Executives now face a transformed advisory landscape, with private equity challenging incumbents and accelerating the competition for AI-enabled services. Recent deals such as Cinven’s stake in Grant Thornton Germany are more than standard modernization plays.1 They signal an escalation between traditional partnership networks, public platform providers, and a new class of private capital-backed firms, all competing to become the next generation of AI partners.
Since TowerBrook’s 2021 investment in EisnerAmper, private equity has steadily expanded into professional services.2 By mid-2025, nearly half of the top 25 U.S. accounting and advisory firms had completed or were actively pursuing PE transactions.3 The prize is significant: estimates of the global management consulting market vary by definition but largely exceed $300B for 2025.4 Meanwhile, the global managed services market is projected to more than double, from about $335B in 2024 to ~$731B by 2030.5
The traditional model of buying advice is increasingly obsolete. In an AI-first environment, value comes not from blueprints but from reuse and run: maintaining healthy models, retraining pipelines, and embedding automation into operations. After costly generative AI experiments that stalled in pilot purgatory, boards are no longer buying advice; they're buying production results, measured through business-specific KPIs like process automation rates, decision accuracy improvements, and operational cost reductions.6
This shift explains why all three competing models are racing to bundle advice + platform + operate capabilities:
Understanding how each model approaches this bundling (and their inherent trade-offs) has become critical for decision-makers allocating AI budgets.
Unlike traditional technology implementations that could largely succeed with technical expertise, AI fundamentally requires cross-functional collaboration to deliver value. This collaboration must navigate complex technical challenges, including model versioning, data lineage tracking, and production monitoring at scale. McKinsey research emphasizes that “to succeed at scale, organizations must shift to a cross-functional delivery model, anchored in durable transformation squads composed of business domain experts, process designers, AI and MLOps engineers, IT architects, software engineers, and data engineers”.7
This deep integration of domain knowledge with technical skill is what makes AI implementation unique. This collaboration must navigate complex technical challenges, including model versioning, data lineage tracking, and production monitoring at scale. The strategy is not to replace experts, but to enhance their capabilities, turning AI into a powerful tool. It is the domain specialists who are best equipped to guide this process, as they alone can identify the high-value problems and opportunities where AI can be most effectively applied.8
The complexity of this requirement manifests differently across the three models:
What makes AI particularly challenging is that technical teams alone cannot identify where AI will create value. Domain experts must audit data for gaps, ensure diverse and representative samples, and guide data scientists on how to structure models and interpret results.8 The process requires implementing comprehensive data lineage, privacy controls, and regulatory compliance frameworks. This is especially critical in regulated industries like healthcare and finance, where biased or misapplied AI can have catastrophic consequences.
Financial services firm JPMorgan Chase exemplifies this approach: Their AI-powered fraud detection systems succeeded because “risk analysts, data scientists, and compliance experts pooled their expertise”, achieving a 15-20% reduction in fraudulent activity.12 The key wasn't just the AI technology, it was the seamless integration of domain and technical expertise.
The Model: Publicly traded tech-services platforms like Accenture bring balance-sheet scale, global delivery, and industrialized tools. Their model depends on landing transformation programs and then industrializing them for long-term run. That strategy shows up in the numbers: In the third quarter of fiscal 2025, Accenture reported $19.70 billion in new bookings, with Managed Services accounting for 54% of this total and generative AI contributing $1.5 billion, highlighting how both AI and large-scale operational contracts are becoming integral parts of their business.13
The Trade-off: This model's strength - predictable performance - is also its core constraint. The relentless pressure for favorable quarterly earnings inevitably leads to a reduced risk appetite. This focus favors the optimization of existing platforms over the pursuit of unproven and potentially disruptive technological breakthroughs. As a result, innovation becomes more incremental than revolutionary.14
➜ Best fit if: You want “advise-build-run” with SLA-backed operations and can work within standardized rails.
