Diversification Strategy for Angel Investors: How Many Companies to Back
Data-driven framework for angel portfolio construction — optimal portfolio size, stage mix, sector focus, check sizing, and follow-on allocation by power law.
Whether you're looking for an angel investor, a growth advisor, or just want to connect — I'm always open to great ideas.
Get in TouchAI, startups & growth insights. No spam.
TL;DR: Most angels write 5 to 10 checks and call it a portfolio. The math says that is not enough. Power law return distributions in venture mean your median company returns zero, a few return modestly, and one or two return everything. If you do not have enough shots, the probability that your portfolio contains a real outlier drops to a coin flip or worse. This post covers how many companies you actually need, how to construct a portfolio by stage and sector, how to think about follow-on allocation, and what my own portfolio philosophy looks like after 38+ angel investments across six years.
Before we talk about how many companies to back, you need to viscerally understand what kind of distribution you are operating in. Angel investing does not look like a stock portfolio. It does not look like real estate. It does not look like any asset class most people have previous experience with.
The return distribution in early-stage venture is a power law. Not a bell curve. Not a normal distribution where most outcomes cluster around the mean. A power law, which means the vast majority of your returns — and I am talking 80 to 95 percent of the total value you ever realize — will come from one or two companies in your entire portfolio.
Let me make that concrete. Here is what a realistic angel portfolio return distribution looks like across 25 investments:
| Outcome | Number of Companies | Return Multiple | % of Total Portfolio Value |
|---|---|---|---|
| Total loss (0x) | 13 | 0x | 0% |
| Returned capital (roughly breakeven) | 5 | 0.5x – 1.5x | ~4% |
| Modest return | 4 | 2x – 5x | ~12% |
| Strong return | 2 | 10x – 30x | ~24% |
| Outlier (fund returner) | 1 | 50x – 200x | ~60% |
That last row is not a lucky scenario. It is how angel portfolios are supposed to work if they work at all. The outlier company — the one that returned more than your entire fund — is the portfolio. Everything else is context.
This has a critical implication that most angels absorb intellectually but do not internalize emotionally: your job is not to pick winners. Your job is to ensure you are in the game long enough and broadly enough that when a winner emerges, you are in it.
"The biggest mistake I see angels make is not losing money on bad investments. It is having no position in the company that ended up being the breakout. They passed on it, or took a small check that diluted away, or concentrated their capital in deals that felt safer."
The mathematics bear this out. If the probability of any given early-stage investment returning 50x or more is roughly 1 in 30 — which is consistent with what the data from First Round Capital, AngelList, and other large datasets suggests — then you need to make enough investments to statistically capture at least one of those outcomes. With 10 investments, your probability of having even one outlier in your portfolio is about 28 percent. With 25 investments, it rises to about 57 percent. With 50 investments, it reaches about 82 percent.
Those numbers should bother you if you are running a 10-company portfolio and expecting venture-style returns.
The power law is not an opinion. It is an empirical property of venture returns that has been measured consistently across datasets covering thousands of early-stage investments over multiple decades. The Kauffman Foundation's research on angel returns, the Cambridge Associates data on venture fund performance, and the AngelList portfolio analytics all point to the same structural reality: the top decile of outcomes in any early-stage portfolio generates a disproportionate share of total returns, and that effect becomes more pronounced — not less — as you move earlier in company formation.
What this means in practice is that the strategy question for angel investors is not "how do I identify the winners?" It is "how do I maximize the probability that I am invested in the winners when they emerge?" Those two questions sound similar but lead to very different portfolio behaviors.
The first question leads you toward higher conviction, fewer investments, and more time on diligence. It is how smart, experienced people with a prior investing framework get into trouble in angel investing, because they apply analytical frameworks built for later-stage investing — where fundamental analysis is more predictive — to a domain where the future is genuinely unknowable.
The second question leads you toward appropriate breadth, disciplined initial check sizes, and a follow-on reserve that lets you concentrate into the companies that prove themselves after the fact. That is the structure that actually works.
This is where angels genuinely disagree, and the debate is not purely theoretical — it has real consequences for how you structure your time, your capital, and your relationships with founders.
The 20-company camp argues that portfolio size beyond this point creates meaningful costs that offset the mathematical benefit of broader diversification. Their argument:
You cannot maintain meaningful relationships with more than 20 portfolio companies. And if you cannot add value, you should not be investing. Due diligence quality degrades as deal volume increases. At 50 investments, you are effectively pattern-matching on pitch decks rather than doing real analysis. At 20 companies with $25K checks, that is $500K deployed — a reasonable annual commitment for a high-income angel investor. Signal value matters. Being selective about what you invest in affects which deals you see. If you say yes to everything, the best founders stop bringing you their best opportunities.
