Angels and Micro-Funds: Running Real Diligence Without an Analyst Team
An angel or solo GP has the same job as a fund with fifteen people and none of the leverage. No analyst to screen, no associate to take reference calls, no platform team to maintain a CRM. Often it is not even a full-time role.
The instinct is to compensate with judgement and a narrow network. That produces a small, homogeneous portfolio and a process that quietly degrades whenever the week gets busy. This is the alternative: a system sized for one person.
Key takeaways
- Budget four hours a week, fixed, and design the process to fit inside it.
- The constraint is screening cost. Solve that and everything else becomes affordable.
- Never delegate or automate: founder reference calls, the thesis, the decision.
- Syndicates work when the division of labour is explicit, and fail when everyone assumes someone else did diligence.
- Write one page per decision, including passes. It is the only feedback loop you have.
The four-hour week
| Block | Time | Job |
|---|---|---|
| Screen | 60 min, once | Everything new since last week, thesis-fit then quality |
| Calls | 90 min | Two founder calls, 45 minutes each |
| Diligence | 60 min | Reference calls and verification on live deals |
| Admin | 30 min | Update pipeline, write memos, log passes |
Four hours, on the same day every week. The fixed slot matters more than the total: a process that only runs when you have spare capacity is a process that stops in month three, and the companies that arrive during your busy weeks are not systematically worse than the ones that arrive during your quiet weeks.
Make screening cheap or nothing else works
For a solo investor, screening is the binding constraint. At twenty minutes per company, sixty minutes covers three companies a week. That is 150 a year, which is not enough funnel to build a portfolio and forces you straight back onto warm intros - with all the costs that carries.
Three levers:
1. Publish a specific thesis. Ten lines: sector, stage, geography, cheque size, hard exclusions. Put it on your profile and your site. It pre-filters inbound before it reaches you, which is the cheapest screening available because someone else does it.
2. Use structured sources. A company arriving as a standard field set takes ninety seconds to screen. The same company as a deck takes twenty minutes. That single difference turns 150 companies a year into well over 1,000.
3. Apply a written gate. Four or five binary questions, applied identically. Judgement belongs in the 45-minute call, not in the 90-second screen. See pre-screening at scale.
What to automate, and what never to
| Automate | Never automate |
|---|---|
| Extracting facts from a company record | Founder reference calls |
| Normalising metrics into comparable units | The investment thesis |
| Flagging missing or contradictory information | The decision |
| Drafting the memo skeleton | Judging the team |
| Pipeline reminders and stage tracking | Deciding what risk you accept |
AI is genuinely good at the left column and produces a reliable first-pass summary from a structured record. It removes the blank-page cost that stops most solo investors writing memos at all.
It is not good at the right column, and the accountability for those cannot be delegated anyway. An AI-generated thesis reads plausibly and commits to nothing, which is the opposite of what a thesis is for.
Proportionate diligence for a solo investor
The full pre-seed checklist is a week of work. For a $25K angel cheque, compress to the four non-negotiables plus two conversations:
- Entity and good standing. Five minutes, registry search.
- Cap table reconciles. Fifteen minutes with the founder's own table.
- IP assignment confirmed. One question, one document. The most commonly skipped and most expensive omission.
- Two founder references. One hour. The highest-value hour in the process.
- One customer conversation. Thirty minutes.
- One metric verified at source. Thirty minutes.
Under three hours. It catches the structural problems that are genuinely fatal and skips everything that cannot yet be known.
The temptation is to skip steps 1 to 3 because they feel like paperwork and you trust the founder. They are not about trust. Founders lose track of their own cap tables routinely and rarely know whether a contractor from 2024 signed an assignment.
Syndicates: make the division of labour explicit
Syndicates are the natural leverage for angels, and they fail in a specific, predictable way: everyone assumes someone else did the work.
The lead is presumed to have done diligence. The lead presumes the members will do their own. Nobody calls a reference. The round closes on collective enthusiasm.
The fix is a stated split, written into the deal note:
| Role | Owns |
|---|---|
| Lead | Entity, cap table, IP, founder references, the memo |
| Domain member | The technical or market-specific verification |
| Members | Read the memo, ask one hard question each, decide independently |
And one rule that does most of the work: the memo states what was verified and what was not. A member reading "IP assignment confirmed; customer references not taken" can make an informed decision. A member reading "diligence complete" cannot.
Write one page per decision
Including passes. Especially passes.
Your investments are a biased sample of companies you liked. Your passes are the full distribution, and they are where any pattern in your judgement is actually visible. If you keep passing on a founder profile that keeps working, the pass archive is the only place you will ever see it.
Pass memos can be three lines: company, category, one-sentence reason. Use a controlled vocabulary - out of thesis, team gap, market too small, no demand evidence, deal structure, valuation, too early, cannot verify - so that after a hundred passes you can count them.
Almost no solo investor keeps this archive. It is the cheapest edge available.
What the system buys you
Run this for a year and you will have: a specific thesis that filters inbound, several hundred companies screened rather than a few dozen, a written record of every decision and its reasoning, and a portfolio that is not simply a map of your existing network.
The last one is the point. A solo investor's structural disadvantage is not capital or brand. It is that an unsystematic process defaults to proximity, and proximity is not a strategy.
Frequently Asked Questions
Under three hours for a small cheque: entity, cap table, IP assignment, two founder references, one customer call, one verified metric.
On speed and decisiveness, yes. On coverage, only by using structured sources that make wide screening affordable.
For extraction, normalisation and first-pass summaries, yes. Keep the thesis and the decision in human hands.
Enough that you have a basis for comparison. Fifty screened before a first cheque is a reasonable discipline.
Assuming diligence was done by someone else. Require the memo to state what was verified and what was not.
Read next
The Modern Dealflow Stack: From Inbox Chaos to a Scored Pipeline
Proportionate Diligence: A Pre-Seed Checklist That Is Not a Series A Checklist
The Investment Memo, Standardised: What Belongs in a Pre-Seed IC Memo
Adverse Selection: The Hidden Cost of Warm-Intro-Only DealflowRaiseTalks gives solo investors a filterable startup catalogue, an eight-stage dealflow board and a one-click AI pre-score on companies that have granted access, so a four-hour week covers a thousand companies a year. Start your 30-day trial.