Dog Training & Behavior · Article

When to Raise Criteria and When to Go Back to Basics

When to Raise Criteria and When to Go Back to Basics

I use a rough rule of thumb with my clients: if your dog is succeeding 80% of the time, you can think about raising the bar. If they're succeeding less than 50% of the time, go back a step. Between those two numbers, stay where you are and keep building.

It sounds simple. It's harder to apply than it looks, because most people raise criteria based on impatience rather than the dog's actual success rate.

What "Raising Criteria" Means

Criteria is what you're asking for in a given training step. For a stay, criteria might be: 5 seconds, you standing 3 feet away, no distractions. Raising criteria means increasing one of those variables — duration, distance, or distraction level.

The foundational rule is raise one variable at a time. When you increase duration, keep distance and distraction constant. When you add distance, drop duration back down. When you add a distraction, work at short duration and close distance until the dog is stable.

People get into trouble by raising all three at once, then wondering why the dog is failing.

The 80% Rule

At any given criteria level, your dog should be succeeding roughly 8 out of 10 repetitions before you raise the bar. Not 10 out of 10 — that might mean you're not challenging them enough and the progress is slower than it could be. Not 5 out of 10 — that means you've moved ahead of where the dog's understanding actually is.

Eight out of ten is the sweet spot: the dog knows the behavior, understands what's being asked, and is ready to be challenged.

Signs You've Raised Criteria Too Fast

  • The dog starts making errors they weren't making before
  • The dog shows stress signals (yawning, lip-licking, avoidance, disengagement)
  • The dog "checks out" mentally during sessions
  • A behavior they had reliably starts falling apart
  • The dog starts offering alternative behaviors (hedging)

When you see these, don't push through. Drop back to where the dog was succeeding and build more slowly toward the new criteria.

Building advanced behaviors requires patience in this process

In building advanced behaviors, the most common mistake is moving forward when the foundation isn't actually there. A dog that looks like they know a behavior might only know it in one specific context at one specific criteria level. Changing anything — new location, new distraction, your mood on a given day — can expose that the behavior is less solid than it appeared.

This is why going back to basics isn't failure. It's recalibrating. You're finding the level where the dog is actually fluent and building forward from there with accurate criteria.

When to Go Back to Basics

  • New environment: Every new location effectively resets criteria somewhat. Start easier than you think you need to.
  • Long gap since training: If you haven't practiced a behavior in weeks or months, expect some regression. Refresh from the beginning, not from where you left off.
  • After a stressful event: A dog that had a scary experience may regress on behaviors. Rebuild confidence in the basics before pushing forward.
  • Consistent failures: If you're failing more than two or three sessions in a row, the dog is telling you the criteria is wrong. Listen.

Maintaining trained behaviors also involves knowing when to step back

Part of good maintaining trained behaviors practice means occasionally testing behaviors at increased criteria to see where the edge is — and then being willing to back off when you find it. (See also: puppy regression.)


The trainers I admire most are the ones who go back to basics without ego. The dog doesn't know you're regressing. They just know training is working and they're succeeding. That's all that matters.