SKIN OS — Decision Frameworks & AI Schema

Trigger condition

?

Use this playbook when

You're generating consistent revenue or strong engagement, you personally are the bottleneck to growth, and you're debating whether to hire before raising or wait until after.

Decision framework

A

Pre-hire readiness gates

Before any hire, all three gates must be green.

›

Gate	Signal required	If missing
Role clarity	You can write the job in one sentence with a measurable 90-day outcome	Wait — undefined roles produce mis-hires
Revenue floor	12+ months runway at current burn even after adding the hire's salary	Wait — hire under pressure = bad hiring decisions
Founder bottleneck	You can identify a specific task consuming 30%+ of your time that a hire would take over	Pause — the bottleneck may be process, not headcount

B

The hire decision tree

Step through these questions in order. Stop when you hit a "Wait" or "Hire" conclusion.

›

?

Question 1

Is the task you need help with in your core zone of genius — or outside it?

→

Outside your zone

You're creating a structural weakness. Hire as soon as revenue floor is met. This is highest-leverage.

→

Inside your zone

You're at capacity in your strength area. Move to Question 2.

?

Question 2

Does hiring now make your fundraise more likely to succeed — or less?

→

More likely (strategic hire)

A CTO, VP Sales, or domain expert that strengthens the story. Hire pre-raise if you can afford it. The equity premium is worth the narrative unlock.

→

Less likely or neutral

Raise first, then hire. Don't dilute runway when the hire won't accelerate the raise.

?

Question 3

Can this need be met with a fractional or contractor arrangement instead?

→

Yes — role can be fractional

Use fractional. Test the role definition and the need. Convert to full-time only when you can't get enough of their time. Preserves runway and reduces mis-hire risk.

✓

No — needs to be full-time

All gates are green, the hire is justified, and fractional won't cut it. Hire now. Move fast. A 30-day mis-hire costs less than a 6-month vacancy.

Hire now — signals

12+ months runway post-hire · Role defined with 90-day outcome · Strategic hire unlocks the raise · Founder is genuine bottleneck

Wait — signals

Under 12 months runway · Role unclear or vague · Hire driven by founder anxiety, not evidence · Process problem masquerading as headcount problem

C

The mis-hire early warning system

Check at 30 and 60 days post-hire.

Critical

›

Week	Green signal	Red signal — act immediately
Week 2	Asks clarifying questions about priorities, not just tasks	Replicates what you were already doing rather than improving it
Week 4	Has identified one thing to change or improve that you hadn't seen	Requires daily direction — no autonomous initiative
Week 8	You've stopped worrying about their area of ownership	You're doing their job alongside them

!

Rule

If Week 8 red signal is present: the hire is wrong, not the person. Address immediately. Every week of a mis-hire costs more than the exit severance.

Core diagnostic

A

The four raise readiness tests

You need to pass at least 3 of 4 to raise with conviction. Raising at 2 or below will cost you 6 months and dilute relationships.

›

1

Evidence test

Do you have at least 3 repeatable customer wins (same profile, same motion, same value delivered) — not 10 one-off experiments?

2

Milestone test

Can you articulate exactly what this round gets you to — a specific, investable milestone that changes your valuation story — in 2 sentences?

3

Runway test

Do you have at least 9 months of runway remaining? Raising below 6 months is survival mode — investors can smell it and it compresses your leverage.

4

Investor access test

Do you have at least 3 warm paths to target investors — not cold outreach? If all paths are cold, the raise will take 2x as long as expected.

0–1 tests passed: Don't raise 2: Raise with caution 3–4: Raise now

Raise now

3–4 tests passed. You have leverage. A raise now will be faster, cleaner, and produce better terms. Don't wait for perfect — perfect doesn't close rounds.

Grow first

0–2 tests passed. Going to market now costs relationships and time. Identify which test you're closest to passing and target that milestone specifically. 60–90 day sprint.

B

If you decide to raise — the 90-day sprint

Fundraising is a full-time job. Running it poorly while also running the company is the most common execution mistake at this stage.

Process

›

Week	Focus	Output
1–2	Materials prep	Deck locked, data room complete, 50-name investor list prioritized by warmth
3–5	Soft launch	First 10 meetings — friends, angels, advisors. Refine pitch from real objections.
6–10	Primary process	Target funds in parallel. Create momentum — no sequential conversations.
11–12	Term sheet pressure	Use first term sheet to accelerate remaining conversations. 2-week close window.

!

Critical rule

Never run sequential raise conversations. All target investors must be in process simultaneously. Sequential raises give each investor infinite time and zero urgency.

The pivot vs. persist diagnostic

A

Five-signal test

Score each signal. 4+ pointing to "pivot" is a clear signal. 2–3 means optimize before you pivot.

›

Signal	Persist	Pivot
Customer retention	Users return without prompting. Cohort retention is stable or improving.	Churn is consistent and the explanation keeps changing. No user has renewed twice without a discount.
Conversion signal	You can reliably predict which prospects will convert and why.	Every deal feels different. No repeatable pattern to who buys or why they say yes.
Competitive response	Customers choose you over alternatives for a specific, named reason.	Customers don't compare you to anything — they evaluate you against doing nothing.
Founder energy	You still find the problem genuinely interesting. Setbacks fuel you.	You're building to prove you were right, not because the problem matters to you.
Market signal	Inbound interest from buyers you didn't pitch. The market is finding you.	Every customer required founder-level sell. Zero organic inbound after 12+ months.

