Test anfragen
✅ Check up last week: how we looked at why share class rights matter.
These rights can introduce non-linear payoff profiles in an exit scenario. IPEV's guidelines suggest that "when estimating the fair value of an Investment, the valuer should determine how each class of equity would participate in distributions from a sale or other liquidity event and the implications for the fair value of each class of equity." The same exit value can yield very different payouts to different share classes depending on where it falls relative to the preference stack.
No single method captures that equally well across every outcome. Some assume a single snapshot in time; others are built for a binary world where the company either succeeds outright or fails completely; others still weigh a fuller range of in-between scenarios. Each has merits and trade-offs, and the right choice depends on the company's stage, the capital structure's complexity, and how close a liquidity event actually is.
CSE, or the fully diluted approach, treats every class of shares as if already converted to common and values them all identically, based purely on fully diluted ownership. It's easy to implement and requires almost no modeling complexity. The catch is that it ignores the value of liquidation preferences entirely.
CSE only holds up as a reasonable methodology in a narrow set of circumstances: when the capital structure is genuinely simple (no classes with meaningfully different economic rights), when the market clearly expects all preferred to convert to common anyway (typically because an IPO exit is expected, or because equity value is high enough that converting is an obvious economic choice for preferred holders), or when outcomes are expected to be "bimodal," meaning the company will either succeed so completely that everyone converts, or fail so completely that there's nothing left to distribute regardless of preference stack. Outside those cases, CSE will systematically overstate what common is worth.
CVM allocates the current company value across classes based on the greater of each class's liquidation preference or its as-converted value, as if the company were being sold today. It's a snapshot, not a forecast: precisely its strength and its weakness. Like CSE, it's straightforward to build, but it isn't forward-looking, and it can understate the option-like value of junior classes if there's a real chance the company appreciates significantly before an eventual exit.
CVM may be appropriate when a liquidity event is genuinely imminent, when the fund holds a controlling position in the company, or otherwise has the ability to force an exit, or when equity value is already well above the liquidation preference stack, such that conversion is the obvious rational choice for preferred holders.
OPM takes a different approach: it treats each class of equity as a call option on the company's total equity value, with strike prices set at the breakpoints where the payoff to each class changes, the liquidation preference thresholds and conversion points. A Black-Scholes-type framework then prices each of those options, and the incremental value between breakpoints gets allocated to whichever class holds that layer of the capital structure.
OPM's advantages lie in being forward-looking, considering future values, and relying on inputs that are largely observable rather than purely judgment-based. Instead of allocating a single snapshot of today's value, it treats the exit as a continuous distribution of outcomes stretching from now to a future liquidity event, using volatility and time to exit as the key inputs. Several of these inputs are market-observable: volatility and the risk-free rate can be benchmarked against a peer group and market, while time to exit is the one deal-specific input, tied to this particular company's expected path.
OPM's limitations lie in relying on volatile inputs, being moderately complex to build, and assuming returns follow a lognormal distribution that won't always match reality. Expected volatility and time to exit are both judgment calls, and small changes to either can move the allocation meaningfully. Mapping the breakpoints across every class and running them through an option-pricing framework is also a heavier lift than a straightforward waterfall calculation. On top of that, OPM considers only a single liquidity event, so it doesn't fully capture the differences between specific potential future events, an IPO versus a sale, for example, at various time horizons.
Given the number of assumptions, practitioners typically anchor them through an OPM backsolve: deriving the company's equity value from an actual, recent transaction in its own stock, most often the last financing round, by holding the other assumptions at best estimate and solving for the one unknown, total equity value, so the model reproduces the price actually paid. That only holds up if the transaction was arm's length; a distressed or related-party round needs adjusting before it's used as the calibration anchor. That equity value is then carried forward and adjusted at each later measurement date.
OPM tends to fit best when the capital structure is complex, or when a specific future liquidity event is hard to forecast.
Two other approaches are worth knowing even if they come up less often in day-to-day practice. One is "scenario-based methods": Simplified Scenario Analysis, Relative Value Scenario Analysis, and Full Scenario Analysis, better known as the Probability-Weighted Expected Return Method (PWERM), the most detailed of the three.
PWERM identifies two or more distinct future scenarios, typically an IPO, a strategic sale, a dissolution, or continued operation until a later exit. Each scenario gets its own exit value (or range), timing, discount rate, expected dilution from future financing rounds or option pool issuances, and probability of occurring. The equity value each class would receive is then worked out under each scenario based on its rights, weighted by that scenario's probability, and discounted back to the present. It reflects how investors actually think about outcomes, as a mix of possible futures rather than a single point estimate.
The cost is subjectivity and effort. Every one of those inputs, exit value, timing, discount rate, dilution, and probability, is a judgment call, and building out several scenarios multiplies how many of them have to be estimated and defended compared with a single-point method. That same flexibility cuts both ways: because the scenarios and their probabilities are chosen by the valuer, it's easy, deliberately or not, to lean the weighting toward whichever outcome supports a preferred answer. As with OPM, calibrating the model to a recent transaction helps keep that initial set of assumptions in check, even if it doesn't remove the judgment involved in choosing the scenarios themselves.
PWERM is well suited to companies with genuinely distinct, identifiable exit paths where probabilities can be estimated with some confidence.
The Hybrid method combines two or more of the methodologies above, weighting their results together rather than relying on just one. The most common pairing is a scenario-based method with OPM, using OPM to model the range of outcomes within a scenario that's too uncertain to pin to a single value. The specific blend, and the weighting between the pieces, depends on the facts and circumstances rather than a fixed formula. By nature, it also blends together the full assumption set of whatever it combines, inheriting the pros as well as the cons of each: more inputs to defend, but also more ways to cross-check the result.
It's a reasonable answer when the facts don't cleanly fit any single method alone.
None of these methods is right by default: the method has to match the reality, how close the company is to a liquidity event, how complex the preference stack is, and how much confidence there is in specific future scenarios versus a general distribution of outcomes. Two practical checks worth applying on top of that: the method should be one another valuer could independently replicate or approximate using the same data and assumptions, and it should be consistent with whatever valuation technique was used to arrive at the total equity value being allocated in the first place. Mixing a post-money equity value with an OPM-style allocation, for instance, may produce nonsensical results.
The rights tell you what each class is owed. The method tells you what it's actually worth. What ties the two together is consistency: the approach has to reflect how market participants would actually view this company today, not just whichever model produces the most sophisticated-looking number. A simpler method that captures how the deal would really play out beats a complex one that doesn't.
If that list of judgment calls sounds like your Friday afternoon, we hear you, and we're building something to help. More on that soon.
There is no universally best method. OPM is often appropriate for complex capital structures with uncertain exit timing, while PWERM is better suited to companies with clearly identifiable future scenarios. CSE and CVM can be appropriate where the preference structure has limited economic impact or a liquidity event is imminent.
OPM models a continuous range of potential future equity values using option-pricing techniques, while PWERM models specific future scenarios such as an IPO, strategic sale or dissolution and assigns probabilities to them.
CVM is most appropriate when a liquidity event is imminent, the investor can effectively force an exit, or the company’s equity value is already sufficiently high that liquidation preferences have little practical effect.
CSE can be appropriate when preferred shares are economically expected to convert into common shares, particularly where an IPO is expected or the company’s value is sufficiently high that conversion is clearly preferable to exercising liquidation preferences.
An OPM backsolve derives the implied total equity value from a recent arm’s-length financing transaction. The transaction price is used as the calibration point, with the model’s other assumptions held at best estimates.
%20Loans%20in%20Private%20Credit.png)