There is a version of this argument that I respect, particularly the point about value-add. If you have a very specific, high-leverage skill — say, you ran enterprise sales at a category-defining SaaS company and your active involvement genuinely moves the needle for B2B SaaS founders — then concentrating your capital and time into fewer companies where you can be deeply useful is a legitimate strategy.
The problem is that most angels overestimate their value-add. They believe their intros, their advice, and their operational knowledge will meaningfully accelerate their portfolio companies. Sometimes that is true. More often, what angels actually provide is social proof, a warm intro to one or two people, and occasional sounding board conversations. That is valuable, but it is not so scarce that you can only provide it to 20 companies.
The 50-company camp has the math on its side, at least in aggregate, and argues several things:
The uncertainty at the pre-seed and seed stage is so high that any pretense of picking winners is hubris. The data consistently shows that the best professional VCs cannot reliably predict outliers at the seed stage. Angels, with less deal flow and shorter runways on diligence, are even less equipped to do so.
With 50 companies at $10K to $25K per check, you cover the probability distribution adequately while remaining within the capital budget of an affluent professional.
A larger portfolio creates a compounding network effect. Each founder introduces you to other founders. Each investment is a warm referral to the next. The deal flow flywheel rewards size.
You can still add meaningful value to your 5 to 10 most relevant portfolio companies even in a 50-company portfolio. You do not need to be equally valuable to all of them.
This is roughly where I land. The counterintuitive insight that pushed me toward this position: the angels in my network with the best track records are not the ones who claim to have picked the best companies from a small set. They are the ones who made enough investments that the probability of catching at least one exceptional company was high, and then provided genuine support to the companies where they could actually help.
This is the AngelList syndicate lead position, and some angels have built significant track records here. The argument:
At $5K to $10K checks across 100 companies, you are running something closer to an index fund with an information advantage. You are capturing the tail distribution in full. The absolute capital required is manageable at $500K to $1M over several years.
The counterintuitive point: a 100-company portfolio with concentrated follow-on in the best performers often produces better returns than a 20-company portfolio with uniform sizing, because you are maintaining optionality to concentrate after the information asymmetry has been partially resolved.
The challenge at this portfolio size is that your influence on any individual company becomes negligible. You are a name on a cap table, not a genuine stakeholder. Some founders do not mind this. Others actively prefer it — a clean, simple investor relationship with no obligations. But it does limit the deals you see, because founders who are choosing between similarly-priced angels will generally prefer one who brings more than just capital.
I sit firmly in the 50-company camp with a structured follow-on reserve. I have seen too many angels with 15 or 20 company portfolios confidently declare that their portfolio is "working" because they have two or three companies growing nicely — and then a year or two later, those same companies return modest multiples while the company they passed on raises a Series B at a $400M valuation.
The variance at early stage is too high to bet on your own judgment. What you can do is make enough bets that the math works in your favor, and then use follow-on capital to concentrate into the companies that prove themselves with actual traction.
| Portfolio Size | Probability of 1+ Outlier | Capital Required at $25K avg check | Value-Add Capacity |
|---|---|---|---|
| 10 companies | ~28% | $250K | High |
| 20 companies | ~49% | $500K | Medium-High |
| 50 companies | ~82% | $1.25M | Medium |
| 100 companies | ~97% | $2.5M | Low per company |
Not all investments are created equal. The stage at which you invest fundamentally changes the risk profile, the typical check size, the expected return multiple, and the timeline to liquidity. A thoughtful portfolio construction approach has to account for stage as a structural variable, not just an afterthought.
Pre-seed is where binary risk is highest. Most of the companies you invest in at this stage will not make it to a seed round. The ones that do will generally raise at a valuation that gives your initial check a paper markup, but the path from seed to anything resembling liquidity is still years away.
The case for pre-seed investing is compelling on the math side. Valuations are lowest here, which means the potential multiple is highest if the company succeeds. You often have the most access — if you have a strong network with operators, founders, and accelerators, the best pre-seed deals flow to you through warm intros before they are broadly marketed. You can sometimes get into companies at valuations below $5M or even below $3M, which creates 100x or more in return potential if the company reaches a $300M to $500M outcome.
The case against heavy pre-seed concentration is equally real. The failure rate is extraordinarily high at this stage. You are investing in an idea, a team, and early signals — often before there is any meaningful product validation. Pre-seed companies need the most support. They need help hiring, fundraising, and making basic operational decisions. That is fine if you have capacity and genuine expertise. It is a drain if you are writing 40 checks a year across multiple stages and sectors.