B

Types of pivot — not all are equal

Before deciding to pivot, be specific about what you're actually changing.

Framework

›

1

Customer pivot (lowest risk)

Same product, different buyer. Often the right move when the product has real fans but they're not who you've been selling to. Audit your best customers — is there a pattern you've been ignoring?

2

Problem pivot (medium risk)

Same customer, different problem. You have distribution and trust but you're solving the wrong pain. What do your best customers actually complain about? Shift to that.

3

Technology pivot (medium-high risk)

Same problem, different solution approach. You've validated the problem but your solution isn't the right vehicle. Common in technical startups where the approach was limited by early assumptions.

4

Full pivot (highest risk)

Different customer, different problem, different solution. Only justified if you have a genuine team asset (technology, data, network) that transfers. Otherwise this is starting over — be honest about that.

Price raise readiness signals

A

When to raise price

Any two of these signals is sufficient justification. Waiting for all four is leaving money on the table.

›

✓

Signal 1 — close rate too high

If your close rate is above 60% on qualified leads, you are underpriced. The market is saying yes too easily.

✓

Signal 2 — no price objection

If fewer than 30% of prospects mention price in the sales conversation, your price isn't high enough to be a real consideration.

✓

Signal 3 — customer ROI >> price

If your best customers can articulate an ROI that is 10x+ your annual fee, you are dramatically underpriced relative to value delivered.

✓

Signal 4 — waitlist or backlog

Demand is exceeding your delivery capacity. Price is the lever to slow demand while you scale. Raise it.

→

How much to raise

Raise by 20–30% in one move. Small increments signal uncertainty. Test the new price on 5 new prospects before rolling back to existing customers. If close rate holds at 40%+, you're still underpriced.

Focus vs. diversify framework

A

The channel commitment rule

You should be running a single primary channel until it breaks — not until it slows down.

›

?

Are you below $1M ARR?

If yes: you should be on exactly one channel. No exceptions. Diversification before channel mastery is the single most common GTM mistake at early stage.

?

Is your primary channel at capacity?

Capacity means: you've extracted all the volume available, CAC is rising despite optimization, or the audience is saturated. "Slowing down" is not capacity — it's a test of whether you've found the ceiling or just the plateau.

→

If at capacity — test before committing

Run a 30-day experiment on one adjacent channel with a fixed budget. Commit to the expansion only if CPL or CAC from the new channel is within 40% of your primary channel's performance.

Readiness conditions

A

You are not ready to hire sales until

All four must be true. Hiring before this point produces a rep who can't succeed — and a founder who blames the hire.

›

1

The pitch is documented

You've written down the exact words that work. Not bullet points — the actual script. A new rep needs a starting point, not a conversation with the founder's memory.

2

You've closed 10+ of the same deal

Not 10 different deals. 10 deals to the same ICP using the same motion. Repeatability is the signal. If you haven't found the repeatable motion yet, a rep won't find it for you.

3

Objections are known and handled

You have a written objection-handling guide covering the top 5 objections you encounter. If you can't write this, you don't know the motion well enough to teach it.

4

You can afford 6 months of ramp

A new sales rep needs 3–6 months to reach quota. Budget for that ramp explicitly. If you can't afford it, the hire will fail due to financial pressure regardless of talent.

Churn diagnosis tree

A

The four churn root causes

Each has a different intervention. Treating the wrong root cause wastes 90 days.

›

?

Root cause 1 — Wrong customer

Churn concentrated in one customer profile. Intervention: tighten ICP, stop selling to that profile. Don't try to save customers who were never right.

?

Root cause 2 — Expectation mismatch

Customer churns saying "it wasn't what I expected." Intervention: fix the sales process, not the product. You oversold or misdirected. The fix is qualification, not features.

?

Root cause 3 — Onboarding failure

Churn happens at 30–60 days. Customer never reached their first value moment. Intervention: redesign onboarding to deliver a specific, undeniable win in the first 7 days.

!

Root cause 4 — Product doesn't deliver

Customers who used the product well and still churned. This is the hard one. Intervention requires honest product diagnosis. No amount of CS or sales fix will resolve a core product value gap.

The critical distinction first

A

No market need vs. bad marketing — not the same problem

Founders consistently misdiagnose this. The wrong diagnosis wastes 3–6 months on the wrong fix.

Critical distinction

›

A

No market need

People don't want what you're building, even when they understand it. You can explain it perfectly and they still don't buy. The problem you're solving isn't painful enough, frequent enough, or worth the switching cost. Fix: change what you're building, who you're building for, or both.

B

Bad marketing (misdiagnosed as no market need)

People want what you're building, but they can't find it, understand it, or trust it from the way you describe it. The demand exists — you're just invisible to it or unintelligible. Fix: change how you describe, reach, and convert, not what you build.

Signs of true "no market need"

Users understand the product but don't pay · Churn happens immediately after first use · Users can't explain what problem it solved · No inbound interest after 6+ months · Discount doesn't increase conversion

Signs of "bad marketing" instead

Users who reach you do convert at decent rates · Satisfaction is high but acquisition is broken · Word of mouth is strong but small · The competitor who "stole your customers" isn't better — just louder

B

The four diagnostic tests

Run these before deciding whether to fix the product, the market, or the motion.

