Most allocation tables are designed to close a funding round, rather than to survive the subsequent market. These are two different tasks, and only the initial one is completed.
The table is written during the fundraising process, when the only thing that matters is what your investors will sign: a percentage that looks substantial, a lock-in period that looks committed, and a valuation that holds the round together. Every number is chosen based on what the other side will agree to.
Then the round closes and the counterparty changes. The same numbers now face a market that can sell on the same day โ a different test to that of a partner who expected to wait years. The investor allocation was sized against dilution rather than the depth needed to sell into the market, and the cliff was set to demonstrate commitment rather than to coincide with a potential sale.
Nobody notices, because the table still adds up and the percentages still look fine. It's not the document that has changed, but the question: it used to be about what investors would agree to, and now it's about how much this market can absorb.
The solution is an order, not a number. Work out what the market can absorb before negotiating what anyone will receive, ensuring that the raise happens within these limits rather than setting them. Then set a date to re-derive the table before anything is deployed.
A design that was right for the raise is not necessarily wrong; it is simply out of date. You will find this out when changing it is no longer free.