My general recommendation: pre-seed should represent 30 to 40 percent of your total deployment by number of companies, but only 20 to 25 percent of your total capital. This means writing smaller checks at this stage — $10K to $20K — into more companies. You are buying optionality and relationship-building, not making concentrated bets.
Seed stage is the sweet spot for most angels. There is enough signal to distinguish between teams that can execute and teams that cannot. Revenue — or at least meaningful usage data — is often present. The founding team has demonstrated they can build and ship something. But the valuations are still low enough that a successful outcome returns real multiples.
Most angel portfolios should be weighted toward seed. My target is 45 to 50 percent of the portfolio by number of companies and 50 to 55 percent of total capital. This is where I write my largest initial checks — $25K to $50K — and where I am most deliberate about spending time on diligence before investing.
The seed stage also tends to have the best balance of signal quality and valuation. Pre-seed is noisy because there is so little data to evaluate. Series A and beyond compress the potential multiple because the company is already priced for some of its success. Seed sits in the middle: enough signal to make a real judgment, still early enough to get genuine upside if you are right.
Some angels argue you should never invest at Series A because the return multiples are too compressed. A company raising at a $30M to $50M valuation would need to reach a $300M to $500M exit just to return 10x on your initial check, and the probability of that outcome is lower than at earlier stages simply because more capital is required to generate the needed appreciation.
That argument is mathematically correct in isolation. But there are reasons to have some Series A exposure in an angel portfolio.
Lower variance is a real benefit. A company that has raised a Series A from a credible lead investor has cleared a significant filter. The failure rate drops materially. If you are trying to smooth out the volatility of your portfolio, a small Series A allocation acts as ballast against the binary outcomes at pre-seed.
Access to deals you could not enter earlier is another reason. Some of the best companies become visible as great companies only after they have broken through at the seed stage. If you see a Series A company with genuine product-market fit and a clear path to $100M in revenue, passing because the multiple looks lower may be a mistake.
Portfolio signaling matters too. Being in a founder's round at Series A — even at a small check — opens doors to their Series B and to their network of founders who are pre-seed or seed stage.
My target: 10 to 15 percent of total capital at Series A, concentrated in a small number of companies where I have high conviction and specific knowledge.
Here is how I think about stage allocation for a $1M angel portfolio:
| Stage | % of Companies | % of Capital | Avg Check Size | Target Return Multiple |
|---|---|---|---|---|
| Pre-Seed | 35% | 22% | $15K – $25K | 50x – 200x (on winners) |
| Seed | 50% | 55% | $25K – $50K | 20x – 100x (on winners) |
| Series A | 15% | 23% | $50K – $100K | 5x – 25x (on winners) |
One nuance worth noting: these percentages describe intention at deployment time. In practice, if your pre-seed companies mature and you exercise pro-rata rights at their seed and Series A rounds, the effective stage distribution of your portfolio will shift toward later stages over time. That is not a problem — it is the follow-on strategy working as intended.
This is one of the most contested questions in angel portfolio theory, and the honest answer is that both approaches can work — but for different reasons, in different hands.
The argument for sector concentration is essentially an information advantage argument. If you spent 10 years in enterprise sales and then became an angel investor, you have a real edge evaluating B2B SaaS companies. You know what a good enterprise sales motion looks like, what the competitive dynamics are in various subcategories, and which founders understand buyer psychology versus which ones are guessing.
That information edge is real and valuable. Your diligence is better — you catch things that generalist angels miss. You see better deals, because founders in your sector actively seek you out knowing you understand their world. You add more value through introductions, feedback, and operator advice that is actually useful to your portfolio companies. Your thesis is testable: if you are right about a specific market shift, you capture returns across multiple companies in that sector.
Several of the best-performing angel investors I know are deeply sector-focused. One writes exclusively into climate tech hardware companies. Another focuses entirely on fintech infrastructure. Both have outperformed their peers with comparable capital deployed, primarily because their deal access within those sectors is exceptional and their diligence is genuinely better than what a generalist investor could do.
The argument against concentration is a correlation argument. If you have 30 investments in a single sector and that sector goes through a downturn — whether because of regulatory change, a market shift, or simply because the macro cycle turns — your entire portfolio suffers simultaneously.
The regulatory crackdown on crypto in 2022 and 2023 is an example. The collapse of BNPL consumer sentiment when interest rates rose. The pullback in edtech companies post-pandemic as schools reopened and the emergency tailwind disappeared. Each of those sector-specific shocks hit concentrated portfolios hard.
