L1product

Prompt Deck

The glue between ideas and execution. Five cards that compress your world into what matters — rendered on PRD pages, venture pages, the app, landing pages, marketing sites, and diagnostics. One component, every context. The structure is deterministic (five positions never change), the content is probabilistic (emerges through team competition and blind voting). Solo mode fills the grid. Multiplayer mode validates it.

1,200

Priority Score

Pain × Demand × Edge × Trend × Conversion

Customer Journey

Why should I care?

Five cards that sell the dream

1Why

Clarity that compounds, not decays.

When was the last time a strategy session produced something that still mattered three months later?

The friction: Every strategy offsite, venture plan, and pitch deck creates clarity once. Then the artifact freezes. The world moves. The deck doesn't. 2% flow efficiency.

The desire: An instrument where compression compounds. Fill what you know. The gaps become your strategy. Weekly return sharpens the grid.

The proof: This page is built with the instrument it describes. The first sale is to yourself.

The valuable stories behind this slide

Picture

A founder standing at a glass wall covered in five red threads connecting notes — dark room, single overhead light, the gaps between threads glowing. Cinematic, 16:9

2Evidence

One return visit proves everything.

What metric separates a novelty from an instrument?

Every journaling app, strategy tool, and planning instrument dies at the same point — the first week after the first fill. D7 return is the make-or-break.

What we measure	Target
Return within 7 days	> 35%
Grid cells filled	> 60%
External users who fill a grid	10

Performance thresholds

Picture

A scoreboard in a dark control room — five metrics in crimson on black, one number pulsing brighter than the rest. Minimal, cinematic, 16:9

3Platform

Templates that get smarter each use.

What if every business you helped made the next one easier?

The pattern: Standard templates + repeatable process + multiple perspectives = playbooks that evolve. Each completed venture improves the template for the next one.

1,200+ pages structured on the same 5P pattern. 15 PromptDeck implementations across industries, ventures, and PRDs. The mental model site IS the filled grid at portfolio scale.

The legacy rule: When you finish a job, improve the template for the next agent.

The maps that prove the platform

Picture

An architect's desk with a half-built model — crimson wireframe overlay on dark surface, tools arranged precisely. Top-down, cinematic, 16:9

4Loop

Fill, find gaps, return, compound.

When your campaign ends, does your pipeline end with it?

Campaigns end when budget stops. Instruments compound whether you're running a campaign or not. The weekly loop: fill → gaps → questions → research → update → compare.

Without the loop	With the loop
Clarity evaporates	Clarity sharpens
Pitch freezes at v1	Pitch evolves weekly
Strategy is a meeting	Strategy is a habit
Gaps stay hidden	Gaps become the plan

The process that compounds

Picture

A feedback spiral drawn in crimson light on a dark floor — seen from above, the loop tightening toward center. Abstract, cinematic, 16:9

5People

Founders first, everyone eventually.

What would change if everyone in your team could see the same five priorities?

You're in this moment	You need
Starting a venture, drowning in priorities	Five things on one page
Preparing a pitch with stale slides	A living deck that compounds
Life transition, too many dimensions	Five priorities, weekly check-in

Hidden objection: "I don't need another tool." The unlock: creating the deck IS the thinking. The tool is invisible.

Who we built this for

Picture

Three people around a table lit by a single crimson lamp — faces visible, hands on the table, five cards spread between them. Intimate, cinematic, 16:9

1 / 5

The instrument and the wiring are not the same thing.

Frontstage

The coaching instrument

A prompt in the right direction when you need it most. Senses drift, fires the right question, shows the depth. The user sees potential they couldn't see before. That's what a great coach does.

Backstage

The 5x5 grid editor

Create slides, edit prompts, manage content. Admin only. This is what currently lives at /insights/priorities. Users should never see the wiring — only hear the music.

Same five positions. Different seat. The customer sees the dream. The builder sees the process.

Feature Dev Journey

How does this get built?

Five cards that sell the process

1Job

The job: compression that compounds.

Can you state the struggling moment in one sentence?

Situation: People compress their world into five things — then lose the compression. Every tool creates clarity once, then the artifact freezes.

Intention: A living grid where empty cells highlight gaps as strategy and weekly return makes the grid sharper each pass.

Obstacle: Zero external users. No persistence layer. The prototype explored 5 patterns but none persist.

Intent contract

Picture

A single card illuminated on a dark stage — the text reads 'When I..., I want to..., so I can...' with blanks glowing. Cinematic, 16:9

2Stories

Every story has a number.

If the THEN doesn't name a data source and threshold, what are you actually testing?

Stories are test contracts. Each row = a test file. Tests must be RED before implementation starts. Tests going GREEN = value delivered.

Story	THEN assertion	Forbidden
S1: Grid persists	Cell state survives page reload	LocalStorage fallback (no real persistence)
S2: Gaps visible	Empty cells render distinct visual	All cells look the same empty or filled
S3: Weekly return	Snapshot diff shows changes since last visit	No diff — just the current state

Full story contracts

Picture

A test dashboard — rows of story IDs, each with a red or green indicator, thresholds visible. Dark background, terminal aesthetic. Cinematic, 16:9

3Build

60% composition, 40% new code.