›

1

The Sean Ellis test

Survey your active users: "How would you feel if you could no longer use this product?" If fewer than 40% say "very disappointed," you have a need problem, not a growth problem. Don't add features or run ads until this number moves.

2

The substitution test

Ask churned users: "What are you doing instead?" If the answer is "nothing" — they went back to doing without — the problem wasn't painful enough. If the answer is a competitor, the market exists and you lost the sale.

3

The willingness-to-pay test

Remove the free tier or trial. Charge from day one at a price that requires a real decision. Users who pay without being asked are revealing genuine need. Users who engage for free but won't pay are giving you social proof, not market validation.

4

The inbound test

Stop all outbound for 30 days. Does anyone find you? YC data shows 70% of top YC companies had organic demand from day one — the market pulled them, they didn't push. Zero inbound after 6 months is a hard signal.

C

If it is no market need — the exit options

Failing slowly is the worst outcome. If you've confirmed no market need, the decision tree below prevents wasted runway.

Use only after B is confirmed

›

?

Do you have customers who love a specific part of the product — even if they don't love the whole?

If yes: that's your beachhead. Narrow to it, kill the rest. This is a customer pivot — lowest risk, preserves the most.

?

Do you have a distribution asset (audience, channel, relationships) that has value independent of the product?

If yes: this is your real asset. What problem does this distribution give you unfair access to solve? Build that instead.

?

Does your team have a technical capability that's hard to build — even if the application was wrong?

If yes: technology pivot. Find a different problem that genuinely needs this capability. The infrastructure is proven; the application wasn't.

!

None of the above — full pivot or wind down

If you have no customers who love anything, no distribution, and no transferable technical asset — this is a start-over situation. The honest path is to wind down cleanly, preserve relationships, and start the next thing with the hard-earned clarity of knowing what not to do.

The core decision

A

Own a category vs. join one

The most consequential positioning decision a founder makes. Both paths work — but they require completely different resources and timelines.

›

	Own a new category	Join an existing category
When it works	You have the authority, budget, and time to educate a market. Category creator wins are enormous but take 3–5+ years.	The category is established, buyers have budget, and you're genuinely differentiated within it. Faster to revenue.
What you need	Strong founder POV, existing audience/platform, patient capital, and a clear enemy (the old way)	A sharp "why us vs. them" answer, a beachhead segment the incumbent ignores, and a fast sales motion
Fatal mistake	Creating a category name without the authority to make it stick — you become uncategorizable and therefore unfindable	Entering a category dominated by an entrenched player with no structural differentiation — you compete on features and lose on distribution
SKIN signal	Brand score ≥70, Market score ≥75, founder has existing market authority	Market score ≥65, GTM score ≥60, clear ICP definition against category incumbents

B

The category audit — 5 questions

Run this before finalizing any positioning. Most founders skip it and pay for it in investor confusion and slow sales cycles.

›

Q1What category does Dealroom / Crunchbase / your target investors put you in when they find you? Is that accurate? Does it matter?
Q2When your best customer explains your product to a colleague, what category do they use? Is it the same one you use?
Q3Who are the top 3 companies an investor would compare you to on first glance? Are you happy with that comparison? If not, what has to change?
Q4If your company disappeared tomorrow, what budget line does a buyer move to replace you — and what category does that vendor sit in?
Q5What would have to be true about your company for a Dealroom analyst to categorize you in your ideal category? Are you there yet?

C

Common category mistakes — from Failory's misclassification cases

›

The compound category trap

"AI-powered B2B SaaS platform for the future of work" — three categories in one name. Investors can't place you. Buyers can't find you. Pick one category and be ruthless about it.

The too-early category creator

Creating a new category before the market understands why the old category is broken. You spend all your energy educating rather than converting. The category eventually forms — around someone who entered later with better timing.

The wrong comparison set

Naming the wrong competitors in your positioning puts you in a different mental category than you intend. RethinkDB famously named the wrong market in their postmortem — their category framing made investors and buyers think smaller than the opportunity was.

The narrative architecture

A

The five elements of a pitch that converts

Based on Wefunder campaign analysis and YC pitch data. All five must be present. Missing even one creates a credibility gap that investors feel but can't always articulate.

›

1

The personal origin — why you, why this problem

Not your credentials. The specific moment or sustained experience that made this problem personal to you. Investors fund conviction, and conviction is only believable when it has a real origin. Generic origin stories ("I noticed a gap in the market") are the leading credibility destroyer in early-stage pitches.

2

The specific evidence — not claims, proof

The three most credible data points you have, stated without hedging. Revenue, retention, a customer quote that uses language you didn't give them, a wait list that formed without ads. The specificity of the evidence is itself a signal — vague claims signal vague traction.

3

The transparent risk — what could go wrong

Wefunder data consistently shows that campaigns which name their risks outperform those that don't. Investors know the risks exist. Naming them signals sophistication and honesty. Hiding them signals naivety or worse. State the top two risks and what you're doing about each.

4

The specific use of funds — milestone, not category

"To grow the business" fails universally. "To get from $180K to $1M ARR by hiring one account executive and running 90 days of channel experiments" works. The specificity of the milestone signals that you know your business. Investors are buying the milestone, not the money.

