Judging criteria builder
Start from a template, adjust the weights, hand your judges something unambiguous.
How weighted judging works
Each judge scores every project on every criterion, using the same scale. Each score is multiplied by that criterion's weight, and the weighted scores are added up.
Weighting is what stops a polished demo from beating a genuinely better project on presentation alone — you decide up front how much presentation is actually worth.
Publish the rubric before judging starts. Teams build towards what you tell them you are measuring, and judges who have seen the criteria in advance score far more consistently.
Writing criteria judges can actually apply
Keep it to four or five criteria. Past that, judges stop discriminating between them and everything drifts toward the middle of the scale.
Say what you are looking for, not just what the criterion is called. "Innovation" means five different things to five judges; "solves the problem in a way we have not seen before" means one.
Avoid criteria a judge cannot observe in a three-minute demo. If nobody can see it, it will be scored on vibes.
Questions
- Do the weights have to total 100?
- It is not enforced, but it makes scores far easier to explain. The builder shows your running total and can even them out for you.
- What scale should I use?
- A 1–5 scale is easiest for judges to hold in their heads. Wider scales feel more precise but mostly add noise.
- Is anything stored?
- No. Your rubric stays in your browser until you copy or download it.
Judging is the part that goes wrong
HackathonHost runs weighted rubrics live: judges score on their phones, scores normalise across panels, conflicts of interest are blocked, and the leaderboard updates as results come in.
Every tool here runs entirely in your browser. Nothing you type is sent to us or stored anywhere.