Every training plan has slow weeks. A dog that was reliable at home suddenly hesitates in the garden, a recall that was improving plateaus for a fortnight, a settle routine that was going well has an off few days. None of that means the plan has failed. The harder judgement call is recognising when a plan has genuinely stalled, rather than simply needing more repetitions, and working out what to actually change rather than concluding the whole approach does not work.
This guide sets out how to tell the difference between a plateau and a genuine stall, the reasons plans commonly stop working, what to try before switching approach entirely, and how to judge fairly whether a different method or a different trainer is actually warranted.
Plateau or genuine stall
A plateau is a period with no obvious improvement using a measure you can actually track: the dog holds a settle for the same length of time across several sessions, or a recall works at the same distance it worked at last week, without getting worse. That is normal. Skill acquisition in dogs, as in most learning, does not move in a straight line, and a week or two of no visible gain, particularly after a jump in difficulty, is expected rather than alarming.
A genuine stall looks different in three ways. First, there is no measurable progress across several sessions using a concrete criterion, not a general feeling that things “aren’t clicking.” Decide in advance what you are actually measuring, whether that is duration in seconds, distance in metres, or the presence or absence of a specific unwanted behaviour, and track it. A feeling of frustration is not evidence either way. Second, a genuine stall often shows regression rather than a flat line: the dog was managing fifteen seconds of eye contact around distraction and is now managing five, repeatedly, rather than holding steady. Third, watch the dog’s affect, not just the behaviour. A dog that is gradually improving usually looks more relaxed and more willing to engage over time. A dog showing increasing stress, avoidance, shutting down, or reluctance to start a session at all is telling you something is wrong with the plan itself, even if the target behaviour occasionally still appears.
Common reasons a plan stalls
Difficulty raised too fast is the single most common cause. A structured plan increases duration, then distance, then distraction, one variable at a time, as described in structure, boundaries, rewards and correction. Skipping straight from a quiet room to a busy park, or adding a new distraction before duration is solid, asks the dog to solve two or three problems simultaneously and often produces exactly the plateau or regression described above.
Reinforcement that is not actually rewarding to this particular dog in this particular context is another frequent cause. A dog that will work happily for kibble at home may need higher-value food, a toy, or access to sniffing outdoors where competing smells make kibble far less interesting. What counts as reinforcing also changes with arousal and stress: a treat that works when a dog is calm may be ignored entirely when the dog is anxious or overstimulated, which is not the dog being stubborn, it is the reinforcer no longer functioning as one in that state.
The environment can simply be too hard for where the dog currently is, independent of any single training error. A recall that has only ever been practised in a fenced garden is being tested, not trained, the first time it is tried at an off-lead park with other dogs running past. Inconsistency between the people practising the plan also stalls progress quickly: if one household member releases the dog from a stay early and another insists on holding it, or one person allows jumping up as a greeting and another does not, the dog is working with contradictory information rather than a clear pattern.
It is also worth asking honestly whether the behaviour was ever actually taught, or only prompted. A dog that reliably sits when a treat is held above its nose has learned to follow a lure, not necessarily to respond to the word “sit” on its own. Moving too quickly from luring to a verbal or hand cue, without deliberately fading the lure and testing the cue alone, can look like a stalled plan when the real issue is that the underlying behaviour was never fully taught in the first place, only prompted into existing.
Finally, rule out a physical or medical cause, particularly when a previously reliable dog suddenly struggles with something it used to do easily. Pain, reduced hearing or vision, hormonal changes, or general ill health can all reduce a dog’s ability or willingness to perform a learned behaviour. A dog that starts refusing to sit, hesitating on stairs, or struggling to hold a down for as long as it used to is giving a signal worth taking to a vet before assuming the training has gone wrong.
What to try before abandoning an approach
Drop back to an easier version of the exercise and rebuild more gradually, rather than repeating the same failing rep at the same difficulty. If a recall has stalled at six metres with one dog visible in the distance, go back to three metres with no other dog present, get several confident successes, and increase far more slowly than before.
Change the reinforcement being used. Trial higher-value food, a different toy, or a different delivery style, such as feeding in a steady stream rather than a single piece, before concluding the dog is not motivated to work at all. Change the environment: move a session that has stalled in one location to a different, easier location and see whether progress resumes, which tells you the problem was environmental difficulty rather than the exercise itself.
Get a second pair of eyes. An experienced trainer can often spot handler-side issues that are invisible to the person doing them: cueing a fraction of a second late, tensing the lead just before a reactive moment, or rewarding after the dog has already started to disengage rather than while it is still engaged. Timing and communication problems like these are genuinely difficult to see in yourself, and a short session with a competent trainer often resolves a stall that home practice alone could not. How to assess whether a dog trainer is competent covers what to look for when choosing someone for that second opinion.
When switching method or trainer is genuinely warranted
Switching is worth considering when a plan has been given a fair, structured attempt using an appropriate difficulty progression and consistent practice, and still shows no measurable progress or ongoing regression after that fair attempt, rather than after a single difficult week. It is also warranted if the current approach is producing rising stress or avoidance in the dog that is not resolving as difficulty is reduced, since continuing to push an approach the dog is finding aversive is not simply slow progress, it is actively working against the dog’s welfare.
Evaluate a different approach on the same basis you would want the first one judged on: a clear rationale for why the dog is likely to respond differently, a trainer or method that can explain what it will do differently and why, and a realistic timeframe rather than a promise of a fast fix. Switching because the first approach felt slow, without identifying what specifically stalled and why, risks repeating the same underlying handler-side or environmental problem under a different label. A poorly delivered reward-based plan can stall into repetitive food offering with little real progress, and a poorly delivered balanced plan can stall into repeated correction with no teaching behind it; neither outcome means the whole approach is wrong, it means that particular delivery of it needs fixing.
Genuine aggression, severe fear, established reactivity or resource guarding are not questions a change of training method resolves on its own. If a stalled plan sits in any of these areas, when to contact a behaviour professional is the more useful next step than continuing to troubleshoot a general plan, and getting that assessment early tends to save time rather than cost it.