There is also a hubris check embedded in the diversification argument. Sector expertise does not guarantee you identify the right companies within that sector. It may actually make you overconfident in your ability to predict which companies will win, causing you to pass on outliers that do not fit your mental model of how the sector should develop.
I am primarily focused on B2B SaaS and AI infrastructure, which is the domain where I have the most direct operator experience. Roughly 70 to 75 percent of my portfolio is in these sectors.
I deliberately keep 20 to 25 percent of my portfolio in sectors I understand less well — healthcare technology, fintech, and climate. This forces me to be more honest about what I am evaluating when I write those checks, and it creates correlation protection in the portfolio.
The one sector I actively avoid as a standalone investment theme is consumer social. The failure rate is extraordinarily high, the network effects that drive success are impossible to predict, and the market timing dependency is brutal. I have made one exception in six years and I would not repeat it.
How much you write into the first check matters, but how you think about follow-on capital matters more. This is where most angels get the math badly wrong.
Your initial check should be sized so that you can make your target number of investments without running out of capital before you have deployed across a full portfolio. If your target is 40 to 50 companies and you have $1.5M to deploy over three to four years, your average first check should be in the $15K to $25K range.
The mistake I see constantly: angels writing $100K checks into their first few companies because each deal feels compelling in isolation, then running out of capital before they have built a real portfolio. They end up with seven to ten companies and no reserve — the worst of both worlds. They have sacrificed portfolio breadth without gaining the concentrated ownership percentage that would make the concentration worthwhile.
The even bigger mistake: not reserving follow-on capital at all. An angel who deploys 100 percent of their capital in initial checks has no ability to double down into their best performers when the information has substantially improved. They are locked into the exact position they had when uncertainty was highest.
Initial check sizing should also be consistent across deals. I have a range and I stay in it regardless of how excited I am about a particular deal. The times I have broken this rule and written a larger first check because the opportunity felt exceptional, I have generally regretted it — not always because those companies failed, but because concentrating at the beginning forecloses optionality.
Reserve 30 to 40 percent of your total fund for follow-on investments. Deploy it selectively, with a high bar.
The logic is straightforward: follow-on investments are made with substantially more information than initial investments. By the time a company in your portfolio is raising its next round, you know whether the founding team can execute under pressure, whether the product is actually solving a real problem (customers are either renewing or they are not), whether the go-to-market motion is working, and who else wants to invest in this company and at what valuation.
That information asymmetry compresses dramatically between pre-seed and seed, and again between seed and Series A. When a company you invested in at pre-seed is raising a strong seed round from credible institutional investors, the probability of a meaningful outcome has gone up substantially. You should be concentrating capital into those companies.
The practical rule I use: every company in my portfolio that raises a subsequent round with meaningful institutional validation gets evaluated for follow-on. I invest in roughly 30 to 40 percent of those companies on follow-on, typically maintaining my pro-rata or a meaningful fraction of it.
Here is how a $1M angel portfolio with follow-on discipline might look:
| Deployment Category | Capital Allocated | Purpose |
|---|---|---|
| Initial checks (40 companies) | $600K | Build the portfolio broadly |
| Follow-on reserve | $400K | Concentrate into best performers |
| Weighted follow-on (15 companies) | $400K | Avg $25K–$30K per company |
The result is a portfolio where you started at $15K in 40 companies but ended up with $40K-plus positions in the 15 that proved themselves. That is the structure that gives you both the breadth — statistical probability of catching an outlier — and the depth that creates meaningful ownership in the company when it actually works.
If you can negotiate pro-rata rights in your initial investment, do so. Pro-rata rights give you the contractual option — not obligation — to invest your proportional ownership percentage in subsequent rounds. This is one of the most valuable things an early-stage investor can secure.
The reason: as a company succeeds and raises larger rounds, the institutional investors leading those rounds will often want to keep the cap table clean by limiting small angels from following on. Pro-rata rights override this dynamic. They are not always honored in practice, but they are worth negotiating for.
At very early stages — pre-seed SAFEs — pro-rata rights are often not included as a standard term. You can sometimes negotiate them as a side letter. At seed and above, they are more commonly included for angels who write meaningful checks.
Every time someone advocates for portfolio diversification in angel investing, someone else says: "That sounds like spray and pray." I want to address this directly because the critique, while emotionally resonant, is mostly wrong. The 5 checks I run before writing a check is precisely the diligence framework that separates thoughtful diversification from careless spraying.
