Medallia was once one of the most respected names in customer experience management. A pioneer of enterprise voice-of-the-customer programs, it helped define a category that became a cornerstone of how large organizations listened to and acted on customer feedback. So, when news broke this week that Thoma Bravo is handing Medallia to its creditors — wiping out $5.1 billion in equity from its $6.4 billion take-private in 2021 — it landed as more than a financial headline. It landed as a signal.

The mechanics are straightforward. Medallia’s annual earnings, estimated at around $200 million, could no longer cover debt servicing costs approaching $300 million. When a group of lenders led by Blackstone refused to extend further payment-in-kind relief, the math became inescapable. Thoma Bravo chose to walk away rather than invest further to service the debt and retain ownership. That’s not a vote of confidence from the people who understand the company’s performance better than anyone.

It’s tempting to view this purely as financial engineering gone wrong, and there’s truth in that framing. The 2021 vintage of PE-backed software deals was built on cheap debt, peak multiples, and the assumption that recurring revenue would paper over almost any capital structure. According to McKinsey’s 2026 Global Private Equity Report, buyout funds have now underperformed public equities for three consecutive years. More than half of all buyout-backed companies have been held for over four years — the highest proportion on record — because they cannot be sold at prices that justify what was paid.

But the financial story only goes so far. In my experience, private equity firms buying software companies tend to reach instinctively for cost restructuring as their primary value creation lever. Revenue growth, customer-led growth, the kind of expansion that comes from making a product genuinely indispensable — these tend to receive less attention. The prevailing model is: cut costs, optimize margins, and hope the multiple holds long enough to exit. When that approach was underwritten by near-zero interest rates, it worked often enough. In today’s environment, with rates higher and exit windows narrower, it’s exposing a fundamental gap. PE-backed software companies need a growth story, and cost-cutting alone doesn’t write one.

I’m not claiming that a customer-led growth strategy would have saved Medallia under its existing debt load. That ship may have sailed the moment the deal was structured. But the broader question is worth asking: if existing PE approaches aren’t generating the returns investors expect — and the data increasingly says they aren’t — maybe it’s time the industry reconsidered what it means to create value in a software business.

Medallia’s collapse also needs to be understood in the context of a much larger reckoning in private credit. The same week the Medallia deal was announced, Project Syndicate ran a major feature asking whether a private credit crisis is imminent. The private credit market has surged from around $300 billion in 2010 to as much as $3.5 trillion today, much of it flowing through exactly the kind of leveraged software buyouts that produced Medallia’s capital structure. Software companies account for an estimated $600–750 billion of private credit exposure. The IMF has warned that 40 percent of private credit borrowers now have negative free cash flow. UBS projects default rates could double to 9–10 percent this year. Morgan Stanley has warned they could go even higher.

What makes this genuinely concerning — as Brian Judge of UC Berkeley argues — is the structural opacity. When the same firm originates a loan, holds it in a fund it manages, values it using its own models, and reports that value to an insurer it owns, the result is unlikely to reflect what an independent buyer would pay. The layers of leverage, self-dealing, and offshore regulatory arbitrage bear an uncomfortable resemblance to the structures that preceded 2008. The gains accrue to Wall Street, but since private lending is still tied to government guarantees, the potential losses may once again fall on retirees and state insurance systems. Medallia’s lenders had already marked the debt down to between 69 and 79 cents on the dollar well before the restructuring became official. That’s not a confidence-inspiring signal about the broader portfolio.

The more sanguine view, articulated by the economist Dambisa Moyo, is that private credit is not yet approaching crisis conditions and that its scale remains small relative to the mortgage market before 2008. She may be right about the aggregate numbers. But as former central bank governors Agustín Carstens, Stijn Claessens, and Klaas Knot warned in the same feature, warning signs are multiplying and the window for preventive action is narrowing. For any enterprise software buyer whose vendor is financed by private credit, the relevant question isn’t whether the system as a whole will survive. It’s whether your particular vendor’s particular capital structure will allow it to keep investing in the product you depend on.

Bill Staikos, a former Medallia SVP who has been writing thoughtfully about the CX software market, recently argued that the real story is bigger than one company or one financing deal. He’s right about that. But we should be careful not to understate what a creditor takeover actually means in practice. When Blackstone, KKR, Apollo, and Antares Capital become your owners through a debt-for-equity swap, their mandate is to recover their capital. That means further cost restructuring is virtually guaranteed. And cost restructuring, in a software company, means some combination of fewer people, less investment, and less attention to the things that made customers choose the platform in the first place.

Medallia’s customers — many of them large, sophisticated enterprise buyers — chose the company for its depth, its data infrastructure, its ability to run complex feedback programs at scale. Those capabilities require sustained investment. Credit-fund owners are not product-vision owners. R&D, AI innovation, and roadmap ambition will be subordinated to cash flow recovery. Anyone who has lived through a PE-owned software restructuring knows how this plays out for the people who built the product and the customers who depend on it.

The timing compounds everything. The entire CX software landscape is being reshaped by artificial intelligence, and not in the incremental way that legacy vendors have been positioning it. Wall Street is repricing the traditional software model in real time — software company stocks fell nearly 30 percent between October 2025 and February 2026, and BDCs with heavy SaaS exposure have underperformed their peers. JPMorgan recently downgraded collateral valuations on loans to software companies. As Howard Davies of Sciences Po noted, the big investment banks have started making a market in credit default swaps against funds marketed by Apollo and others — insurance that, until recently, nobody thought was necessary. The fear isn’t irrational: AI is changing what’s possible, who can deliver it, and what buyers should expect to pay.

In the B2B CX world specifically, the disruption is structural. The traditional model — deploy surveys, collect responses, aggregate scores, present dashboards — was built for an era of data scarcity. That era is over. Companies are sitting on vast operational and behavioral datasets that, combined with modern machine learning, can predict customer outcomes like renewal probability and revenue risk without ever sending a survey. In consumer research, digital twins and synthetic data are beginning to replace large-scale polling. The survey-centric model isn’t just dated; it’s dating faster with every advance in AI capability.

The idea that enterprises will simply use AI to rebuild their CX platforms internally is, in my view, a fallacy. The failure rates for internal software builds remain stubbornly high, and CX is a domain where getting the model wrong has real revenue consequences. But new competitors are absolutely entering the market with AI-native architectures that don’t carry the technical debt or the business model assumptions of the incumbent platforms. That competitive pressure is real, and it compounds the challenges facing any legacy vendor — especially one entering a period of ownership focused on cost recovery rather than product investment.

Blackstone’s own leadership drew a useful distinction earlier this year, attributing Medallia’s underperformance to execution-driven issues rather than AI disruption specifically. That’s probably accurate as a proximate cause. But execution problems and structural market shifts aren’t mutually exclusive. They compound each other. A company struggling to execute while its market is being reinvented, owned by creditors whose primary concern is recovering their capital, operating within a private credit ecosystem that is itself under systemic stress — that’s a challenge no amount of financial engineering can solve.

Medallia deserves respect for what it built. For over a decade, it was a genuine leader in helping large organizations take customer feedback seriously. But the chapter that’s closing this week isn’t just about one company’s debt stack. It’s about what happens when an industry’s financial sponsors optimize for extraction overgrowth, when a technology model ages out faster than expected, and when the buyers who ultimately fund the whole ecosystem start asking whether they’re getting what they’re paying for. Those questions don’t have comfortable answers for anyone still running the old playbook.