How judges actually score — and the ten ways teams die
Reverse-engineer the scorecard, then audit your deck against the kill-list.
A screening judge is a manager at the sponsor, reading deck #67 of 100 at 11pm on a weeknight. Everything about how you win follows from that single image.
TL;DR · Key Takeaways
Key takeaways
- Judges are tired insiders: design for the skim, source every number, and never recycle a deck they can pattern-match.
- The rubric is your deck-space budget; the printed one always overrides the typical weights.
- Run the ten-point kill-list audit before every submission, and debrief after every result — that loop is the real coaching.
Who the judges are, round by round
| Round | Who judges | What they reward |
|---|---|---|
| Quiz/screening | Software + junior managers | Format compliance, a skimmable exec summary, no obvious errors |
| Deck rounds | Mid-level managers in the function | Rigour, sourcing, feasibility — they know the business cold, so realism beats flash |
| Finale | CXOs + HR leadership | Clarity, decisiveness, poise under grilling — and whether they can imagine hiring you |
Thinking like judge #67
They skim before they read
Exec summary → slide titles → charts. If that pass doesn't deliver the argument, the careful read never happens. This is why the skim test from the deck module is non-negotiable.
They pattern-match against their job
Sponsor-side judges instantly recognise a team that read their annual report — and a team recycling a deck from a different competition with the logo swapped.
They are hiring, not grading
Especially at finales: the unspoken question on every scorecard is "would I want this person on my team next April?" Composure, honesty about limits, and crisp answers are CV signals being scored live.
What "innovation" means to a judge
Not novelty — non-obviousness that survives feasibility. The winning idea is usually the one a smart insider would call "huh, we could actually do that," not the one requiring three technologies the sponsor doesn't own. Calibrate boldness to one pillar (the structural bet) and keep its math honest.
The pre-submission audit
One hour before submitting, the whole team runs the kill-list as ten yes/no questions against the deck — does the exec summary carry the full answer, does every number have a source, is the format spec met to the letter, would a hostile reader find a recommendation or a menu. Then score yourselves 1–10 on each rubric line. Anything under 7 either gets fixed or gets a prepared defence in the appendix. Teams that audit honestly stop being surprised by results.
After every result: the debrief
Win or lose, spend 30 minutes: what got asked in Q&A, which slides judges lingered on, what the winning team did that you didn't (their decks are often shared or presented publicly — study them). A team that debriefs improves a full round-depth per competition; one that doesn't replays the same exit forever.