"Spray and pray" as a pejorative describes investing in companies without any real diligence, any thesis, any pattern recognition, or any value-add — simply deploying capital as fast as possible into anything that shows up in your inbox. That is a real thing and it produces poor results.
But diversification is not the same as spray and pray. The distinction matters enormously.
Spray and pray: Write 100 checks with no diligence, no thesis, and no follow-on framework. Forget about 95 of them. Hope something works.
Thoughtful diversification: Write 40 to 50 checks with consistent diligence standards, a clear thesis for each investment, meaningful support where you can genuinely add value, and a disciplined follow-on reserve deployed selectively into the best performers.
The second approach does not feel as intellectually satisfying as "I carefully picked 10 companies and all of them are winners." But it produces better returns in expectation for the simple reason that the uncertainty at early stage exceeds anyone's ability to reliably pick winners in small samples.
There is an irony here worth pointing out: the "spray and pray" critique is most often made by angels who have written 10 or 15 checks and want to feel good about that decision. It is a rationalization of concentration dressed up as a critique of diversification. The implication is that "my investments are thoughtful; other people's diversified portfolios are undisciplined." This is almost always backwards. A 10-company portfolio that happened to include one strong performer looks smart in retrospect. It was not necessarily a better strategy at the time of construction.
The real question is not how many companies you invest in. It is whether you are applying consistent standards, actually thinking about each investment on its merits, and building a follow-on process that concentrates into the companies that deserve it. You can do all of that with 50 companies. You can fail to do any of it with 10.
"I would rather be right about the structure of the opportunity than right about any individual bet. The structure says: make enough bets, maintain your standards, follow on aggressively into the winners. Everything else is noise."
There is also a practical counterargument to the "spray and pray" critique that is rarely made explicitly: the cost of saying no to a company that becomes a breakout is permanent. The cost of saying yes to a company that fails is bounded. At a $15K check, losing the whole thing is painful but survivable. Missing the company that 100x's is devastating to your total return — and you may never know you missed it because you will not be getting those investor update emails.
After 38+ angel investments over six years, I have developed a fairly specific view of how to construct a portfolio that gives you a real shot at venture-style returns without requiring you to quit your job to manage it.
Here are the principles that actually guide my decisions:
I do not invest in anything I cannot explain in writing. Not because the thesis will be right — most of the time, the way a company succeeds looks very different from how I expected when I invested — but because writing the thesis forces me to be honest about what I am actually betting on. Am I betting on the team? The market timing? The specific technical insight? The go-to-market wedge? All of these are legitimate bets. "This just feels right" is not a thesis. It is a feeling that I have learned not to trust at face value.
This is consistent with my view on the power law math. I want the statistical probability that I have caught at least one outlier to be above 75 percent. That requires roughly 45 to 50 companies at typical early-stage success rates. I am not perfectly at target — I am at 38 across six years, which means my pace has been slower than planned — but the direction is right and I have the follow-on reserve intact.
I have a range — currently $15K to $35K for pre-seed and $25K to $50K for seed — and I stay in it regardless of how compelling any individual deal feels. The times I have broken this rule and written a larger first check, I have generally regretted it. Not necessarily because those companies failed, but because concentrating at the beginning forecloses optionality and exposes me to more variance than the information at that stage warrants.
The threshold for follow-on is simple: has this company raised a subsequent round with meaningful institutional support, and is the core thesis still intact? If yes, I follow on. I do not over-engineer this decision. The institutional round is a validation signal I am relying on rather than trying to out-think.
If a founder is asking me to invest because they have been rejected everywhere else, that is information. It might be wrong information — genuinely great companies get rejected at seed all the time. But it is information that should raise my bar substantially, not lower it. If the best validators in the market do not see what I see, I need a very specific reason why I have an informational edge they do not. Usually I do not.
I use a simple spreadsheet that tracks every investment, the current round based on update emails or public announcements, whether I have follow-on rights, what my pro-rata would be at the next round, and my current mark-to-market valuation. I review this quarterly. I do not look at it every week because early-stage investing works on a 7 to 10 year timeline, and short-term noise is not information.
This is not pessimism. It is calibration. If you build an angel portfolio expecting most of your companies to succeed and produce meaningful returns, you will make bad decisions: holding off on follow-on rounds because the earlier companies have not returned yet, selling secondary positions too early because you are nervous, or adding more capital to clearly failing companies because you cannot psychologically accept the loss. Accepting the loss structure in advance frees you to play the long game that early-stage investing requires.
Two underappreciated dimensions of portfolio construction that most angels do not think about systematically: geography and team composition.
