When the capitalization table and the option ledger are placed side by side in an investment review, the characteristic finding is rarely that the two documents disagree; it is that nobody can say precisely where the difference resides. Between the number of shares recorded as granted in the formal pool and the entitlement employees believe they were told they hold, there typically sits an inventory of promises that has never been consolidated into a single record. That inventory carries, at once, a percentage spoken aloud in a hiring conversation, a line in an email undertaking that the position will be corrected at the next round, and an allocation named during a promotion discussion but never carried into a board resolution. The size of that inventory looks like a bookkeeping problem, yet the first signal it sends to the reviewing party originates elsewhere entirely: there is no single answer to the question of who determines eligibility, and according to what rule.

Asked at three different desks inside the same company, that question tends to produce three different answers, and this is the most reliably recurring form of the pattern. The founder ordinarily reasons along an axis of contribution and loyalty; the human resources side constructs a logic out of tenure and title bands; the finance side recognizes only those allocations tied to a board resolution and treats the remainder as nonexistent. That each answer is internally coherent is precisely what establishes that none of them is the company's operative eligibility rule. A plan document may well exist at this stage, drafted without legal defect; but what the document defines is the mechanics of the plan, not eligibility itself. Eligibility continues to be manufactured outside the document, one decision at a time, at the moment each decision is made.

The mechanism beneath this configuration is not negligence but a shortcut that genuinely lowers cost in the early period. With a critical hire days from closing, an option promise is the fastest instrument available — it generates no cash cost today, requires no legal preparation, and produces immediate effect on the other side of the table — whereas writing the rule in advance consumes time, advisory fees and board agenda. Keeping the promise bespoke also preserves negotiating flexibility, since a published rule creates precedent across every subsequent hire and narrows the bargaining range. The difficulty lies not in the shortcut itself but in its persistence after the underlying conditions have changed. Once headcount crosses a threshold and the company begins dealing with institutional capital, discretion that once functioned as flexibility converts into an unquantified liability.

It also becomes clear at this stage that eligibility is not a single layer. Within any plan, eligibility waits to be defined at no fewer than four distinct thresholds: who qualifies to receive a grant at all; through what vesting schedule and cliff period the entitlement matures; whether acceleration on a change-of-control event operates on a single trigger or a double trigger; and under what circumstances a departing employee is treated as a good leaver rather than a bad leaver. Where three of these four thresholds are written and the fourth is left to discretion, the plan is treated as discretionary in its entirety, because the weakest link determines the auditability of the whole structure. What the review desk is looking for is exactly this: four thresholds closed in a manner that is interlocking, internally consistent and independent of any individual.

The first channel through which the gap reaches valuation runs directly through the capitalization table. Placing the option pool inside the pre-money valuation is settled negotiating practice in institutional rounds; accordingly, every undocumented grant promise surfaced in diligence becomes a demand to enlarge the pool, and every enlargement of the pool becomes dilution of the existing shareholders. That dilution is invisible in the headline valuation, since the headline figure remains unchanged while the economic outcome per share moves. Uncertainty in the promise inventory further supplies the counterparty with a justification for conservatism; a liability that cannot be measured is typically priced at its upper bound. The absence of an eligibility rule therefore tends to result in a pool sized more generously than the underlying facts actually require.

The second channel opens in the contractual architecture. Unquantified option promises enter the representations and warranties package as a bespoke provision under the share capital and liabilities heading, ordinarily backed by a dedicated indemnity and, in some transactions, by a ring-fenced escrow tranche. Once confirmation of historical grants by board resolution and collection of employee waiver letters are added to the conditions precedent, a measurable delay appears on the timetable. On the accounting side, share-based payment expense that has never been recognized forces a restatement of adjusted operating earnings and pulls down the base to which the multiple is applied. On the tax side, the failure to establish share value independently at the date of grant is separately priced, in United States structures, as withholding and penalty exposure.

The third channel operates more quietly and is frequently the most expensive. For as long as the eligibility decision remains within the founder's discretion, the plan reads as a mechanism binding the team to the founder rather than to the company, and that reading constitutes one of the more concrete pieces of evidence supporting the founder-dependency finding that triggers a valuation discount on the human capital side. The absence of a measurement layer reinforces the impression: where pool utilization, the overhang that expresses total dilutive pressure, the distribution between vested and unvested options, the forfeiture rate realized on departures, and the elapsed time between grant and acceptance go untracked, the only evidence that the plan works is management's assertion. For the investor side, assertion does not substitute for a verifiable indicator. A structure built to incentivize can, in review, become the exhibit for a deficit in management maturity.

The intervention that neutralizes this tendency operates through architecture built around the plan rather than through personal discipline, and it separates into four components. The first is that eligibility be written as a rule rather than as a list: which role band, past which tenure threshold, upon clearing which performance gate, qualifies for a grant is defined in advance, and exceptions sit inside the rule as a documented exception procedure rather than alongside it. The second is a single instrument set: no text other than the standard grant letter, the standard vesting schedule and the standard leaver matrix enters circulation. The third is the approval chain; no grant is communicated before it is tied to a resolution of the board or a delegated committee, and the grant date is the date of the resolution, not the date of the conversation. The fourth is a reconciliation rhythm, under which the option ledger, payroll and capitalization table are squared against one another on a fixed calendar, and no cycle is treated as closed until the variance report is cleared.

BEIREK's intervention in this area begins by reducing the existing promise inventory to a single record; every undertaking scattered across correspondence, offer letters and promotion documents is drawn into one schedule with its amount, date, condition and degree of legal enforceability, since an unquantified liability becomes negotiable only once its boundary has been drawn. The eligibility rule is then rewritten against the role bands and seniority structure the company actually uses, and calibrated so that it can explain historical grants retrospectively, because a rule that cannot account for the past simply generates a new inconsistency item in diligence. What these two steps produce is not a text that displaces the plan document, but the governance layer that makes the plan document operable in practice.

The second stage establishes ownership and rhythm. Authority over the eligibility decision is moved out of the founder's hands into a narrow committee combining finance, human resources and board representation; the committee's meeting frequency, decision threshold and exception authority are put in writing, and the rationale for each decision is recorded at the point of proposal rather than at the point of approval. An accompanying indicator set is kept deliberately spare and bound to the reporting cycle: pool utilization, overhang, vesting distribution, forfeiture rate and grant-to-acceptance interval. Where such a structure exists, the reviewing party asking who determines eligibility receives a procedure as the answer rather than a name, and that difference is precisely what the continuity dimension measures.

Independence from the founder is a measurable capacity here, not a symbolic requirement. A plan passes the continuity test when grants can be made in a quarter during which the founder is absent, under the same rule, on the same documents and through the same approval chain; once that condition holds, the ESOP begins to represent the company's institutional grip on its talent rather than the team's attachment to an individual. The same structure produces a difference in secondary transactions as well, since where a buyer deals directly with vested option holders the accuracy of the ledger can become the single variable governing transaction speed. The indicator of scalability is not the size of the plan but whether a mechanism that works across a hundred employees continues to work unchanged across three hundred.

What determines a company's valuation on the human capital side is, more often than not, the quality of the team but the demonstrability of the mechanism binding that quality to the company. ESOP eligibility is the most legible surface on which that demonstration takes place: where the rule is written, the approval chain functions, the records reconcile and the indicator set is monitored, the plan ceases to be a cost line and becomes a verifiable retention capacity. Where the rule is unwritten, the same plan continues to distribute the same shares, yet is priced by the counterparty as a liability that cannot be measured. The difference lies not in the number of shares but in whether the question of who decides can be answered at all.