The beginner question about reserves is whether to hold any capital back at all. If you've answered yes, the harder question comes straight after: how much, and how do you model the split so it holds up across a whole book? That is what a reserve ratio is for.
The Carry's take: a reserve ratio is really a bet on your own follow-on discipline, and the common failure is holding powder that never reaches the winners it was meant for.
A reserve ratio sounds like jargon, and at heart it's just one number: how much you keep back for later rounds against how much you put into the first cheque. But that single number quietly decides how many companies you can start with, how hard you can back the ones that work, and whether the power law ever gets a chance to pay you. It's worth modelling on purpose rather than discovering it by accident.
The ratio is one number. It decides how many bets you place, and how hard you can back the few that win.
What is a reserve ratio, and why does it exist?
A reserve ratio is the share of a planned allocation an angel holds back for follow-on rounds in the companies they already back, framed against the size of the first cheque. It's the number that turns a vague intention to keep some powder dry into an actual plan for how the book gets built.
The reason it exists sits in the shape of the returns. Early-stage outcomes follow a power law: a few names carry almost the whole book while most return little or nothing. If that's the distribution, the ability to put more money into the handful that are working is worth a great deal, and a reserve is how you keep it. The beginner-level case for holding capital back is covered in our piece on reserves and dry powder. This is the layer above it: not whether to reserve, but how much, and how to model the split.
How much do angels typically reserve?
There's no correct figure, and nothing here prescribes one. In practice the framings you hear run along a spectrum: at one end, reserving roughly as much as you put into the first cheque; at the other, holding back two or three times it for your strongest names. These are illustrations of how angels talk about the split, not a target to copy.
The mechanics are simple arithmetic. Reserve nothing and the whole programme goes into first cheques. Hold back the same again and you're planning to back roughly the same companies twice. Hold back two or three times the initial cheque and the reserve becomes the larger part of the programme, with fewer first cheques at the top. The table shows how one notional £100,000 programme splits under each framing. Read it as a worked illustration of the arithmetic, not a steer about where you should land.
What are the trade-offs in a reserve ratio?
Every pound reserved is a first cheque you don't write, so the ratio is really a choice about the shape of the book. Reserve heavily and you get depth: fewer companies, more capital behind the ones that work. Reserve nothing and you get breadth: more shots on goal, but no way to lean into the winners when they raise again.
Three forces pull on the decision. The first is dilution. If you hold a stake and sit out the next round, your percentage shrinks as new shares are issued, and a pro-rata right is only worth as much as the reserve that funds it. The second is signalling: an existing backer following on tells a new lead something, and stepping back can say something too. The third is opportunity cost, the plainest of all. Money waiting in reserve for eighteen months is money not working in a new deal you'd happily do today. Reserves also ride the J-curve, the early stretch where a book looks like losses before the winners mark up; the reserve is what lets you add to those names on the way out of the dip. Writing that second cheque well is its own discipline, covered in follow-on investing.
Why do most angels get the reserve ratio wrong?
Most angels miss in one of two opposite directions. The first is reserving nothing, then watching a company that's clearly working raise at a higher price with no capital left to follow. The winner gets backed by someone else and the early stake dilutes down. This is the more common mistake, and it usually comes from spreading every pound across first cheques because a full portfolio feels like progress.
The second failure is the mirror image: reserving so heavily that the programme never gets enough first cheques into the ground to find a winner worth following. Capital sits idle, the book stays small, and the tail the reserve was meant to catch never appears because too few bets were placed. The portfolio maths is unforgiving here. If most investments return little and a small tail carries the rest, you need enough first cheques to have a real chance of holding one of the names in that tail before reserves can do anything for you.
How should you model a reserve ratio for your own book?
Model it against your own numbers rather than a rule of thumb, because the right split depends on things specific to you. The factors worth weighing include the total size of your programme, how many first cheques you need to give the power law a chance, your appetite for concentration, your own liquidity and time horizon, and how the tax reliefs land on any follow-on tranche.
That last point is worth a line. A follow-on is a fresh share issue, so any SEIS or EIS relief on it is judged on the company's eligibility at the time, not carried over from your first cheque. Confirm the current scheme rules at GOV.UK before you count on relief for a second tranche.
A closing note on what this is. This piece is general editorial information, not investment advice, and the ratios above are illustrations of how the arithmetic works, never a figure to adopt. How much you reserve is a decision for you and an FCA-regulated adviser, taken against your own circumstances. The one outcome worth avoiding is the accidental one: finding the reserve empty just as your strongest company opens its next round.
Frequently asked questions
How much should an angel reserve for follow-ons?
There is no correct figure, and this is not advice. Common framings run from reserving roughly the same as the first cheque up to two or three times it, but the right split depends on your programme size, appetite for concentration, liquidity and time horizon. Model it against your own numbers rather than adopting a ratio, and take FCA-regulated advice before committing capital.
What is a reserve ratio in angel investing?
A reserve ratio is the share of a planned allocation an angel holds back for follow-on rounds in companies they already back, usually expressed against the size of the first cheque. Reserving the same again as the first cheque is one common framing; some angels hold back two or three times it for their strongest names. It is a modelling choice, not a formula.
Should angels keep dry powder?
Dry powder is the capital held in reserve for later rounds. The case for it comes from the power law: because a few names carry most of the return, being able to follow the winners when they raise again can matter more than writing another first cheque. The cost is that reserved capital cannot also fund new deals, so it narrows how many companies you start with. Whether to hold it, and how much, is a personal decision.
What happens if I don't follow on?
If you hold a stake and skip the next round, new shares are issued and your percentage falls, which is ordinary dilution. A pro-rata right lets you invest again to defend your percentage, but only if you have reserved the capital to exercise it. Skipping a follow-on is a legitimate choice; the point of a reserve is to make it a choice rather than something forced on you by having run out of money.
Is this reserve-ratio guidance financial advice?
No. This is general editorial information, not investment or financial advice, and the ratios described are illustrations of how the arithmetic works, not figures to adopt. How much to reserve depends on your own circumstances. Confirm the current position at GOV.UK and speak to an FCA-regulated adviser before committing capital.