Most angel investors default to investing heavily in their local market — wherever they live and work. This is rational from a deal flow perspective: warm intros come from your existing network, and your network is geographically concentrated.
But it creates a portfolio that is more correlated than it appears. If you are a Bay Area angel and your portfolio is 80 percent San Francisco companies, you are exposed to Bay Area talent costs, Bay Area real estate impacts on employee quality of life, Bay Area regulatory dynamics, and the Bay Area fundraising market. When things are going well in the Bay Area, everything in your portfolio looks fine. When the market tightens — as it did materially in 2022 and 2023 — every company in your portfolio faces the same headwinds simultaneously.
I try to keep no more than 50 percent of my portfolio in a single geography. In practice, I have companies in San Francisco, New York, London, Toronto, Bangalore, and a handful of other markets. The international exposure has added operational complexity — different regulatory environments, currency considerations, time zone management — but has also introduced me to companies and founders I would never have encountered if I had stayed local.
One specific note on emerging market investing: the risk profile is genuinely different, not just cosmetically different. Regulatory unpredictability, currency risk, talent market dynamics, and exit market availability all look quite different in Southeast Asia or Sub-Saharan Africa compared to the US or Europe. I am not saying do not invest there — some of the most interesting opportunities are in markets that are underserved by capital — but do it with eyes open about the specific risks involved.
Early in my angel investing, I was largely pattern-matching on team composition based on what I had seen succeed before: two co-founders, one technical, one business, both with prior operator experience in the relevant industry. That pattern works. But it is not the only pattern that works, and treating it as a hard filter causes you to miss things.
Some of the strongest performers in my portfolio have had solo founders who compensated for the partnership gap with unusually clear vision and fast hiring decisions. Others had three-person founding teams that included a deep domain expert alongside the technical and business co-founders. Several had founding teams with no prior startup experience but strong industry relationships that unlocked enterprise contracts other companies simply could not have gotten.
The more useful lens than team composition pattern-matching is this: does this founding team have a specific, defensible advantage in the market they are entering that goes beyond "they are smart and hardworking"? Smart and hardworking is a table stake, not an edge. What is the edge that this specific team has in this specific market at this specific moment that makes them the right people to build this company? Building an advisory board is one way founders with narrower networks compensate for distribution gaps early on.
Diversity in team composition within your portfolio also creates valuable cross-portfolio learnings over time. If you have invested in 15 technical solo founders, 15 traditional two-person co-founder pairs, and 10 non-technical CEOs with strong CTO hires, you are building real data on what patterns work in what contexts. That pattern recognition makes you a better investor over time.
Everything above advocates for broad diversification at initial deployment. But there are circumstances where concentration is the right answer, and I want to be honest about when those are.
If you built and sold a company in a specific vertical, and a founder in that exact vertical with a closely related approach is raising a round, you have information that the broader market does not. You know the buyer, the sales cycle, the likely objections, and the competitive dynamics from the inside. That is a fundamentally different situation from the average angel investment.
In this case, writing a larger check — and potentially taking a board seat or advisory role that lets you actively accelerate the company — may produce better returns than splitting that capital across five investments where you have no comparable edge.
I have done this twice in six years. Both times, the companies I concentrated into had specific technical or market characteristics I understood deeply from prior operator experience. One has produced a paper markup of roughly 8x on the position at the last round valuation. One returned capital at roughly 1x in an acqui-hire. The concentration risk was real in both cases; I got reasonably lucky on the outcome.
Follow-on is the one place in angel investing where concentration is not just acceptable — it is mandatory. If a company in your portfolio has proven its thesis with real traction and is raising a round at a valuation that still implies significant upside, you should be concentrating aggressively into that company.
This is the mechanics of how professional early-stage fund managers make their money. They cast wide nets at the seed stage and then write meaningful checks into the 15 to 20 percent of companies that prove themselves. The initial checks are an access mechanism for the follow-on investment that actually drives the return.
If you have already built a 50-company portfolio and you have two or three companies that are clearly outperforming, the calculus for new capital changes. Rather than deploying additional capital into new companies — adding more of the same diversification you already have — it may make sense to buy secondary shares in your top performers if secondary market access exists, or to deploy the marginal dollar into follow-on in your best companies rather than new initial positions.
This is a second-order portfolio management decision that most angels never reach because they do not track their portfolios with enough rigor to see the pattern clearly.
The one thing I want to be very clear about: concentration as a response to conviction is different from concentration as a response to loss aversion, and it is different from concentration as a response to a compelling pitch.
