient build-up.
What if my dog only holds stay when I have treats visible?
Move to intermittent reinforcement — reward every 3rd or 5th stay, then variable. Keep treats out of sight (in a pouch or behind you). Reward unpredictability creates more durable stays than predictable food-visible reward patterns.
Can a senior dog learn stay?
Absolutely. Senior dogs often learn duration behaviors more reliably than puppies because the calm-and-still default is closer to their natural state. Down-stay is particularly comfortable. Adapt session length to the dog’s energy and physical comfort.
