Early-stage due diligence is not a box-checking exercise. It's a structured process for identifying the risks that matter most, understanding how the founders are managing them, and deciding whether the deal structure protects you if you're wrong.

This playbook is organized by risk domain. Within each domain, you'll find the key questions, the common failure modes they're designed to catch, and how to weight your findings when you're making a final decision.

How to use this framework

Not all risks are equal. Before you dive in, assign each domain a weight based on stage:

Domain	Pre-seed weight	Seed weight
Team	Very high	High
Market	High	High
Product	Medium	High
Traction / Financials	Low	High
Competition	Medium	Medium
Cap table & legal	Medium	Medium

At pre-seed, you're betting primarily on the team and the market. At seed, product-market fit signals and unit economics start to matter more. Adjust your time accordingly.

Domain 1: Team

The team is the single most important factor at early stage. Products pivot, markets shift, and timelines slip — but exceptional founders find a way through. Mediocre teams don't.

Key questions:

Why is this specific founder uniquely positioned to win this market? What do they know or have access to that a competitor can't easily replicate?
Have they worked together before? What's the co-founder relationship history? If solo, why couldn't they recruit a co-founder?
What did they do in their last 12 months before starting this company?
Have they failed before? How did they handle it?
Who do they go to for advice? Are those people actually accessible to them or just logos on a slide?

Reference check protocol:

Do not skip references. And do not only call the references they give you — call people who have worked with them or for them who are not on the list. LinkedIn makes this easy. The difference between a glowing listed reference and an unsolicited reference from a former colleague is often the difference between a great investment and a disaster.

Specifically ask references:

"What is this person like when things are going badly?"
"Did they take responsibility for failures or externalize blame?"
"Would you work for them? Would you invest alongside them?"

Common failure modes:

Charismatic but execution-light founders: Great pitchers who can't execute. Usually revealed by a career history of many short tenures or projects that never shipped.
Domain tourists: Founders who discovered the problem recently and are building on shallow understanding. Ask very specific operational or technical questions about the market. Tourists reveal themselves quickly.
Founder conflict risk: Co-founders who met recently, have no shared history, or have very different expectations about equity, roles, and decision authority. Ask: "How do you make decisions when you disagree?"

Domain 2: Market

The best team building the right product in the wrong market generates weak returns. Market selection is one of the most important and least discussed aspects of early-stage diligence.

Key questions:

Is this a genuine hairsocket problem — one that the target customer actively has, thinks about, and has budget to solve?
What is the existing solution? How is the customer solving this today?
What is the realistic serviceable market — not the theoretical TAM, but the set of customers the company can actually reach and sell to in the next three years?
Is the market growing, stable, or contracting? What's the secular tailwind (or headwind)?
Is there a regulatory, geographic, or structural moat that makes this market hard to enter — and if so, does this team have access to it?

Market size sanity check:

Build the number yourself. Take their claimed TAM, strip out every geography and segment the company isn't actually selling into, and divide by a realistic ACV. Compare to any public competitor revenue data you can find. If your bottoms-up number is less than 30% of their claimed TAM, that's a meaningful signal.

Common failure modes:

Regulatory markets the team doesn't understand: Healthcare, financial services, and government procurement have acquisition cycles and regulatory requirements that kill startups who underestimate them. Ask detailed questions about the sales cycle and regulatory pathway.
Platform-dependent markets: Building on top of a platform (Amazon, Shopify, Salesforce) that could shut you out with a product update or policy change. How platform-dependent is the business model?
Markets where the entrenched buyer is also the incumbent: Enterprise markets where the buyer and the incumbent vendor have a relationship that goes back 10 years and has a $2M switching cost. Startups often underestimate how hard it is to displace these relationships.

Domain 3: Product

At pre-seed, the product may barely exist. At seed, you should see something real — and you should use it.

Key questions:

What does the product do that existing solutions don't? Is this a 10x improvement or a 30% improvement?
Who built it? Is the technical co-founder or CTO equity-holding or salaried? (Big difference in long-term retention risk.)
What is the product's core technical advantage? Is it reproducible by a well-funded competitor?
What is the product's biggest current weakness? (Watch how they answer — founders who can't name one are either not self-aware or not honest.)
What does the product roadmap look like for the next 12 months? Is it driven by customer demand or internal engineering preference?

Trial the product:

If the product exists and you're putting in meaningful capital, use it. Sign up for a trial, complete a workflow, ask your team to use it. You will learn more in 20 minutes of use than in two hours of slides.

Common failure modes:

Vendor-dependent architecture: Core functionality built on a third-party API (OpenAI, Twilio, Stripe) where a pricing change or policy update could destroy margins or break functionality. This is particularly acute for AI-native companies in 2026 where model costs are a major COGS line.
Technical debt at the foundation: A v1 built for speed that can't scale. Usually surfaces as "we need to rebuild the infrastructure in the next 12 months" in the roadmap. Ask: what does that rebuild cost in time and headcount?
Product-founder fit but not product-market fit: The founders love the product. Early design partners gave polite feedback. But the product hasn't been stress-tested against real price-sensitive buyers who have alternatives.