When an angel writes a large initial check into a company because it "feels different" from all the other companies they have seen, what they are usually doing is rationalizing an emotional response to a compelling pitch. The companies that have the most compelling pitches are not systematically the companies that produce the best returns. Some research suggests a mild negative correlation — the companies that are easiest to explain and most emotionally resonant in a pitch tend to be slightly overcrowded with capital and slightly overvalued relative to the expected outcome, while the companies that are harder to pitch are undercapitalized and cheaper.
Do not concentrate because a founder gave you the best pitch of the year. Concentrate because you have actual, verifiable information that distinguishes this company from the broader opportunity set.
One of the most underappreciated operational disciplines in angel investing is the portfolio review cadence. Most angels track their investments poorly — they wait for update emails from founders, mentally add up paper gains when a company announces a new round, and have only a vague sense of how their overall portfolio is performing at any given time.
Here is the review system I actually use, which I have refined over six years:
Log any update emails, press coverage, or news about portfolio companies. Note any companies that have been silent for more than 90 days — sustained silence from founders is often an early signal of trouble. Respond to any founder requests for help, intros, or advice.
This is not a deep analytical exercise. It is maintenance: keeping your finger on the pulse of the portfolio without letting it consume your week.
Pull the full portfolio spreadsheet. For each company: what is the current round, what is the last reported mark-to-market valuation, do I have pro-rata rights at the next round, have I heard from the founders in the last 90 days, and based on what I know, has the investment thesis changed?
The quarterly review is also when I make follow-on decisions. If a company is raising a round and I have pro-rata rights, I decide whether to exercise them within the quarterly window. I try not to let these decisions sit because the best rounds close fast and waiting means losing your allocation.
Once per year, I do a deeper review that includes sector allocation — am I overexposed to any single sector? Stage allocation — how is my follow-on capital deployed relative to initial checks? Return attribution — which companies are driving my current portfolio value, and what does that tell me about my thesis accuracy? And new deal pipeline — based on what I have learned, are there thesis updates I should make for the next year of deployment?
The annual review is also when I update my written investment thesis for the year ahead. After six years and 38+ investments, I have learned specific things about what predicts success and failure that I did not know when I started. The thesis should evolve as the data comes in.
This is the hardest part of portfolio management. When a portfolio company is clearly struggling — low growth, high burn, difficulty fundraising, team attrition — most angels do one of two things: ignore it (emotionally easier) or write a bridge check out of loyalty and loss aversion (emotionally compelling but often economically irrational).
The right framework is harder than either of those. Have an honest conversation with the founder about the current state. Assess whether the core problem is fixable with your help, your money, or your introductions. If yes, consider a bridge — but only with a clear, time-bound thesis for what changes and how you will know whether the intervention worked. If no, accept the loss and stop spending time on it.
Time is your most limited resource as an angel investor, not capital. Spending 10 hours per quarter on a failing company that is going to fail regardless is 10 hours you are not spending on your best companies, your deal pipeline, or your relationships with founders who have not yet pitched you.
The discipline to accept losses cleanly is one of the hardest things to develop in angel investing. It runs against every instinct of optimism and loyalty that makes you a good investor in the first place. But it is essential for portfolio hygiene, and the angels I know who have built genuinely strong track records have all figured out how to let go of the failures without excessive hand-wringing. The portfolio company failure case study is one honest walk-through of exactly what that looks like in practice.
How much total capital do I need to build a real angel portfolio?
The honest answer is $500K to $1M deployed over three to four years. That gives you enough to make 30 to 50 initial investments in the $10K to $25K range and maintain a meaningful follow-on reserve. Some angels build real portfolios with less — AngelList and syndicates have lowered the minimum check size in many deals to $1K to $5K — but the relationship quality and pro-rata rights you get at those levels are typically minimal. Below $500K total, you are more of a participant in the venture ecosystem than a portfolio manager within it. That is still valuable as a learning exercise, but you should be honest with yourself that you are not yet running a portfolio with the statistical properties that make the power law work for you.
Should I join a syndicate or invest independently?
Both, ideally. Syndicates give you access to deals you would not otherwise see, particularly if you are earlier in your investing career and have not yet built a network that generates strong inbound deal flow. Independent investing gives you more control over check size, timing, and the depth of your relationship with founders. A reasonable approach: use syndicates for 20 to 30 percent of your initial deals in sectors where you have less direct network access, and invest independently in the deals where your network gives you genuine access to strong companies. Over time, as your network strengthens, shift more of your activity to independent investing where you are not paying carry to a syndicate lead.
How do I know if my portfolio is performing well before exits happen?
