Author
TL;DR A private equity sponsor sees pricing across a software portfolio as a stack of reported figures: average selling price, average discount, net revenue retention. Each was produced by a company that set its pricing alone, on its own value metric, off its own pricebook, so identical figures can describe opposite economics. A shared pricing function standardizes how decisions are made and leaves what each holding sells untouched. The readable view sits one layer down: the value metric each company sells, what its license grant covers, whether its editions were designed or accumulated, and where realized prices sit against list. Three questions at the end put that layer on the table.
Portfolio pricing strategy in private equity software is the sponsor’s attempt to read pricing across holdings that each set it alone. The figures the fund receives are visible as an aggregate and unreadable as a comparison. Each was produced on a different value metric, against a different competitive set, off a pricebook built for a different market.
Why that stack cannot be read as one picture, what sits underneath it, and what a view across the portfolio can establish one layer lower than the reported number is the subject here. The AI risk discount is read on one company at a transaction. The portfolio read below runs across holdings during the hold.
Where a Sponsor Sees Pricing
A portfolio company settles its pricing decisions once, or updates them episodically, and reports the results upward in a board deck. What reaches the sponsor is the output of a pricing decision.
What a board deck carries, and what it leaves behind
The board deck carries realized figures: average selling price for the period, win rate, discount rate relative to list. These are accurate records of what closed. They are silent on the question that determines whether those figures are comparable to anything else in the fund.
The patterns in our library show value capture tracking how critical the application is to the customer’s operations. A commodity application and a critical-path application can report identical average selling prices and have opposite economics. The board deck never surfaces the structural difference.
Pricing moves EBITDA without adding customers. Every advisory page in this market says so, and it is true. What those pages leave out is which of the licensing, packaging, and pricing decisions the assumed uplift rests on. Each costs something different to change, and the board deck never distinguishes between them.
For what a structured read of those three decisions inside a single holding covers, see private equity pricing diligence. The problem here is a different one: why a fund’s holdings cannot be read side by side.
Why portfolio companies have no reason to coordinate on pricing
Each portfolio company set its pricing architecture years before the current fund held it. The decisions were made by a team responding to its own competitive set, its own buyer vocabulary, and its own packaging at the time. Nothing in the reporting chain asked that team to reconcile its choices with any other holding in the fund.
Portfolio companies vary widely in pricing maturity, and the advisory market uses that observation to justify a shared capability. The observation is real. Maturity variance is visible in the reporting, though, and unit variance is invisible, and unit variance is what makes the comparison fail.
One pattern from our library shows what sits under a figure that reads consistently. A client priced by territory, where the customer’s own regional map set the count and the license fee followed it. A new chief revenue officer on the customer side consolidated the map into fewer regions and asked for the fee to fall in proportion. Usage had not fallen; the population running the software was unchanged, and only the label counting it had moved. On a board deck that account read as every other, a unit count and a fee; underneath, the metric was one the customer could relabel at will.
Why Two Portfolio Companies Reporting the Same Discount Are Not Comparable
This is the mechanism. A reported figure can be identical across two holdings and describe entirely different economics.
The same discount figure off three different pricebooks
Take one average discount reported by three portfolio companies in the same fund. The first discounts off a per-seat list price. The second discounts off a committed-volume edition whose floor already includes a volume concession. The third quotes from an edition whose list price was constructed to absorb a standard discount before the negotiation begins.
The reported discount is the same figure in each case. Pricebook deviation, the gap between the undiscounted list price and the price the customer pays, measures something different in each. A comparison across pricebooks built on different units is a comparison of three different quantities that happen to share a label.
Pricebook deviation is the more precise lens, and it still depends on the pricebooks being comparable before the comparison means anything. Unifying pricing across product lines covers the same problem inside a single company with several products; the portfolio problem is that problem one level up.
What average selling price means when the unit changes
Average selling price collapses under the same pressure. Two holdings can report the same figure while selling entirely different units. One sells seats. One sells API calls. One bills against a data-volume threshold the customer’s operations may or may not reach in a given period.
The figure reported as average selling price is the average of what closed. It says nothing about the value metric each company sells, meaning the unit the licensing model selects as the basis for price. When the value metric differs, the figures are incomparable, and averaging them at the fund level produces a number that describes no individual holding.
Why does portfolio pricing strategy break down at the reported number?
Because the reported number is the output of a pricing architecture. Two holdings can produce the same output from architectures that are structurally incomparable, and a fund-level view of portfolio pricing strategy that reads only the outputs cannot tell them apart.
What a Pricing Center of Excellence Answers, and What It Leaves
The standard advisory answer to this problem is organizational. Stand up a shared pricing capability across the portfolio, give each holding access to common people, process, and tools, and standardize how pricing decisions get made.
