pitch-deck

Mode skill. Default-enabled on the Stage specialist.

When to use

Use when the deliverable is the actual deck — slides in order, with speaker notes. Covers seed and Series A. For live delivery and Q&A, switch to demo-day-and-q-a. For the shift sentence anchoring slide one, run narrative-shift first.

A pitch deck is a sequence of claims, each earning the next.

The ten-slide arc

Sequoia template, refined with Raskin's spine. Eleven slides if the ask warrants its own page.

1. The shift. One sentence: what changed in the world. No product. No company name in the headline.
2. The stakes. Who wins and who loses under the new rules. The slide where the room leans in or checks their phone.
3. The problem. The new winner criterion, stated as a problem the buyer feels today. Customer-voice phrasing if available.
4. The old way and why it fails. Name the incumbent approach. State the dimension on which it falls short of the new criterion. Do not insult the incumbent — explain its design assumption and why the shift broke it.
5. The new way. Name the new approach as a category, not a product. "We're building a [category] for [audience] in the era of [shift]."
6. The product. One screen, one workflow, one frame. The product is the proof of the category, not the category itself.
7. Traction or design-partner evidence. Revenue, users, retention, design partners, waitlist — in that priority. Real numbers only.
8. Market built bottom-up. Number of buyers × annual willingness to pay × adoption assumption, with the source for each. Top-down TAM goes in the appendix.
9. Business model. Pricing, unit economics, payback period. Coin sets these; you place them.
10. Team. One sentence per founder naming the earned advantage relevant to this problem.
11. The ask. Round size, runway it buys, three milestones it will produce.

Procedure

1. Lock the shift sentence first. Run narrative-shift. Do not draft slides until the shift survives the three-question test.
2. Draft slide one as a headline plus a single proof point. No bullets. The headline is the shift; the proof point is the dataset or behavioural change that makes it undeniable.
3. Work slides two through five before the product slide. Founders skip to the product. You force the narrative spine first.
4. Pull traction numbers verbatim. If they say "around 40 users," the slide says "40 users" or you ask for the exact number. No rounding up.
5. Get the market build from Coin. Ask Coin to produce it; you place it.
6. Write speaker notes in the founder's voice. Read the slide aloud first; the notes are what the founder would say next, not a rephrase.
7. Stress-test against the twenty Q&A questions. Run demo-day-and-q-a. Any question the deck cannot answer in one breath becomes a slide edit or a prepared line.
8. Hand the visual template to Mira. You deliver content and structure; Mira sets the look. Do not pick fonts.

Decision rules

• One claim per slide. If a slide makes two claims, split it or cut one.
• The product is slide six. A stranger should be able to summarize the shift, stakes, and problem before they see your product.
• Bottom-up market only. Top-down sizing belongs in the appendix, never as the primary slide.
• The ask names milestones, not categories of spend. "$2M for 18 months to reach $1M ARR" beats "$2M for engineering, sales, marketing."
• No "vision" slides at the end. The shift slide carries the vision. A second vision slide signals the first is weak.

Anti-patterns

• Logo-clutter "as featured in" slides used as proof. Press is not traction.
• Team slide with five faces and three sentences each. One sentence, founders only.
• TAM/SAM/SOM dolls without a bottom-up build. Investors discount top-down to zero.
• Product screens with feature labels and arrows. One screen, one workflow.
• Competitive matrix scoring green on every row. No reader believes it; some leave on it.

Before / after

Brief: "Seed deck for a B2B tool that pulls product analytics into Slack."

Before (product-first, no shift, generic stakes):

Slide 1: "Acme — product analytics, simplified." Slide 2: "The problem: product analytics is hard." Slide 3: "Our solution: Slack-native dashboards."

After (shift-first, stakes named, product on slide six):

Slide 1: "In 2025, every product team moved standups into Slack. Their analytics stayed in a browser tab nobody opens." Slide 2: "Teams that act on data daily will outship teams that check it weekly. The week-checkers are losing." Slide 3: "Product managers want one number every morning. They have eleven dashboards and check none."

The "after" earns slide six. The "before" dies on slide three.