Get a panel to judge the same thing and you get a dozen opinions. Waza has them score independently, then shows where they truly agree — and where they don't. Judgment you can stand behind, and explain.
“And at the end, we sit in a room and try to agree on who makes the team.”
Same player, same drill — three evaluators, three different numbers. Without a shared scale, you collect noise.
A mean of 3 can mean everyone agreed on 3 — or that nobody agreed on anything at all.
When a parent asks why their kid didn't make it, that's all you've got. No trail, no defence.
Waza gives the panel one structured way to score, then surfaces where they agree and where they don't — so the call you make is the one the evidence actually supports.
“Three coaches scored independently. Here's where your kid stood on each skill, where we agreed, and where we saw it differently. We can talk about any of it.”
When the answer is transparent, a hard conversation earns respect instead of starting an argument.
Everyone scores independently — and no one sees another's numbers until every evaluation is in. You capture what each person actually saw, not who scored first.
The loudest skater on the ice pulls every eye — and one big impression quietly seeps into every score. A rubric makes each criterion stand on its own, so a dazzling personality can't inflate the skills underneath it, and the quiet, complete player never disappears into someone else's spotlight.
Each skill is scored deliberately, one at a time — so a strong overall impression can't stand in for skills nobody actually watched. A complete read of the whole player, not just the parts that grabbed the room.
If one evaluator's eye got caught and the others' didn't, that gap surfaces as divergence on a specific criterion — not a tidy average that quietly absorbs it.
When charisma or stage command is exactly what you're judging, put it on the rubric. It should count — but only in the column where you decided it counts.
Define your criteria and 1–5 scales — skating, shooting, game sense. Write your own or clone one from the Marketplace.
Attach the rubric, add the roster, assign your evaluators. Each organization sees only its own events.
Score each criterion by form or by voice — at the boards, in the wings, at the back of the hall, with or without signal.
Waza weights the scores, surfaces agreement, and flags where evaluators diverge — for the whole panel to see.
Half the time there's no Wi-Fi at the venue. Doesn't matter — score how you like, sync when you're back.
Tap through the criteria, score 1–5, submit. Installs as a phone app — keep scoring offline and it syncs the moment you reconnect.
Press and hold to talk. Transcription runs on your own hardware, Waza extracts structured scores from what you said, and you confirm before anything is recorded.
Waza weights each evaluator's scores and runs disagreement analysis, so a split panel looks like a split panel — never a tidy mean that hides it. And when one evaluator sees what the others missed, that signal stays in front of you — not flattened into the average.
Not every evaluator counts the same. Consensus reflects how the panel is weighted, not a flat average.
Where the panel splits on a criterion, Waza flags it — so you discuss the real contention instead of burying it.
A crowded tryout overloads any clipboard and any memory — the obvious standouts get kept, the rest blur together. But the lone high score nobody else gave isn't noise to average away, and the player who makes a team isn't always the highest overall — it's the one with the factor you needed. Scored criterion by criterion, that diamond in the rough lives in the data, where memory can't lose it.
Tap a term for the short version — or learn all of the terms used here, each with a plain definition and an example, and the story below reads itself.
“Would you like to evaluate your players beginning, middle, and end of the season? I've got a platform that makes that very, very easy.”
— the question that started it, asked to a mites coach
Same rubric, three checkpoints, one clear line of growth. Development stops being a feeling and becomes something you can show a player, a parent, a board.
It started at a rink — one coach, one clipboard, and a tryout that came down to memory. Waza exists so the call you make about someone else's kid is one you can defend. [PLACEHOLDER — founder's origin story to come]
It started at a rink. It works anywhere a panel has to turn what they saw into a fair, defensible call. Tryouts, camps, and full seasons — the same rubric, start to finish.
Don't build a rubric from scratch — clone one and tweak it to fit. The Marketplace holds shared rubrics you can freely adapt — heavier or lighter — alongside agreements that carry a platform baseline an organization can only strengthen, never weaken.
Enforcement is a ladder. The platform sets a minimum rung for each agreement; an organization can only move it stricter. Clickwrap signing and login gating make sure the right terms are accepted before work begins.
The sensitive work stays in-house, and the platform denies by default whenever anything is uncertain.
Transcription and analysis run on your own hardware. Audio and personal data are processed in-house and don't leave the building.
HIPAA boundaryBuilt-in detection screens content and personal information as it enters the platform, before it spreads.
Guardian controls and wellbeing rules look after the people doing the evaluating, not just the data they produce.
Every permission check denies on error. The creator can't be the approver. Every action is written to an access audit.
One platform, many orgs — each sees only its own events, rubrics, and people.
Admin, event manager, evaluator, viewer — permissions match the job.
Install as a phone app, evaluate offline, sync on reconnect.
An evaluation platform only earns trust if it's built like one. Every service follows the same disciplined pattern, gets tested against expected outcomes, and ships through a deliberate, reversible release path.
Every service is built to a single template: identity comes from a signed token, never from request parameters; authorization denies on error; and every action is written to a structured audit log.
A scripted acceptance playbook checks role permissions, multi-organization isolation, and full evaluation flows against expected outcomes — and we evaluate Waza with Waza before anyone else does.
exit gate · zero P1 bugsChanges move through development, validation, and production in reversible steps. Nothing auto-deploys to production — every promotion is a deliberate, approved action, with secrets isolated per environment.
SAST · dependency · secret scanA bidirectional access audit shows exactly who can do what — and separates standard role-based grants from any that bypass them. The creator can't be the approver, and access outside its scope is denied by default.
Build the rubric. Score independently. See where you agree.
That's Waza.