5

The timing argument — why now, not eventually

The structural change (regulatory, technological, behavioral) that makes this moment different from 3 years ago and 3 years from now. YC specifically trains founders on this: name recent events — not macro trends — that have opened the window. Windows close. Investors fund windows.

B

Valuation red flags — sourced from Wefunder data

These patterns appear repeatedly in pitches and campaigns that fail to close. Each one is a credibility destroyer that experienced investors recognize immediately.

Credibility destroyers

›

Pattern	What it signals	Fix
Hockey-stick projections with no named inflection driver	Founder hasn't thought through the mechanics of growth. The model is aspiration, not analysis.	Replace with a specific driver: "When we add the second sales rep in month 4, based on current rep productivity, we expect X." Tie every inflection to an action.
Revenue multiples above 30x at early stage without a clear growth rate	Valuation is comp-based, not evidence-based. Wefunder cases show 165x+ multiples at early stage consistently fail to close institutional follow-on.	Anchor valuation to milestone math: "At the end of this round we'll have $X ARR, which at industry comps of Y multiple = $Z valuation." Show the work.
Raising "to extend runway"	No milestone attached. The business isn't creating its own momentum and the founder knows it. Investors smell survival mode immediately.	Only raise when you can articulate the specific proof point the capital gets you to and why that proof point changes your valuation story.
Cold outreach as primary investor access strategy	Signals lack of network or credibility in the space. Cold raise success rates are 5–10x lower than warm introductions.	Map the 3 warmest paths to each target investor before launching the process. One strong warm intro beats 50 cold emails.

Playbooks / Source-Informed / 11

Founder Resilience

YC Startup School data across 15 years and 5,000+ companies identifies resilience — not intelligence, pedigree, or confidence — as the #1 predictor of founder success. Startups.RIP postmortems confirm: team cohesion under stress is the leading indicator of survival. This playbook operationalizes it.

Team dimension Source: YC Startup School · Startups.RIP 1,737+ postmortems

What resilience actually means operationally

A

The three resilience failures — from Startups.RIP patterns

These are the specific ways resilience collapses in early-stage companies. Each has a different intervention.

›

1

Narrative collapse under pressure

Founder changes the company's story when challenged by an investor, customer, or bad month. The narrative was never anchored in conviction — it was anchored in optimism. When optimism gets tested, the story changes. Symptom: each version of the pitch is subtly different. Intervention: identify the one thing that will still be true about this company even if the first 3 approaches fail. Anchor there.

2

Co-founder fracture under stress

Startups.RIP postmortem pattern: co-founder conflict rarely begins in good times. It surfaces when the company hits its first serious obstacle — and the two founders disagree on how to respond. The undeclared CEO problem becomes live. Intervention: run the hard conversation before the hard moment. Agree now on: who decides when you disagree, what the shutdown criteria are, and what each person's non-negotiable is.

3

Motivation drift

YC data: founders who started a company to prove they were right — rather than to solve the problem — run out of fuel when the problem proves harder than expected. The extrinsic motivation (validation, status, money) doesn't sustain through the 18-month stretch where nothing works. Intervention: the founder needs to find genuine interest in the problem, not just the outcome. If the problem itself has become uninteresting, that's a signal worth taking seriously.

B

The resilience check — quarterly self-audit

Run this every 90 days. Not as a performance review — as an early warning system.

›

Q1What is the hardest thing that has happened in the last 90 days? How did you respond — and would you respond the same way again?
Q2Is your co-founder relationship stronger or weaker than it was 90 days ago? What specific thing changed it?
Q3Do you still find the problem you're solving genuinely interesting — not just the outcome of solving it, but the problem itself?
Q4What would have to be true for you to walk away from this company? Have any of those conditions changed?
Q5When you describe what your company does to someone new, does the story feel true — or does it feel like a pitch?

→

SKIN scoring note

These questions feed directly into Team dimension signals: conviction under pressure, co-founder cohesion, and founder-market fit. A founder who can answer Q3 and Q5 with specificity scores significantly higher on the Team dimension than one who deflects or gives generic answers.

C

The W12/S12 pattern — what early cohort failures teach us

Startups.RIP analysis of YC batches shows a disproportionate failure rate in the W12/S12 cohort. The lesson is generalizable.

Source: Startups.RIP

›

→

What happened

The 2012 YC cohorts coincided with a post-2008 funding surge that attracted a higher proportion of "opportunity founders" — people who entered markets because funding was available, not because they had lived the problem. When growth slowed (2013–2015), these founders had no conviction reservoir to draw from. The story changed, the co-founders split, or the company simply wound down quietly. The pattern is not specific to 2012 — it repeats in every hot funding cycle.

✓

The SKIN implication

When a company raises during a hot market cycle (AI, 2023–2025 being the most recent), SKIN analysts should apply additional scrutiny to Team dimension signals — specifically: did the founder enter this market because of funding availability, or because of genuine problem proximity? The funding environment does not change the underlying signal requirements.

Playbooks / CB Insights Data / 12

Unit Economics

19% of VC-backed startups that shut down since 2023 cited unsustainable unit economics as a primary cause. The pattern: founders knew the numbers were shaky and scaled anyway. This playbook tells you when the unit is proven enough to scale — and when scaling will accelerate the collapse.