Domain 4: Traction and Financials

At seed, you should have data. If you're looking at a pre-seed deal with $0 revenue, adjust this section accordingly — the questions still apply, but you're diligencing the hypothesis rather than the evidence.

Key questions:

What is MRR, and what is the month-over-month growth rate over the last six months?
What is the churn rate? Monthly churn above 3% is a warning sign for SaaS. Monthly churn below 1% is exceptional.
What is NRR (net revenue retention)? Above 120% suggests the product delivers ongoing value. Below 100% means the bucket leaks.
What is CAC by channel? How many months to CAC payback?
What is the current burn rate? What is the runway from this raise?
What are the top three line items in COGS, and how do they scale?

Financial model review:

Ask for the model in a spreadsheet format, not just as slides. Look at the assumptions tab. The assumptions reveal the founder's mental model of their business — which is often more informative than the outputs. Specifically:

What growth rate is assumed, and what justifies it?
What headcount is assumed, and when are key hires made?
What happens to gross margin at 3x current revenue?
What event triggers Series A, and is it achievable on this runway?

Common failure modes:

Revenue from related parties: Friends, family, or advisors who "paid" for the product as a favor. Ask for customer names and whether any customers have a relationship with the founders outside of the product.
Pilots that haven't converted: Five pilots in progress, zero paid customers. Pilots are not traction. When do pilots convert? What has the conversion rate been historically?
Burn that doesn't match headcount: High burn with a small team suggests founder compensation, contractor spend, or infrastructure costs that aren't being highlighted. Ask for a headcount-to-burn reconciliation.

Domain 5: Competition

Founders consistently underestimate competition. Your job is to find what they've missed.

Key questions:

What does the customer currently use to solve this problem? (The answer is always a competitor — even if it's a spreadsheet or an incumbent's legacy tool.)
What are the strongest one or two competitors, and what do they do better than this company right now?
Is there a platform player (Salesforce, Microsoft, ServiceNow, AWS) that could ship this feature as an add-on? What would that do to the market?
What is the company's defensible moat in 24 months? Network effects? Proprietary data? Regulatory access? Switching costs?

Do your own competitive research:

Search the problem, not the solution. Google "how do [target customers] manage [problem]" and see what tools, workflows, and providers appear. Talk to one or two people in the target customer role at non-portfolio companies. Twenty minutes of customer research will often surface a competitive dynamic the founder hasn't mentioned.

Common failure modes:

Well-funded stealth competitors: Larger companies with similar products that are 6–12 months from launch. Check Crunchbase, recent conference demos, and job postings (a company hiring aggressively in a specific product area is often building in that direction).
Open-source alternatives: In infrastructure and developer tools especially, open-source alternatives can gut the addressable market. Ask: "What is the open-source equivalent and why don't customers use it?"

Domain 6: Cap Table and Legal

This is the domain most investors rush through because it feels bureaucratic. It is not. Cap table problems are one of the most common reasons good companies can't raise Series A.

What to request:

Full cap table, including any SAFEs, convertible notes, and option pool
All SAFE and note conversion terms (cap, discount, MFN, pro-rata)
Any side letters with existing investors
Founder vesting schedules and cliff dates
Any IP assignment agreements (especially if founders worked at a prior employer in a related area)

Key questions:

Are all founders on standard 4-year vesting with a 1-year cliff? Founders with no vesting or fully vested equity create an enormous flight risk.
Are there any shares held by departed co-founders, employees, or early investors who could create friction in future rounds?
Are there any outstanding convertible instruments with punishing terms (full ratchet anti-dilution, high multiple liquidation preferences) that will make Series A difficult?
Has the company properly assigned all IP from founders and early contractors? (Standard PIIA agreements should be in place for everyone who touched the product.)

Common failure modes:

Departed co-founder with uncancelled equity: Extremely common. A former co-founder who left after three months but kept 20% is a liability that will surface at Series A. Ask directly: "Has anyone left the company? What happened to their equity?"
Angel investor terms that block future rounds: Some early angels negotiate terms (super pro-rata rights, approval rights on future rounds) that make them impossible to bring along in future financings. Surface these early.
IP ownership gaps: Code written by a contractor who never signed an IP assignment agreement. A founder's core technology that they built while employed at a previous company. These create legal exposure that can derail acquisitions and later rounds.

Putting it all together: the risk-weighted decision

After completing diligence, score each domain on a simple 1–5 scale (1 = serious unresolved concern, 5 = exceptional) and weight by stage appropriateness.

Then ask three questions:

What is the single biggest risk here? Can I live with it? Have I structured my terms to protect against it?
What would have to go wrong for this to be a total loss? How likely is that scenario? What early warning signs would I watch for?
What would have to go right for this to be a fund returner? What has to be true about the team, market, and product, and how confident am I in each?

If you can answer these three questions clearly and the deal still makes sense, you have done your job. If you can't answer them, you need more information — regardless of how much pressure you're feeling to decide.

The investors who generate consistent early-stage returns are not the ones who move fastest. They are the ones who know exactly what they're buying — and what they're not.

PitchVault's AI deck analysis surfaces the structural risks in a pitch before you spend hours on diligence calls. Start reviewing deals →