The GTME Cup

Quarterfinals

Research: The Wide Net

Nooks sells an AI dialer to outbound teams. Step one: who's even worth a look?

8 signals entered. Pick who you think survives to the Semifinals:

Company size / ARR SDR headcount SDR job openings SDS leadership new hires SDS leadership job openings RevOps headcount Sales tech stack Outbound motion evidence

5 advanced. 2 became hard gates next round. The other 3 didn't vanish — they changed jobs:

Company size/ARR → shaped the segment Tech stack → competitor read, not a score RevOps headcount → cut entirely

Semifinals

Hypothesis: The Bet

4 signals + 2 gates. What's the real risk?

The bet: outbound motion + SDR growth signals = real buying intent, better than firmographics alone.

1

No confirmed outbound motion → row stops here.

2

10 or fewer SDRs → row stops here.

Everyone who clears both gets scored on what's left.

Which counter-hypothesis do you think was the real threat?

Signals mean nothing Gates too strict Wrong ICP for segment

Real answer: a mix of both — the bet beat all three challengers, but it was close against two of them. Check the match cards to see how close.

Evidence — Gate 1 sample rows

Company	Outbound Motion	Gate 1
GitHub	Weak signal	Held
Turing	Strong signal	Cleared
Docker, Inc	Confirmed	Cleared

Real rows, Quarterfinals workbook (500 companies).

Final

Build: Two Ways To Score

Same hypothesis, same gates — the scoring math is its own contest.

Scoring 500 companies on 3-4 signals — which approach wins?

Equal-weight sum score Weighted fit score

Weighted won — on try #2. Try #1 (equal weights) drowned the qualified table in false positives.

Scored

SDR job openings SDS leadership new hires SDS leadership openings

Enrichments — not AI

Company data Headcount by criteria 3rd-party tech stack ID Funding recency

Enrichments — AI

Outbound motion read Tech stack via JD scrape Weighted fit score

Run conditions

<10 SDRs → don't run No outbound motion → don't run Fit score <2 → don't send

Evidence — disqualification flags

Company	Fit Score	Flag
Docker, Inc	Ran	—
Movate	Ran	Disqualified
Cvent	Ran	Disqualified
Citrix	Ran	Disqualified

"Send SDR Headcount 11-50" workbook — passing the formula ≠ passing the gate.

The Champion

Result & Reflection

500 entered the 11-50 SDR segment bracket. Here's who lifted the cup — and why.

Guess how many of the 500 survive both gates:

Under 50 50–150 150+

500Entered

→

129Cleared Gates

→

80Score > 7

→

40Selected

Entered

500

Cleared Gates

129

Score > 7

80

Selected

40

Evidence — real fit scores

Company	Score	Why
BigPanda	9	Strongest signal combo
Teleport	9	Only match in cohort
Coralogix	9	Confirmed cold-calling motion
Andela	8.5	Well-corroborated evidence
Contentstack	8	Medium-confidence motion

"Nooks — Prompt 8 — SDR 11 to 50" workbook, 129/129 rows.

Open QuestionGates cut noise fast. Not fully proven: was the AI's read on outbound motion and fit score actually accurate? Next thing to stress-test.

🏆

"Build the hypothesis before you build the table. Without one, you can't know what worked — or what to fix."

Research finds the field.

Hypothesis gives you something to be wrong about.

Build is just the mechanism.

Result means nothing without the bet.

Reflection is the adjustment.

THE GTME CUP