Financials dimension GTM dimension Source: CB Insights · 431 failures · 2023–2026

The core principle

A

What "unsustainable unit economics" actually means

Most founders frame unit economics as a spreadsheet problem. It's actually a timing problem — the unit is only unsustainable if you scale before it's proven.

›

1

CAC (Customer Acquisition Cost)

The fully-loaded cost of acquiring one customer — including all marketing spend, sales salaries, tools, and attributed overhead. Not just ad spend. Most founders undercount CAC by 30–50% by excluding salaries. Your CAC number is almost certainly wrong unless you've included every cost that wouldn't exist without the acquisition effort.

2

LTV (Lifetime Value)

The total net revenue from a customer over their lifetime with you. The critical word is "lifetime" — not what they've paid so far. LTV based on less than 12 months of cohort data is a guess. Churn patterns frequently accelerate at the 12–18 month mark in ways that early cohorts don't reveal. Don't use LTV to justify CAC until you have at least two full cohort cycles of data.

3

Payback period

How many months of gross margin from a customer it takes to recover their CAC. This is the most actionable unit metric because it tells you how much capital you need to grow. A 6-month payback means you need 6 months of working capital per customer to grow. A 24-month payback means you're essentially lending money to your customers — and you'll run out of cash before they pay it back.

!

The scaling trap

The CB Insights pattern: companies raise capital to scale growth, growth increases CAC (channel saturation, less-qualified leads), margins compress, LTV assumptions prove optimistic, payback period extends, runway shortens. They raise again to cover the hole — and the second raise buys 22 months before death (the CB Insights median). Every dollar of growth capital poured into a broken unit makes the eventual reckoning worse.

B

The unit economics readiness test

Run this before committing to any growth spend increase. Five questions — they'll tell you whether the unit is ready or broken.

›

Q1What is your fully-loaded CAC — including all salaries, tools, events, and agency costs attributed to acquisition? If your answer changed when you added salaries, recalculate before proceeding.
Q2What is your payback period in months? (CAC ÷ monthly gross margin per customer.) If it's over 18 months, stop here — the unit needs fixing before the scale conversation.
Q3What does your LTV calculation assume about churn — and how many months of data is that based on? Any assumption based on less than 12 months is a guess.
Q4As your customer count has grown, has your gross margin per customer gone up, stayed flat, or gone down? Declining margins at scale is the single clearest indicator the unit won't survive growth.
Q5If you doubled growth spend next month, what would happen to your CAC? If you don't know — or if the honest answer is "it would go up" — you don't have a repeatable acquisition channel yet.

C

Fix-it hierarchy — broken unit economics have four root causes

The fix depends on which part of the unit is broken. Don't treat all unit economics problems the same way.

Use only after B identifies the gap

›

1

CAC too high → channel or conversion problem

You're paying too much to acquire. Either your channel is expensive (move to organic, referral, or lower-CAC channels), your conversion rate is low (qualification, demo, or sales process problem), or you're counting too many leads as qualified. Fix the funnel before increasing spend.

2

Payback period too long → pricing problem

Even if CAC is reasonable, if monthly gross margin is too small, payback extends painfully. This is almost always a pricing problem. You are undercharging. A 30% price increase with 10% customer loss usually improves unit economics dramatically — and the customers you lose were probably the least valuable ones anyway.

3

LTV too low → retention or expansion problem

Customers aren't staying long enough or expanding enough to justify acquisition cost. Churn before value realization is an onboarding problem. Churn after 12+ months is a product depth problem. Neither is fixed by acquiring more customers faster.

4

Margins deteriorating at scale → structural business model problem

If unit economics worsen as you grow, you have a structural issue: either your costs scale with revenue (services business masquerading as SaaS), your channel is saturating, or you're discounting to hit growth targets. No amount of operational efficiency fixes a structurally broken model. This requires a strategic decision about the business model, not a tactical fix.

D

The CB Insights benchmark — what the data shows

Source: CB Insights · 431 VC-backed failures · 2023–2026

›

→

19% cited unsustainable unit economics as a primary cause

This makes it the 4th most common failure cause — behind capital running out (70%), poor PMF (43%), and bad timing (29%). Importantly, unit economics failures appear disproportionately at later stages: companies that raised enough to scale a broken model before it became visible.

→

Median time from last raise to death: 22 months

This is the signature of a unit economics death spiral: the raise provides enough runway to scale the problem, but not enough to fix it. By month 22, the company is out of options. The SKIN signal: a company raising "to grow" without a clear unit economics proof point is often buying 22 months, not 22 months of growth.

→

Fintech skews heavily: median equity of $4M vs. $11M dataset-wide

Fintech unit economics failures tend to be earlier-stage — companies that never proved the unit before scaling acquisition. The implication: you don't need to be heavily funded to have a unit economics problem. Early-stage fintech, marketplace, and consumer subscription companies are particularly exposed.

Playbooks / CB Insights Data / 13

Market Timing

29% of recent VC-backed failures cited wrong market timing or adverse macro conditions. It's the second biggest failure cause — and the hardest to diagnose because being early and being wrong look identical from the inside for the first 18 months. This playbook gives you the diagnostic framework to tell them apart.