The short answer is: you mostly cannot, and anyone who tells you otherwise is either lying or measuring the wrong thing. Paper valuations from subsequent rounds are a reasonable signal but are not realized returns. What you can track as leading indicators: which of your portfolio companies are raising subsequent rounds from credible institutional investors, which are growing revenue or usage in a way that suggests real product-market fit, and which are retaining their founding teams and key early employees. Companies that score well on all three of those dimensions two or three years after your initial investment have a substantially better probability of producing real returns than companies that are struggling on any of them.
What is the right number of new investments per year?
If you are targeting a 50-company portfolio built over four years, that implies roughly 12 to 15 new investments per year — approximately one per month, which is a reasonable pace for someone doing angel investing as a serious side project rather than a full-time job. More than 20 new investments per year starts to challenge your ability to do genuine diligence and maintain meaningful relationships with founders after the investment. Fewer than 8 per year means it will take you six or more years to build a portfolio with adequate statistical coverage.
Should I invest in companies where I know the founder personally?
Personal relationships with founders create both advantages and risks. The advantage: warm relationships give you information about the founder's character, work ethic, and intellectual honesty that you simply cannot access from a three-meeting pitch process. The risk: personal relationships create pressure to invest when you should pass and to hold on when you should accept a loss. My rule: personal relationships should move a founder from the "I would consider looking at this" pile to the "I will make time to look at this carefully" pile. They should not move a company from a no to a yes in the absence of actual investment merit.
How do I think about SAFE notes versus priced rounds when structuring angel investments?
Most pre-seed and early seed investments are structured as SAFE notes because they are fast, cheap to document, and founder-friendly. The main risk of SAFEs for angels is that your ownership percentage is not determined until the subsequent priced round, and if that round happens at a high valuation or with a large raise, your effective ownership can be significantly diluted. The practical implications: prefer SAFEs with a valuation cap that reflects reasonable terms rather than an uncapped SAFE, which is almost always a bad deal for the angel investor. A $10K investment at a $5M cap is meaningfully different from a $10K investment at a $20M cap even if both are described as "pre-seed."
When should I sell secondary shares rather than holding to full liquidity?
Secondary share sales in private companies have become more accessible in recent years through platforms and direct LP secondary transactions. The general principle: if a company in your portfolio has grown significantly and you are offered the opportunity to sell 25 to 50 percent of your position at a price that represents a meaningful return on your initial investment, taking that liquidity is usually rational. The argument for holding everything to primary liquidity — acquisition or IPO — assumes that the current valuation significantly underestimates the eventual outcome. Sometimes that is true. Often it is not. Realizing partial gains when they are available is not a failure of conviction; it is good portfolio management.
What is the most common mistake you see angels make with portfolio construction?
Two mistakes tie in my experience. First: deploying too much capital too early, in too few companies, leaving nothing for follow-on in the best performers. This is the angel equivalent of going all-in on the first hand of a long poker game. Second: letting loss aversion drive ongoing decisions — holding on to clearly failing companies beyond the point where any intervention will matter, and spending disproportionate time on the bottom of the portfolio instead of the top. Both mistakes stem from the same underlying issue: not having a clear, written framework for how you make decisions before you are in the middle of an emotionally charged situation. The framework does not have to be rigid or complex. It just has to exist, and you have to actually look at it when you are tempted to deviate from it.
How do you handle founder relationships when a company is failing?
With directness and care, in that order. I have been on the receiving end of investor silence during difficult periods as a founder, and I know how corrosive it is. If a portfolio company is struggling, I try to have honest conversations with the founder about what I am seeing and what my capacity to help looks like. Sometimes that means making introductions, helping with a bridge round, or working through a strategic pivot. Sometimes it means telling a founder that I think the company is not going to work and that they should think carefully about their options, including winding down. Those conversations are hard but they are respectful. What is not respectful is ghosting a founder because their company is going through a difficult period.
I have made 38+ angel investments over six years across pre-seed, seed, and Series A rounds. The views here reflect my personal experience and what I have observed across the broader angel investing community. Nothing in this post constitutes investment advice, and past patterns in angel investing do not guarantee future outcomes. Early-stage investing involves substantial risk of loss.
A rare look inside a 38-company angel portfolio: the tracking system, vintage analysis, wins, losses, and the decision framework that evolved over seven years.
A brutally honest case study of a portfolio company failure — what we missed, what went wrong, and what I changed about my investing process after.
A post-mortem across 38+ angel investments: the 3 patterns behind every win, 3 patterns behind every loss, and the risk framework that changed everything.