Stay at the Table

The volume ends here. Founders often ask us: after all these approaches, which line do you actually want me to remember?

Our view first. Every approach in this volume comes down to one line: staying at the table matters more than winning any one hand. Why? Three layers below.

Layer one: the move used most at the table is the fold

There are only four moves at the table: fold, check, raise, all-in. The one we use most is always the first.

That is not caution, it is arithmetic. Glacier does twenty or thirty mandates a year, half new and half existing clients coming back for the next round — not a large number by the standards of this business. Small on purpose. Each mandate runs sixty days across eight milestones, from D01 building the fact base to D60 closing the funds, and at every milestone a partner has to be sitting at the table. Three partners across twenty or thirty mandates is around ten each a year, which means holding two or three tables at once every month. Judgement, research, coordination and a partner's attention are all non-renewable. Spread across a hundred projects, it is a broken promise to all hundred.

So we drew ourselves a line: people can grow in number, projects cannot. More people is added capacity. More projects dilutes promises already made. We have held this line for eight years.

Drawing that line means missing some good companies.

So does missing count as a hole? It counts as a cost, not a hole. A man carrying only three tonnes of supplies cannot go in every direction, but he gets to the pole. Repeat mandates from existing clients are half of what we do, and that scorecard cannot be gamed — the market votes for us again every year.

Layer two: winter always comes, and we stock for winter

Markets turn. We do not predict which year they turn. We stock on the basis that they will.

Stocking means keeping headroom while nothing has gone wrong yet. Leave room for the mistakes that will certainly happen, leave time for the processes that will certainly run long. Scott's team broke down on the way back — the reason it broke down was packed in the bags before they set out. If you cannot get back, however fast you went out does not count.

The same holds on the company side. While a technical path is still undecided, the real moat for the companies out in front is often not the specifications, it is the cash in the bank. As we see it, whoever can keep the team fed and outlast a long cycle is still sitting at the table. So fundraising is sometimes the business itself — take the bullets, not the pose.

So what counts as finishing the course? Look at how much headroom is left.

And once it is burnt through? You leave the table.

Layer three: stay present and your turn will come

In our view, returns inside a single cycle depend heavily on luck and timing, and nobody times it right run after run. Which is to say, what decides the long-term result is not which year looked brilliant, it is survival. Whoever stays present will meet the wave that belongs to them; nobody knows which year. This one is our conjecture, but we act on it.

It is also what we ask of ourselves: to be a respected FA with no failed cases that never gives up. No failed cases is the result, never giving up is the reason, and each line proves the other. What a founder really fears is not failing to raise this round, it is being put down halfway. Take it on and it has to work. So we take on sparingly.

There is one more boundary to make clear: we serve deeply, but we do not take everything on. Whatever makes up the company's long-term capability, the company carries itself; whatever is friction in executing the transaction, we absorb. Draw that line badly and both sides get hurt.

The road ahead is still the steep North Slope. Hard to climb, and the view is like nothing else.

We take few because we intend to come back. The next volume is about the people who do this work.