In an investment review, one of the questions put to the commercial team almost invariably stalls at the same point: of the incremental revenue earned from existing customers last year, how much originated in a conversation the company initiated, and how much in an order the customer placed of its own accord. The answer typically opens with a pause, proceeds to the names of two or three large accounts, and drifts toward the strength of the relationships. In the same room, account managers genuinely know where the expansion opportunities sit within their portfolios; asked where that knowledge is recorded, what emerges is not a field in the CRM but a personal notebook, an email thread, or recollection. The capacity to generate incremental revenue exists — but it rests with three or four people at the table rather than with the company.
The quieter half of this picture appears in the accounts. The expansion line from existing customers rarely looks weak; in certain years it contributes more than new-logo acquisition does. The reviewing party sees the figure and accepts it, yet the figure itself is never the answer to the question being asked. What is being asked is whether the same figure remains producible across the next three years, once the founder is no longer making customer visits and two key account managers have departed.
The mechanism beneath this gap is not an oversight but, under particular conditions, an entirely rational choice. In an early- or mid-scale company the customer count sits below the volume that the memory of a few key individuals can carry; who uses which product, when a given contract renews, when a particular customer's budget window opens — all of this is already known. Under such conditions, formally defining, recording, and measuring an upsell process generates a clear cost while returning no new information, because the information is already in the room. The shortcut works not because it is a shortcut but because it fits the condition. The difficulty begins when the same method persists after the product of customer count and product count has exceeded the capacity of memory, a threshold crossed silently and without generating any indicator at the moment of crossing.
The behaviour that emerges once that threshold has been passed is remarkably uniform. Expansion conversations arise not from a defined trigger but from incidental contact: a customer calling about a problem, a renewal date that imposes itself, or a competing proposal landing on the table. What this describes is a sales organisation that does not generate expansion revenue but processes the expansion demand brought to it. The distinction is not a matter of terminology; in the first case the timing and magnitude of incremental revenue sit within the company's control, in the second within the customer's. Controllability carries a higher-than-average weight in an investor's growth plan, since a plan can only be underwritten on variables the management team is in a position to move.
Contrary to expectation, the institutional cost first surfaces not in the revenue line but in the distribution of that revenue. Mapping product penetration account by account typically yields the following picture: a small portion of the customer base consumes most of the range, a broad middle band has remained on a single product, and the fact that this band was never opened reflects not a market ceiling but a contact cadence that was never established. In diligence, that distribution says two things — first, that an unpriced expansion reserve sits within the existing base, and second, that the company has yet to demonstrate the capacity to open it on its own. The buyer writes the unopened reserve into its own investment thesis and does not pay the seller for it.
The second channel runs through measurement, and it is the harsher of the two. Where net revenue retention, cohort-level revenue development per customer, and trigger-to-proposal and proposal-to-close conversion rates are not produced on a regular basis, the expansion component of the growth plan presented by management is treated as unverifiable. The fate of an unverifiable component is seldom rejection of the plan; more often it is a rewriting of it. The buyer trims the expansion contribution in its own model to a conservative level, raises the assumed cost of new customer acquisition, and recovers the difference either through the multiple or through the payment structure. It is at this point that an earn-out arrives on the table as a threshold linking part of the consideration to incremental revenue from the installed base — which, for the seller, amounts to financing its own narrative at its own risk.
The ownership dimension binds these two channels together. In most companies upsell hangs suspended between two functions: the sales team regards it as the account management group's remit, while account management, seeing itself as accountable for satisfaction and renewal, refers the pricing conversation back to sales. The territory in between appears on no one's scorecard, is therefore not measured, being unmeasured is not managed, and being unmanaged falls onto the founder's personal calendar. Identifying this at the diligence table is straightforward: commission plans are requested, and the question becomes whether expansion revenue appears in any role's targets and at what weighting. A responsibility with no counterpart on a scorecard is, whatever the organisation chart states, effectively unowned.
What is sought on the documentation side is not a procedure manual but the trace of decisions. Where the record shows how an expansion opportunity was identified, at what threshold it converted into a proposal, under whose authority the price was approved, and why a declined opportunity was declined, the company can demonstrate that it operates a method. Absent that record, past performance can be narrated but not defended — and performance that cannot be defended, even when supported by the tables presented in the data room, translates into a narrower scope of representations and warranties and a higher escrow percentage. The contractual reflection is equally visible: where expansion pricing, scope variation, and add-on module fees were never structured into the master agreement, every incremental sale becomes a negotiation begun from zero, which is a structural friction that lengthens the sales cycle and erodes margin.
The intervention that neutralises this tendency is not the imposition of larger quotas on the sales team but the conversion of expansion revenue from an individual skill into a system output. BEIREK's work in this area begins with a penetration map of the existing base, separating, for each account, which product or service lines have been taken up, which have never been touched, and whether that gap arises from a technical constraint or from a conversation that simply never took place. A trigger set is then defined — a usage threshold crossed, a budget window opening a set interval ahead of the contract anniversary, a new facility, line, or geography coming into operation on the customer side, support tickets clustering around a particular module — with each trigger assigned to a role and to a response window. Until the trigger is defined no upsell process becomes measurable, since a conversion rate without a denominator cannot be computed.
On that foundation a three-component operating layer is built. The first is a decision record kept at the moment the opportunity arises rather than at the moment of approval, holding in a single entry which trigger was observed, who took ownership, and what the outcome was, so that declined opportunities also generate data. The second is an account review cadence fixed to the calendar — quarterly for the upper band of the portfolio, semiannual for the middle band — in a meeting whose agenda is scope rather than satisfaction. The third is the entry of the measurement set into management reporting, with net revenue retention, products per account, and the trigger-proposal-close funnel tracked under a heading separate from new-logo acquisition metrics. Once these three components are in place, founder dependency is resolved at the level of the record rather than the level of assertion, because what is shown at the diligence table is not a narrative but a revenue series arising from meetings the founder did not attend.
Continuity can be tested cleanly at precisely this point. Where an account manager departs with a portfolio and the successor is able, within the first thirty days, to read the open expansion opportunities from the record and to initiate the same conversations off the same triggers, the capacity resides with the company. Where the successor begins by getting to know the customers again, the capacity left the building with the individual, and the discontinuity in the revenue series becomes visible within two quarters. Rotation here is not a risk but a test; in a well-constructed structure it leaves no trace on the revenue chart, while in an unconstructed one it is the most reliable diagnostic instrument available.
A company's ability to draw incremental revenue from an existing customer is a commercial skill; the company's knowing who will do so, at what moment, and off which trigger is an institutional capacity, and only the second is priced at the valuation table. The question that separates them is not how much incremental revenue the installed base produced last year, but what share of that revenue arose from a conversation the company itself initiated — and who, today, knows that ratio.