Twelve of the thirty-six competencies can only be assessed by a named teacher watching you teach a real class. This is what that review looks like when it comes back — marked per competency, with timestamps, in language you can act on.
Recorded class, submission 1 of 2 · reviewed against Yoga.org Teaching Standard v1.0
08:12 — "Press the outer edge of your back foot down" landed on five of seven mats within one breath, which is the test. You said it once and let it work instead of restating it three ways; that restraint is the harder half of cueing and you already have it.
21:40 — You caught the front knee drifting inward on the student at back-left and cued it as a relationship ("knee over the ankle") rather than a distance. That's principle 2 applied under load. 34:05 — You also let a wider stance stand on the taller student instead of matching them to the room. Principle 6, correctly ignored-on-purpose.
27:18 — Wrist collapse in Plank on the front-right mat; you moved the whole room to forearms rather than singling anyone out. Good instinct, and it protected the other two who were about to do the same thing.
12:00–24:00 — You demonstrated the entry once and then stayed upright and watched for the full twelve minutes. You can't observe a room from inside a pose, and you clearly know it.
31:52 — "Sorry, no, not like that — here, watch me" to the student at back-right. She stopped participating for the next four minutes and did not take the next two invitations to deepen. That is the specific harm this competency exists to prevent, and it's why this one isn't met on the first pass.
44:10 — Second instance, milder: you corrected the same student publicly again rather than moving to her mat. Correcting the same person twice in front of a room reads as singling out even when you don't mean it.
Resubmission, 18:30 — Room-level correction, no name, no demonstration, and the student adjusted and stayed engaged. Met.
04:30 — A student arrived and mentioned a shoulder. You changed the peak from a bind to a supported variation for the whole room within about ninety seconds, and never announced why. Handling it without making her the reason is the part most candidates miss.
58:40 — Asked after class about numbness down the arm, you said you weren't able to assess that and suggested she see a clinician before the next class. You did not name a condition and you did not offer a fix. Exactly right, and this is the competency we're strictest about.
02:10, 39:00, 55:20 — Invitational language throughout and a genuine opt-out at the close ("you're welcome to stay, and you're welcome to go"). No hands-on assist without asking. The choice you offered was real, not rhetorical.
Reviewer summary: a genuinely competent teacher with one habit to break. You see bodies well and you cue economically. The correction habit came from wanting to help fast; the fix isn't kindness, it's address — talk to the room. Nothing here needed a third submission.
Four of the twelve observed competencies are omitted above for length. A real review marks all twelve.
Every other competency in the standard can be assessed with a written response or an artifact you produce. These twelve cannot — they only exist while a real room is in front of you, and there is no test, essay, or model that watches a class and notices that a student went quiet for four minutes after a correction.
That's also the honest limit on our speed. A named human has to watch the whole session, which means certification takes as long as review takes, and we publish the turnaround rather than promising an instant credential.
Still being finalized with faculty: minimum class length, minimum students, reviewer turnaround target, and how many resubmissions are included. All published before any deposit is taken.
