A real seed deck, scored 72, then rewritten to FUNDABLE

We score deck rewrites all day. Most of the time the founder fixes one slide, score moves 4 points, and we never see them again. Occasionally a founder runs the loop fully, rewrites 7+ slides, and lands a fundable score. This is one of those.

The founder gave us permission to publish the teardown anonymized. They are a B2B SaaS company in the operations space, raised pre-seed in 2024 for $400K, and were re-raising a seed of $2M in spring 2026. They closed at $2.1M with 1.5x oversubscription, 19 days from first email to first commit.

Here is what their deck looked like before, what it looked like after, and what we changed slide by slide.

First-pass score: 72/100

Dimension breakdown:

• Problem clarity: 62
• Solution specificity: 68
• Market sizing realism: 75
• Traction proof: 78
• Team credibility: 80
• Ask + use of funds: 65
• Business model fit: 78

Strong on team and traction. Weak on the fundamentals: problem, solution, and ask. The kind of deck where investors say "interesting, but not now" without being able to articulate what would change their mind.

Slide 1 — Problem (was: "We solve [vertical] operational chaos.")

Before: 14 words, generic, no specific pain or population. Scored 62.

After: A specific moment, a specific number, a specific person.

"Mid-market operations leads spend 11 hours a week reconciling vendor invoices in Excel because their ERP cannot read non-standard PDFs. The cost: $34K per ops lead per year in salary on a copy-paste task."

Why it scored higher: it names the buyer (ops lead), the specific pain (PDF reconciliation), the specific number (11 hours, $34K), and an explicit cause (ERP can't read non-standard PDFs). A generalist VC reads this and feels the pain immediately.

Score after rewrite: 88.

Slide 2 — Solution (was: "An AI-powered platform for ops teams")

Before: marketing copy, no wedge, no defense.

After: a single sentence wedge plus a "why us, why now" paragraph.

"We turn any non-standard vendor PDF into a structured ERP entry in 6 seconds. We trained the model on 14M anonymized invoices from a partner ERP integrator nobody else can access."

Why it scored higher: the wedge ("turn PDF into ERP entry in 6 seconds") is concrete and measurable. The "why us" sentence explicitly addresses the obvious objection ("couldn't a big incumbent build this?") with a specific moat (the 14M-invoice dataset).

Score after rewrite: 84.

Slide 6 — Traction (was: 14 metrics on one slide)

Before: ARR, MRR, MoM growth, weekly active, monthly active, retention, NPS, CAC, LTV, sales cycle, expansion revenue, logo retention, GP margin, churn. A wall of numbers nobody could digest.

After: 2 metrics on one chart.

Top line: ARR over the last 9 months. Y-axis: $0 to $440K. The line goes up and to the right with a visible inflection at month 6 when they shipped the auto-categorization feature. Bottom line: Net revenue retention by cohort. Steady at 124%.

The second number is the killer. NRR over 120% in B2B SaaS at this stage is rare. Putting it next to the ARR chart says "this is real and it grows on its own."

Score after rewrite: 94.

Slide 8 — Ask (was: "Raising $2M for product, sales, and marketing")

Before: generic, no milestone, no Series-A trigger.

After: specific milestone tied to next-round metrics.

"Raising $2M to take us from $440K ARR to $2.5M ARR over 18 months. Hits the Series A trigger conditions for B2B vertical SaaS in our segment: $2M+ ARR, 130%+ NRR, payback under 14 months."

Why it scored higher: the ask is no longer "we want money to operate." It is "we want money to hit a specific Series-A-defensible position in 18 months." Investors can underwrite that.

Score after rewrite: 89.

The 11 slide changes that mattered

Beyond the 4 slides above, the founder also rewrote:

5. Title slide tagline: removed "AI-powered" (banned vocabulary), added the 6-second-per-PDF claim. +2 points.
6. Market slide: replaced top-down TAM with a bottom-up "30K mid-market ops teams × $48K ACV at year 3 = $1.4B SAM, our beachhead is the 4K with custom ERP setups." +9 points.
7. Competition slide: removed the 2x2 quadrant. Replaced with one paragraph naming the 3 closest competitors and the specific feature each lacks. +6 points.
8. GTM slide: replaced "channel strategy" buzzwords with the actual playbook: 4 ERP integrator partnerships, 1 paid + 1 organic LinkedIn channel, an inbound trial flow. +7 points.
9. Team slide: re-ordered founder bios so each one led with the relevant scar (CTO ran ERP eng for 6 years; CEO sold a similar product to mid-market ops as her last role). +5 points.
10. Financials slide: added a single CAC payback line, dropped 3 unused projection columns. +4 points.
11. Why now slide: removed entirely. Investors at this stage do not need a "why now" lecture; the traction is the why now.

Second-pass score: 87/100

Dimension breakdown:

• Problem clarity: 88
• Solution specificity: 84
• Market sizing realism: 84
• Traction proof: 94
• Team credibility: 85
• Ask + use of funds: 89
• Business model fit: 86

The deck moved from "interesting, pass" territory to "let's lead the round." The founder sent it to 22 funds. 19 took the meeting. 7 wrote terms. They closed in 19 days.

What this teardown actually proves

Most decks fail at the same 4 dimensions: problem clarity, solution specificity, ask, and business model fit. Fix those, even with average traction, and the score moves 12-15 points. The traction proof slide is the highest-weight, but it is also the only one that is hard to fake; it moves up only when your numbers move up.

The deck that closes a round is rarely the deck the founder started with. The 47% close rate we see for 80+ scoring decks is largely founders who ran this loop two or three times and stopped sending the deck before the rewrite hit.

If you want your own deck scored against the same rubric, upload it. The first score is free; the rewrites are why founders pay us.