Market dimension Product dimension Source: CB Insights · 431 failures · 2023–2026

The core distinction

A

Early vs. wrong — the most consequential misdiagnosis in startups

Founders who are early think they're right but unlucky. Founders who are wrong think they're early. The difference determines everything — especially whether to persevere or pivot.

›

Signs you're early (right market, wrong moment)

Customers recognize the problem immediately · Manual workarounds exist — they're solving it badly · You can name a specific forcing function that will change urgency · A few early adopters are genuinely passionate · The problem's importance is growing, not static

Signs you're wrong (wrong market or wrong problem)

Customers don't recognize the problem without a long explanation · No existing workarounds — it's not being solved at all · No forcing function you can name specifically · Even your best users are lukewarm · The problem is hypothetical, not lived

!

The CB Insights pattern — sectors that got timing wrong at scale

CB Insights found timing failures concentrated in climate & energy, food & agriculture, and blockchain — sectors that attracted heavy capital in 2021–2022 on trends that never materialized. New Age Meats ($32M) and RECUR ($55M) raised at the peak of alt-protein and NFT waves respectively. The lesson: sector consensus is not market validation. A crowded fundraising thesis can masquerade as market timing confirmation right up until it doesn't.

B

The timing diagnostic — five questions

These questions force the distinction between "early" and "wrong." Answer them honestly — the temptation is to rationalize toward "early" every time.

›

Q1Can you name one specific event — regulatory change, technology shift, macro disruption — that has made this problem more urgent in the last 24 months than it was before? "The market is growing" is not an event. Name the event.
Q2How are customers solving this problem today, right now, without you? If the answer is "they're not — they don't recognize it as a problem," that's a wrong-market signal, not an early-market signal.
Q3When you pitch the problem — before showing the product — what percentage of your target buyers say "yes, this is painful for us right now"? If it's under 30%, the urgency isn't there yet.
Q4Is there a specific buyer persona — a specific role, company size, or situation — for whom this problem is urgent right now, even if it isn't for most of the market? If yes, that's your beachhead. Start there.
Q5What would have to be true for this market to be 5x more receptive in 3 years? Is that future plausible — and can you survive until it arrives?

C

Responses by diagnosis

The intervention depends entirely on which timing problem you have. Don't apply the same fix to a wrong-market company that you'd apply to an early-market company.

›

A

Early market — right problem, not yet urgent

Find the 10% of the market for whom it IS urgent right now — usually a specific company size, sector, or situation. Serve only them. Build deep reference customers in that segment. When the broader market catches up, you'll have the credibility and the product depth. Don't try to educate the whole market. You'll run out of money before they catch up.

B

Wrong timing — real problem, adverse macro

If the forcing function you were counting on (regulation, technology unlock, behavioral shift) hasn't arrived, your options are: extend runway to wait it out, pivot to a version of the problem where the forcing function has already arrived, or find adjacent markets where the macro is favorable. Don't raise growth capital against a macro that hasn't turned — you're borrowing time, not buying growth.

C

Wrong market — the problem doesn't exist at scale

This is the hardest verdict. If customers don't recognize the problem, no amount of waiting fixes the timing. This is a pivot situation — not to a different moment, but to a different problem. Use what you've built as an asset: your distribution, your team's expertise, your existing customer relationships. Find a problem those assets uniquely position you to solve.

D

Right timing — execution is the variable

When competitors exist and budget exists and customers recognize the pain — timing isn't the problem. Stop diagnosing and start executing. The window is open but not permanent. Focus on winning a specific segment before the market consolidates around 1–2 dominant players.

D

SKIN scoring implications

Market dimension · 18% weight

›

→

Green signals on Market timing

Founder can name a specific forcing function · Competitors exist (validates timing) · Customers solve this problem today with workarounds · Specific early-adopter persona identified · CB Insights or Dealroom sector momentum data supports the window

→

Red signals on Market timing

No existing workarounds — market unaware of problem · Forcing function is a macro trend, not a specific event · Sector raised heavily in 2021–2022 on trends that stalled · Founder's timing thesis depends on "the market will catch up" · Similar companies have failed citing timing in the last 24 months

Full report object

skin_report.jsonOutput schema

// SKIN Pulse Report — v1.0 schema { "report_id": "rpt_[uuid]", "generated_at": "2026-03-27T09:00:00Z", "company_name": "Acme Inc.", "analyst_version": "skin-os-v1.0", "review_type": "full_pulse", // full_pulse | express | re-pulse "skin_score": { "composite": 76, // Weighted average, 0–100 "mode": "SCALE", // PROBE | BUILD | LIFT-OFF | SCALE | OUTLIER "confidence": "HIGH" // HIGH | MEDIUM | LOW — based on evidence quality }, "dimensions": [ { "id": "team", "weight": 0.20, "score": 84, "confidence": "HIGH", "strengths": ["...", "..."], // 2 items, evidence-cited "risks": ["...", "..."], // 2 items, evidence-cited "signals": { "green": ["fmf", "track_record", "access"], "red": ["credential_sub"] } } // ... dimensions 2–7 follow same structure ], "priorities": [ { "rank": 1, "dimension": "gtm", "action": "Define ICP by situation, not firmographics", "rationale": "Sales motion has no repeatable pattern...", "horizon": "30_days" // 30_days | 90_days | 6_months } ], "kpis": [ { "metric": "Sales cycle length", "target": "≤ 45 days", "current": "~90 days (estimated)", "dimension": "gtm" } ], "mode_narrative": "The company is directionally correct...", "flags": ["gtm_founder_dependent", "financials_burn_unclear"], "reviewer_notes": "..." // Optional human analyst notes }

