Most reviews of online dog training judge one thing: is the method any good. That is half the job. A course can teach a humane, evidence-based method and still fail you completely, because it never reckoned with its second student. Not the dog. You.
This is the idea that makes our reviews different, and it is why every course gets a second score, Teachability and Design, alongside Method and Welfare. This page explains the science behind that score, so our reviews can point here rather than repeat it. For the dog’s side of the equation, see the science of dog training.
The bottleneck is the human, not the dog
Dogs learn constantly and quickly. Drop a dog into a well-run reward-based program and it will usually hold up its end. The hard part is the other end of the leash: a tired person, in a real living room, trying to change their own habits, timing, and reactions, consistently, across weeks, without a coach watching.
That is a human learning problem. And human learning has its own large, well-studied science that most dog courses ignore entirely. A course can be right about dogs and wrong about people, and when it is, you are the one who quits in week three feeling like a failure. You were not the failure. The instructional design was.
What the science of human learning says, and how courses break it
Six findings come up again and again in research on how adults learn skills. Each maps to a specific way online dog courses tend to fail.
Watching is not doing
The gap between understanding a technique and being able to perform it is called the transfer problem, and it is large. Information alone rarely changes behavior. What closes the gap is practice in the real context, with feedback and rehearsal. A course that is ten hours of polished video and no structured practice has handed you a textbook and called it a driving lesson. When we score teaching, a clear practice structure counts for more than production value.
Attention is finite
Cognitive load theory is the robust finding that working memory is small and overloads easily (Sweller). Skills have to be introduced one at a time and built in sequence. A four-hour unstructured “everything about obedience” video is a cognitive-load failure no matter how good its content is, because no one can hold it all at once. Good courses chunk. They teach one criterion, let you practice it, then add the next.
Show, do not just tell
People learn complex physical skills better from clear demonstration with concise narration than from a talking head or a wall of on-screen text (Mayer’s work on multimedia learning). This means editing and demonstration quality are not cosmetic, they are pedagogy. A course that shows the exact hand movement, the timing of the marker, and the dog’s response, from a camera angle where you can actually see it, is teaching better than one that explains the same thing in paragraphs. We judge this directly.
Memory needs spacing and retrieval
Skills stick through repeated, spaced practice and active recall, not a single weekend binge. A course that explains something once and moves on is fighting how memory works. One that builds in short daily repetitions, revisits earlier skills, and asks you to recall before it reminds you is working with it. Spaced, progressive structure is a quiet marker of a course built by someone who understands learning.
Follow-through is designed, not willed
Whether you actually do the training comes down to how easy, prompted, and routine it is, not how motivated you feel. The Fogg behavior model captures this: behavior happens when motivation, ability, and a prompt arrive together. Motivation is unreliable and fades, so good courses lower the difficulty and attach practice to things you already do, the morning walk, mealtimes, the ad break, so a session survives a busy week. Courses that depend on your enthusiasm staying high are designing for a person who does not exist by week two.
Honesty is part of the design
A course that promises a fully trained dog in a weekend is not just overselling. It is engineering your failure, because when reality arrives, you conclude the method does not work and stop. Realistic timelines, honest “this part is hard,” and clear troubleshooting for when things go sideways are features, not weaknesses. We reward them.
What a well-taught course looks like
When a course earns a high Teachability and Design score, it tends to share a recognizable shape:
- Sequenced, bite-sized lessons that introduce one skill at a time and build deliberately.
- A practice plan, ideally downloadable, that tells you what to do between videos and how to tell if it is working.
- High-quality demonstration, shot so you can see the handler’s timing and the dog’s response, with narration that adds rather than repeats.
- Troubleshooting for the common ways each skill goes wrong, because it will.
- Habit design: short daily reps tied to existing routines, not a marathon you will not repeat.
- Honest expectations about how long real change takes.
Notice that none of this is about the method. A reward-based course can be a teaching disaster, and a balanced trainer can be a superb instructor. That is exactly why we grade the two axes separately, and why a course’s two scores often diverge.
How this becomes a score
Teachability and Design is not a vibe. We look for the concrete features above and weigh them: structure and sequencing, practice and feedback, demonstration quality, habit and follow-through design, and honesty about scope and timelines. A course that nails the method but skips the teaching gets told so, in the score and in the words. The full breakdown is on our methodology page, and the dog’s half of the rubric is covered in why positive reinforcement wins.
The reason this matters is simple. You are not buying information. Information is free and everywhere. You are buying a change in your own behavior that sticks long enough to change your dog’s. A course that understands that is worth your money. One that does not will leave you exactly where you started, a little poorer and convinced it was your fault.
The learning principles referenced here draw on established work in instructional design and adult learning: cognitive load theory (Sweller), the multimedia learning principles (Mayer), research on spaced and retrieval practice, and the Fogg behavior model. Full reading list accompanies our methodology.