What standardizing pricing process across private equity software portfolios delivers
A pricing center of excellence standardizes the inputs to a pricing decision: the frameworks teams use, the data they collect, the approval chains that govern discounting, the review cadence that keeps decisions current. A fund that runs consistent pricing process across its holdings produces more defensible decisions at each one.
It also lowers the cost of pricing governance. Each holding no longer has to rediscover how to run a competitive analysis or structure a packaging review, because those capabilities travel through the shared function.
Each holding priced against its own competitive set, and a shared function gives each one better tools for that work. Competitive analysis for B2B software covers what the analysis requires. The shared capability makes the analysis more consistent; it leaves each company’s competitive set exactly where it was.
Why common process does not produce comparable numbers
The shared function standardizes how decisions get made. What is being decided stays the same. The first company still sells seats, the second still sells committed volume, and the third still has editions that accumulated rather than being designed.
Process consistency is a property of the decision-making. Comparability is a property of the unit being sold. A fund can run excellent process in every holding and still receive a stack of numbers it cannot lay side by side.
Governance asks who decides. Comparability asks whether two decisions can be read against each other. The advisory approaches in this market run those two questions together, which is why the org chart answer feels sufficient while the measurement problem stays open. Which axis carries the real upside inside a holding is a separate argument, and the licensing axis question in PE pricing diligence makes it.
A Pricing Center of Excellence Staffs the Function. Who Fixes Each Holding’s Architecture?
A shared pricing capability settles who owns pricing across the portfolio, not whether each holding’s licensing, packaging, and pricing fits its buyers. Describe one portfolio company’s situation and an expert will pinpoint where its architecture breaks.
What a Portfolio Pricing Strategy Can Establish Below the Reported Number
One layer below the reported number, a view across the portfolio becomes readable. Readable here means coherent as a set of named differences, which is a different property from a single comparable number.
The value metric each company sells
The value metric is the unit the licensing model selects as the basis for price: seats, API calls, data volume, workflows, the count of users in the customer’s own customer base. Each holding has one, and most boards receive a reported figure without ever seeing the unit named.
When a reading across the portfolio names each holding’s value metric, the fund-level view changes. Two holdings that looked alike on average selling price may be selling entirely different units. Two that looked different may share a value metric and differ only on price level, which is the most accessible layer to change. Pricing model versus value metric covers the distinction in depth.
What the license grant covers, holding by holding
The license grant is what the contract entitles. It determines what the customer receives for the price paid, and what requires an additional purchase. Two holdings can report the same average selling price while granting entirely different rights: unlimited users on a single instance in one, a named-user count on a multi-tenant environment with usage caps in the other.
A reading that names the license grant holding by holding surfaces whether the fund’s holdings are selling access, seats, outcomes, or some combination. That surface never appears in the board deck.
What realized prices show that list prices cannot
List prices are statements of intent. Realized prices are records of what the market paid. The gap between them, read across a portfolio, shows where each holding’s pricebook holds under negotiation pressure and where it gives.
When realized prices cluster tightly around list, the pricebook is working, and customers who buy the same configuration are paying the same price, the discipline enterprise SaaS pricing describes. When they scatter, something in the packaging, the edition structure, or the competitive situation is driving outcomes the list price never anticipated.
Across our corpus, the groupings that predict willingness to pay are behavioral rather than firmographic. The size-and-vertical splits a board deck reports do not predict which customers pay more, so the Customer Groups under each holding’s revenue stay hidden behind the split. That pattern is invisible from the decks alone, and reading it is work on a specific fund’s holdings. Talk to an expert about what that reading establishes for yours.
The scatter under a clean average has a shape our library carries. A company with a never-discount policy whose reps still needed a lever to close sold customers a less capable edition and then worked with license management to grant entitlements beyond what that edition included. The policy line held on every report, and the average selling price read as disciplined, while the net realized price for the same capabilities scattered underneath it. The editions had accumulated around what reps needed to close, and the accumulation was invisible from the deck.
Questions a Sponsor Can Put to Any Portfolio Company
These questions work across a portfolio regardless of vertical, and each is answerable in a board meeting. They are the portfolio’s own questions. The single-company read at a transaction, the value metric, the grant, the cost move and the substitute, is the four-question set in the AI risk discount piece.
1. Is our list price the price customers see, or does negotiation start from a different number?
2. Were our editions designed for distinct Customer Groups, or did they accumulate over time?
3. Has the value metric changed in recent years, and if so, did the packaging change with it?
The questions neither score the holding nor rank the portfolio. They name what the sponsor does not know from the board deck alone, and they give each portfolio company’s team a clear surface to respond to. For the operating partner’s side of the same question, the read runs inside one holding rather than across the fund.
A sponsor who can answer all three for every holding has a view across the portfolio the board deck never provides. What those answers reveal for a specific fund is what the reading establishes, and we do that reading with sponsors, holding by holding. Investors can see how that engagement is shaped, and the portfolio view covers what a full read across holdings returns.