Field reference

Field definitionsRequired

skin_score.composite integer Weighted average of all 7 dimension scores. Computed — never manually set. Required

skin_score.mode enum Derived from composite score. PROBE (0–39) · BUILD (40–59) · LIFT-OFF (60–74) · SCALE (75–89) · OUTLIER (90–100). Required

dimensions[].confidence enum AI confidence in the score based on evidence quality. LOW when fewer than 3 signals have supporting evidence. Required

priorities[].horizon enum Action timeframe: 30_days (critical), 90_days (important), 6_months (strategic). Max 1 action per horizon per report. Required

flags[] string[] Machine-readable risk flags from the signal rubric. Used to trigger playbook recommendations in the OS. Optional

reviewer_notes string Human analyst overlay. Added after AI draft. Takes precedence over AI narrative when present. Optional

Intake object

skin_intake.jsonInput schema

{ "intake_id": "int_[uuid]", "submitted_at": "2026-03-27T08:00:00Z", "company": { "name": "Acme Inc.", "founded": "2024-06", "stage": "seed", // pre-seed | seed | series-a "sector": "b2b-saas", "revenue_arr": 180000, // USD, 0 if pre-revenue "headcount": 4 }, "documents": { "deck_url": "s3://skin/[id]/deck.pdf", "team_bios_url": "s3://skin/[id]/team.pdf", "financials_url": null, // Optional "other_urls": [] }, "founder_responses": { // Indexed by question ID from knowledge base "team_q1": "I spent 8 years in industrial procurement before...", "team_q2": "We mis-hired a head of partnerships in January...", "product_q1": "We help ops managers at mid-sized manufacturers..." // ... all active questions per dimension }, "nda_accepted": true, "review_tier": "standard" // standard | priority }

Dimension definition object

skin_dimension.json — example: teamInternal schema

{ "id": "team", "index": 1, "name": "Team", "weight": 0.20, "signals": { "green": [ { "id": "fmf", "label": "Founder–market fit", "description": "Lived the problem personally or professionally", "weight": 0.25, // within dimension "rubric": { "90_100": "Founding story is the problem. Deep domain immersion, 5+ years.", "70_89": "Strong proximity to problem. Personal experience or operator background.", "50_69": "Adjacent experience. Logical but not intimate.", "30_49": "Opportunity identification from outside. No lived experience.", "0_29": "No relevant experience. Pure market opportunity play." }, "source_questions": ["team_q1"] } // ... other signals ], "red": [ { "id": "opp_founder", "label": "Opportunity founder syndrome", "penalty": -12, // Applied to dimension score if present "detection": "Cannot explain personal reason for this specific problem" } ] }, "questions": [ { "id": "team_q1", "text": "Why are you uniquely positioned...", "maps_to": ["fmf"] }, { "id": "team_q2", "text": "Describe a decision you made wrong...", "maps_to": ["conviction"] } ], "failure_modes": ["opp_founder", "undeclared_ceo", "credential_sub"], "playbooks": ["hire_vs_wait", "first_sales_hire"] // Linked playbooks }

End-to-end pipeline

Processing stagesActive

01

Intake

Form + docs

→

02

Parse

Doc extraction

→

03

Gap detect

Missing signals

→

04

Score

7 × dimension AI

→

05

Synthesize

Composite + flags

→

06

Human review

Analyst overlay

→

07

Deliver

Report + playbooks

Stage 02 — ParseAI

InputPDF deck, team bios, founder response text

ActionExtract structured data per intake schema. Flag missing fields.

OutputPopulated skin_intake.json with confidence scores per field

Stage 04 — ScoreAI

InputIntake JSON + dimension schema + signal rubrics

Action7 parallel API calls, one per dimension. Each returns score, evidence, flags.

Output7 × dimension score objects, ready for synthesis

Stage 05 — SynthesizeAI

Input7 scored dimensions + company context

ActionCompute composite, assign mode, generate priorities (max 3), select KPIs, trigger playbook links.

OutputComplete skin_report.json draft

Stage 06 — Human reviewHuman

InputAI-generated report draft

ActionSKIN analyst reviews scores, adjusts where evidence warrants, adds reviewer_notes. Uses score engine UI.

OutputFinal approved report. SLA: within 2 working days of submission.

Document parser — Stage 02

prompt:doc_parserStage 02

You are a SKIN intake analyst. Extract structured startup data from the submitted documents. TASK: Parse the uploaded pitch deck and team bios. For each field in the skin_intake schema, extract the best available value. If a field cannot be determined from the documents, set it to null and add it to a "gaps" array. RULES: - Extract facts only. Do not infer, interpolate, or assume. - Quote the source text in an "evidence" field alongside each extracted value. - Flag any claim in the documents that appears unsubstantiated. - Revenue, headcount, and ARR figures must be taken verbatim — do not normalize. OUTPUT: Return only valid JSON matching the skin_intake schema. Include a "parse_confidence" field (HIGH/MEDIUM/LOW) per top-level object. Do not include prose, preamble, or explanation.