The Model: Federated partnerships excel at stewardship and client intimacy through deep industry expertise and multi-disciplinary teams. Some have built sizeable managed services portfolios that seek to provide clients with end-to-end support, from advice and implementation to ongoing operations.15, 16
The Trade-off: Partnership economics and governance create friction in sustaining multi-year platform investments. As The Economist observed, the Big Four and other large networks have “potentially outgrown decentralized governance just as technology choices became existential”.17 Annual partner draws and consensus-driven decision-making can slow platform development.
➜ Best fit if: You need cross-functional advice and complex change management with high-touch customization.
The Model: Private equity's play transcends capital injection. It's a fundamental operating model redesign creating firms that are:
Recent deals involving Blackstone's investment in Citrin Cooperman, New Mountain's stake in Grant Thornton, and now Cinven's acquisition of Grant Thornton Germany show serious capital backing this hypothesis.18, 19
The Trade-off: Sponsor investment horizons (typically 3-7 years), integration complexity during roll-ups, and delivery network flux during transformation create execution risk.
➜ Best fit if: You want a platform-plus-operate partner that can move quickly and price against production KPIs.
A critical risk spanning all three models is the expertise shortage. While many firms can sell “AI,” far fewer can deliver production systems. Research indicates that a significant majority of enterprise GenAI initiatives fail to achieve meaningful business impact. An MIT study found that 95% of enterprise GenAI pilots fail to deliver measurable P&L results, often due to workflow misalignment and weak technical execution.20 Gartner likewise projects that over 40% of agentic AI projects will be scrapped by 2027 due to rising costs and unclear business value.21
This talent challenge manifests differently across models:
For buyers, this means diligence must go beyond roadmaps and pricing. Demand to see architect-of-record and MLOps leader credentials tied to shipped systems, not rebranded slideware. Evaluate their model lifecycle management practices, including containerized deployment capabilities, automated testing pipelines, and real-time performance monitoring systems.
The cross-domain requirement fundamentally changes how organizations should evaluate AI partners:
Perhaps most critically, this cross-domain requirement extends to the buyer's organization. While AI adoption is now widespread, achieving significant business value remains a key challenge. According to McKinsey's 2025 “State of AI” report, the single attribute with the biggest effect on an organization's bottom line is the redesign of workflows to embed AI. However, only 21% of organizations report that they have actually redesigned business processes as part of their AI deployment.24 The difference between success and failure often comes down to this internal, cross-functional commitment to transformation, not just technology adoption.
Organizations must be prepared to:
No consultant, platform, or fund-backed firm can substitute for the internal work of redesigning workflows to leverage automation, establishing governance for AI deployment and risk, and building a culture that embraces continuous adaptation. This requires establishing robust AI ethics committees, risk management protocols, and cross-functional governance structures that can navigate the complex regulatory and operational challenges of enterprise AI deployment.
To navigate these trade-offs effectively, buyers need an evaluation framework that goes beyond surface-level slides and sales pitches:
For decades, companies defaulted to the same trusted brands. AI is different. There are no defaults. And with PE entering the field in meaningful ways, you not only have more options, it also drives competitive innovation which means better results for you. Reassess the field, nobody is “expert-level” at scale yet. Regardless of who leads, vet whether the operating model aligns with your objectives, not legacy rankings.
Private equity isn't a silver bullet. Partnerships aren't obsolete. Public platforms aren't invincible. Each model carries trade-offs that must be evaluated against your specific objectives and readiness.
Think of evaluation less like hiring a consultant and more like vetting a merger partner. M&A diligence asks: how durable are the assets, how aligned are the incentives, and what happens after the deal closes? Ask the same of your AI partner. You need expertise built for what’s needed now, not what worked before. Choose wisely.
Do you need support finding the right AI partner?
Eminence Growth Solutions provides independent AI transformation advisory services including: how to evaluate technology partners, align incentives, and design models that turn AI investments into enterprise value.
References