What's already built that you haven't wired yet?

Exists	Wire	Build
CDD prototype (5 patterns)	Convex persistence	Cell editor
15 page implementations	Stackmates auth	Mode toggle
1,200+ pages domain content	Cinematic design	Card View (Stack)
PromptDeck React component	Image generation	Snapshot comparison

The static instrument works. The living instrument is the product.

Platform state

Picture

Architecture diagram split into layers — interaction, engine, data, platform — with connection lines showing composition. Crimson highlights on new code. Cinematic, 16:9

4Sprints

Four sprints to living instrument.

Which two features prove the thesis fastest?

Sprint	Ships	Proves
0 (2 wks)	Grid + persistence	Cells survive page reload
1 (2 wks)	Card View + modes + gaps	Mobile works, gaps glow
2 (3 wks)	Snapshots + share + export	Someone returns at D7
3 (TBD)	AI rhetoric + analytics	Data reveals which slides lose

Build order and commissioning

Picture

A timeline with four sprint blocks — each block lights up as features ship. Progress bar filling. Dark background, crimson milestones. Cinematic, 16:9

5Validation

10 grids filled or we stop.

What's the smallest experiment that proves people return?

Kill signal: 10 external users fill a grid. If zero return within 7 days, it's a novelty, not an instrument. Stop building.

Hole	Severity	Resolution
People fill once, never return	Critical	Snapshot comparison + gap highlighting
No external users to test	Critical	Founder-led pilots with 5 teams
White-on-white design	High	Room-change effect from cinematic prompt

Who we test with

Picture

A kill switch on a dark panel — crimson countdown timer, the number 10 prominent. Stark, minimal, cinematic, 16:9

1 / 5

The pitch is the shape. The flow diagrams prove the thinking. The VV stories validate the value.

Flow Diagrams VV Stories Design Prompt

Problem

Situation

People compress their world into five things that matter — strategy offsites, venture plans, life transitions — then lose the compression. Every tool (Notion, Miro, slides) creates clarity once, then the artifact freezes. 2% flow efficiency. Step 5 (Update) never happens.

Intention

A living instrument where the 5x5 grid persists, empty cells highlight gaps as strategy, and weekly return makes the grid sharper each pass. Compression that compounds instead of decays. Each card is a heading + image (mantra) that links to deeper thought chains in /docs/.

Obstacle

The frontstage (coaching instrument) and backstage (admin grid) are conflated. The live app at dreamineering.com/insights/priorities shows the backstage — a 5x5 CRUD editor. Users need the frontstage: a prompt in the right direction when you need it most.

Hardest Thing

Getting someone to return. Every journaling app, strategy tool, and planning instrument dies at the same point — the first week after the first fill. D7 return is the make-or-break metric because without it, this is a novelty, not an instrument.

Scorecard

Priority (5P)

4/5

Pain

4/5

Demand

5/5

Edge

5/5

Trend

3/5

Convert

Readiness (5R)

Principles5 / 5

Performance3 / 5

Platform3 / 5

Process2 / 5

Players2 / 5

What Exists

Component	State	Gap
PromptDeck container	Working	Carousel with keyboard nav, 5 pills, opacity transitions.
Slide component	Working	Headline, picture, prompt, mantra, systems props. Vision panel.
SlideDepth links	Working	CTA-styled depth links to spec/pictures.
15 implementations	Working	Industries, PRDs, ventures all using the component.
Visual treatment	Broken	White-on-white. No room-change effect. Blends into page.
Persistence layer	Missing	Zero persistence. No save, no return, no snapshots.
Image generation	Missing	Picture prop accepts URLs but no generation pipeline.
Cinematic design prompt	Working	4 presets documented. Not yet applied to component.

Relationships

PRD	Contributes
Time + Mind	Fractal pair: Prompt Deck = WHAT to hold in mind. Time+Mind = WHEN to wear which hat. Same five positions, different instrument.
Marketing Site Generator	Every generated marketing site has a diagnostic — which IS a Prompt Deck scoring across dimensions.
Better Practice	Gamified blind voting on practices IS the Prompt Deck multiplayer mode. Same thesis: validated through group testing.
Identity & Access	Auth dependency for persistence and team voting.
Agent Platform	Agents need prompt decks too — same instrument, different user.
Sales CRM	PromptDeck renders on CRM PRD page and in sales presentations.
Multi-Tenant Platform	Prompt Deck game needs app context (which venture), org context (which team), user context (who submitted). Three-entity model.

Kill Signal

10 external users fill a grid. If zero return within 7 days, it's a novelty, not an instrument. Stop building and revisit the job.

Questions

If the hidden narrator is the first audience, did creating this PRD change your conviction — or just arrange existing words?

Which of the 15 existing implementations best proves the pitch+prompt pattern — and what makes it work?
When the dynamic version ships (persistence, D7 return), will people return because of Zeigarnik or despite it?
What's the gap between the static instrument (working now) and the living instrument (the product vision)?