Every seed fund spends its first years reporting on itself with numbers that are not yet facts. There is no way around it. A fund that started writing cheques two years ago has companies, it has costs, and it has no exits. What it reports instead is a mark, and a mark is an estimate.
That is not a criticism of the practice. It is just worth knowing what you are holding, because a lot of people read the mark as performance when it is closer to a status update.
What a markup is, and what it is not
When a portfolio company raises again at a higher price, the fund revalues its position and the paper multiple moves. TVPI goes up. Nothing has been sold. No money has come back.
That number is one investor's opinion, on one day, about one company. It is real information. It is not the same kind of information as a distribution.
It also cuts both ways, which is the part that gets skipped. Fast markups can be a sign of hype rather than progress, particularly when the operating performance underneath has not moved much. A company that has been repriced twice in eighteen months while revenue and retention sat still is a company whose next round has a higher bar to clear, not a safer one. An LP who congratulates a manager for that is rewarding the wrong behaviour, and managers notice what gets rewarded.
Why the marks are quieter in Southeast Asia
Marks move when somebody reprices the company. That requires a next round, and next rounds are less frequent here than the headlines suggest.
Early stage capital in Southeast Asia is scarce across the board. Angel networks are thin, grants are fewer and more onerous in practice than they look on paper, and there are not many real accelerators. A good company in Kuala Lumpur or Manila can be growing properly and still go eighteen months without an event that generates a new number.
So a Southeast Asia seed fund tends to look quiet on paper for longer than a fund investing where capital is dense. That quiet is not underperformance. Read the other way, the same scarcity is what keeps entry prices rational, and entry price is one of the few parts of the eventual return that gets fixed early and never improves later.
Four things you can read before there is any DPI
Entry price and ownership. Ask what the fund bought, not what it invested. A first cheque figure means little on its own. The ownership it purchased, and the valuation it purchased it at, is the part that compounds into the return. This is inspectable today, and it is a straight test of whether a manager has price discipline or just enthusiasm.
Access. At the first cheque, the companies worth owning are settled in conversations that never circulate. Ask where the last several investments came from, and whether the answer describes a relationship or a process. Access decides whether the manager ever gets to apply their judgement.
Operating progress inside the companies. This is the closest thing to truth available before exits. Not valuations. Customers added, revenue retained, a second market entered, gross margin that holds as the company grows. A manager who can talk fluently about these across the portfolio is watching the right things. One who reaches for the valuation history is not.
What happens at the next round. Reserve policy and follow-on discipline decide whether the good positions survive dilution. Ask what proportion of the fund is held back, what triggers a follow-on, and what the manager does when a company they like raises at a price they think is wrong.
The uncomfortable part
For the first several years you are not underwriting a portfolio. You are underwriting a manager.
That is an unsatisfying thing to be told, because manager quality resists a spreadsheet. But it is the accurate description of what a commitment to an early fund is. The companies are too young to have proved anything. What you can assess is whether this person can reach good founders before anyone else, whether they pay a price they can defend, whether they are useful to a company between rounds, and whether they will say no to something attractive.
Those four things are visible in a conversation, and they are far more predictive than a paper multiple in year two.
What to do with this
If you are looking at a seed fund now, try reading its most recent report in a deliberate order. Skip the headline multiple. Go to the company progress first and form a view. Then look at what the fund paid to own what it owns. Then, last, look at the marks and ask whether they are consistent with what you just read.
If the operating story and the marks tell you the same thing, the marks are probably honest. If the marks are running well ahead of the operating story, you have learned something more useful than a number.
And if a manager cannot tell the story without reaching for the valuations, that is the finding.
Next time a fund sends you a report, what do you read first?