Building an angel portfolio: why one great pick is the wrong plan
The maths of early-stage returns, why twenty cheques beat two, how to size cheques to your tax position, follow-on strategy, sector spread, and the discipline that separates angels who make money from angels who have stories.
Early-stage returns are not like the stock market. Most investments return nothing; a few return everything. An angel who makes two investments is gambling; an angel who makes twenty, sized properly, is running a portfolio with a real expected return. The difference is not skill at picking — it is discipline about numbers.
The distribution
Across large samples of UK angel investments, roughly half to two-thirds lose most or all of the money, a quarter return somewhere between 1× and 3×, and under one in ten return 5× or more. The 5×-plus cheques produce essentially all of the portfolio's profit. You cannot reliably tell in advance which ones they will be — experienced angels are a little better than chance, not a lot. So the plan is to own enough positions that you hold some of them.
How many, and how much
Twenty positions over three to five years is the usual target. Fewer than ten and the outcome is mostly luck. Decide the total you are prepared to have in early-stage companies — for most people 5–10% of investable assets — and divide by twenty. That is the cheque. If the cheque comes out below £5,000, you are better served by an EIS fund than by direct investment.
Sizing to your tax position
SEIS relief at 50% is only worth having if you pay enough income tax to use it. Work out your income tax for the year, double it if you will use carry-back, and that is the maximum useful SEIS investment. Above it, EIS at 30%; above that, the relief is gone and the arithmetic changes. Most angels invest a SEIS slice every year at the start of their tax planning and fill EIS afterwards.
Reserve for follow-ons
The companies that reach a second round will want you again, and the second cheque is better-informed than the first: you have seen eighteen months of updates. Keep a third to a half of your total back for follow-ons into the companies that are working. Angels who deploy everything in first cheques cannot back their winners and get diluted out of them.
Spread
Across sector — not all software, not all consumer. Across stage — some pre-seed bets on people, more seed with revenue. Across time — a few a year, not twenty in the year the market was hot. Across instrument — some straight equity, some secured loan notes in profitable businesses or buy-outs for income while the equity matures. And within reason, across what you know: the investments where you understand the customer are the ones where your judgement is worth something.
The discipline
- Write down why you invested, in three lines, before you do. Read it at the next update.
- Never exceed the cheque size because you love the deal. The one you love is the one you are wrong about most often.
- Read every update. Reply to the asks. The best deal flow comes from founders who tell other founders you are useful.
- Count the portfolio, not the story. The total return across twenty matters; the one that went to zero does not.
- Expect nothing back for five years. Early-stage money is illiquid; plan as if it is gone.
Questions investors ask
"Should I lead, or follow?"
Follow until you have seen five rounds close. Then lead where you know the sector — see leading a round.
"What about syndicates and EIS funds?"
Both give spread for less work and less control, at a fee. Many angels do both: a fund for the base, direct cheques where they have an edge.
"How do I find twenty good deals?"
That is what Find is for — every listing is a UK business that has been reviewed, with documents and live Companies House data, and you choose who to talk to.