Dimension scorer — Stage 04 (per-dimension)

prompt:dimension_scorerStage 04 × 7

You are a SKIN analyst — an experienced startup operator scoring a company against the SKIN framework. You have operator judgment, not consultant hedging. CONTEXT: Dimension: {{dimension_name}} (weight: {{dimension_weight}}%) Signals rubric: {{dimension_signals_json}} Founder answers: {{founder_responses_for_dimension}} Parsed deck data: {{relevant_deck_fields}} TASK: 1. Score this dimension 0–100 using the rubric. - Green signals pull the score up. Red signals apply their penalty. - Low confidence (missing evidence) should pull the score toward 50, not toward 100. 2. Identify the 2 strongest pieces of evidence (green or red). 3. Identify the 2 highest-risk items with specific citations. 4. List which signal IDs are present (green) and which red flags fired. CONSTRAINTS: - Never score above 80 if confidence is LOW. - Never infer intent from missing data — flag the gap. - Cite specific quotes or data points for every strength and risk. OUTPUT: Return JSON matching the skin_report.dimensions[] schema. No prose.

Report synthesizer — Stage 05

prompt:synthesizerStage 05

You are the lead SKIN analyst synthesizing a full Pulse report from 7 scored dimensions. INPUTS: Scored dimensions: {{dimensions_array}} Company context: {{company_object}} TASK: 1. Compute the composite SKIN Score (weighted average, round to integer). 2. Assign the correct SKIN Mode based on the composite. 3. Write a mode_narrative of exactly 2 sentences. Operator voice, no hedging. 4. Generate 3 prioritized actions — one per horizon (30_days, 90_days, 6_months). - Each action must target a specific dimension and cite evidence. - The 30-day action must be executable without additional capital or headcount. 5. Select 3 KPIs that are leading indicators for the top priority dimension. 6. Fire any applicable flags from the flag registry. 7. Link relevant playbooks from the playbook registry. CONSTRAINTS: - Priorities must be ordered by impact × urgency, not by score rank. - Mode narrative must use the company's specific situation — no generic language. - Do not repeat information from individual dimension summaries verbatim. OUTPUT: Complete skin_report.json. No prose outside the schema fields.

Failure pattern object schema

skin_failure_pattern.jsonInternal schema

// SKIN Failure Pattern Registry — v1.1 // Sourced from: Failory (400+), Startups.RIP (1,737+), Wefunder ($617M data) { "pattern_id": "fp_no_market_need", "label": "No market need", "source": "failory_cemetery", "case_count": 64, // Documented cases in source "frequency_rank": 1, // #1 most common across all sources "dimensions": ["product", "market"], "signals": { "intake_triggers": [ "users_engage_dont_pay", "churn_immediately_after_first_use", "zero_inbound_6_months", "discount_doesnt_increase_conversion" ], "question_responses": [ "Cannot name a user who would be very disappointed if product disappeared", "Churned users say they are doing nothing instead" ] }, "distinction_from": "fp_bad_marketing", // Often confused "playbook_link": "pb-nomarket", "score_impact": { "product": -25, "market": -15 } }

Full pattern registry

All 11 registered failure patternsFrequency ranked

#Pattern IDSourceDimensions affected · Playbook link

1fp_no_market_needFailoryProduct · Market → pb-nomarket

2fp_bad_marketingFailoryGTM · Brand → pb-narrative · pb-channel

3fp_competitionFailoryMarket · GTM → pb-category · pb-pivot

4fp_lack_of_focusFailoryGTM · Product → pb-channel · pb-price

5fp_lack_of_fundsFailoryFinancials · Investor Readiness → pb-raise

6fp_poor_productFailoryProduct → pb-churn · pb-nomarket

7fp_no_pivotingFailoryProduct · Market → pb-pivot

8fp_mismanagement_fundsFailoryFinancials → pb-raise · pb-hire

9fp_resilience_deficitYC / RIPTeam → pb-resilience

10fp_category_misclassDealroomMarket · Brand → pb-category

11fp_narrative_disconnectWefunderBrand · Investor Readiness → pb-narrative

Sector risk profiles — Startups.RIP pattern data1,737+ companies

SectorRisk levelPattern notes

Community / SocialVERY HIGHNetwork effects required. Timing-dependent. Most common "inactive" classification. → flag: fp_no_market_need + fp_competition

MarketplaceHIGHCold start problem. High CAC on both sides. Often acquired before profitability. → flag: fp_lack_of_funds + fp_competition

EducationHIGHLong sales cycles (institutional), low WTP in consumer. Many acquisitions in B2B sub-segment. → flag: fp_no_market_need + fp_bad_marketing

Consumer HardwareELEVATEDCapital intensity outpaces most founding teams. Low acquisition rate. → flag: fp_lack_of_funds + fp_mismanagement_funds

B2B SaaSMODERATEHighest acquisition rate in YC data. Sticky enterprise contracts. Main failure: GTM and ICP. → flag: fp_bad_marketing + fp_lack_of_focus

Developer ToolsLOWERStrong acquirability. Defensible IP. Main failure: monetization timing. → flag: fp_no_market_need (freemium trap)

FintechLOWERHigh acquisition rate. Regulatory risk as wildcard. → flag: fp_legal_challenges (add to registry if encountered)