AI should be a core system

2026 was the year of AI pilots across functions. In 2027, AI becomes a core system, like a CRM or portfolio monitoring tool today. Use cases will be productized across the firm, staffed, and governed. Yet most firms do not have a route from pilot to enterprise-grade tool.

But infrastructure and data must come first

Considerations around buy versus build now lean toward build for edge use cases, as enterprise AI tools gain strength.

Firms want to build a solid infrastructure and deploy multiple use cases, rather than buy point solutions for each use case. Each new use case builds on the last instead of becoming throwaway work.

That infrastructure also requires a consolidated source of truth.

Every use case depends on it, so we see data readiness moving up the agenda. Yet few firms have completed the work, and those that have not are putting it on their 2027 roadmap.

Governance and audit trails will become a priority

Production also raises the standard. A pilot can let AI infer an answer on the fly. Where AI produces figures that touch investments or LPs, firms need deterministic rules they can audit and sign off. In the US, this is an SEC requirement, and fiduciary duty to LPs applies wherever a firm operates. Permissions, security, and approvals need the same standard as any enterprise application.

Holland Mountain’s Private Capital AI experts helps fund managers productionize use cases. Contact us to plan your route to production.

GPs want a partner, not just a processor

Most of our clients outsource fund administration, so the question has changed. GPs now expect a true strategic partner model from their administrator.

Where the relationship falls short, whether on service, SLAs, or fees, firms are looking to fix it.

In 2027, we expect remediation to move up the roadmap.

Decide what to outsource and what to keep

Firms are revisiting what to outsource and what to keep in-house. The answer depends on their technology, operating model, and risk tolerance, which also informs how much oversight they need.

Should you remediate, or switch?

Then comes the harder question: is the administrator still right for the firm? Where it is, firms reset the commercials, the SLAs, and the oversight model. Many are moving from shadowing their administrator to exception-based oversight. Where it is not, switching is easier than it was for firms with access to their validated data.

Getting the outsourced model right delivers ROI at the manager level.

Monitoring has always been about the financials

For years, Private Markets firms have monitored portfolios with a set of financial KPIs from each company, rolled into valuations and fund returns. Firms increasingly treat value creation, board packs, valuation waterfalls, and structure as part of a wider portfolio management function.

Now AI is taking data analysis further

Advances in AI are letting Private Capital firms process more structured and unstructured data, faster than before. Firms can now understand much more about their portfolio than they could. The gains depend on the right technology choices and the processes behind them.

Firms need to know what their portfolio monitoring system can capture, and what needs to live adjacent to it. Without the data in the firm’s systems, AI has nothing to work with.

Managers who combine financials, operating metrics, value creation results, and wider portfolio data can see trends across the whole portfolio and manage it better. That gives them a competitive advantage.

An evolving expectation

We see this becoming an evolving expectation among sophisticated LPs, who invest in a firm’s ability to add value. Portfolio intelligence is direct evidence of that ability. Mid-market and larger buyout firms that cannot show a clear strategy may find it affects fundraising.

Vendors are merging and adding features

There have been many acquisitions among the providers used by Private Capital firms recently. PE Stack tracked 20 M&A transactions and 18 funding rounds in 2026 alone. With many vendors on the market, we expect more consolidation in 2027 and changes to service models. At the same time, vendors are broadening their offer, often with AI. For firms, the question is whether each new capability is the best tool for the job.

Doing more and being the best tool for the job are not always the same thing.

Firms still choose best of breed

Best of breed has won out over all-in-one in recent years, and firms are holding that line. They want the best tool for each job and favor shorter contracts, even if that means more vendors. Broader products can also bring higher license costs for features, such as AI, that a firm already runs elsewhere.

Switching administrator is getting easier

This trend also applies to outsourced fund administration. Switching administrator is easier than it was for firms with an established model for data transfer. They already have access to their validated data, often from inception, meaning that a move no longer depends on the outgoing administrator.

A central data layer makes adding providers routine

With a central data layer, adding a provider is no different from extending the one you have. That holds whether the firm switches administrator, adds Private Credit or merges with another manager. Without it, the vendor that offers the most in one product becomes the default. That can work, but it scales less well and ties the firm in.

This means today’s firms are less stuck on vendors. With their data at the center, they can move to the best tool for the job when it comes out.

Outsourcing of non-core work has more than doubled in last two years

Cost and headcount mandates are not new, but the response is. Private equity firms win on investing, fundraising and value creation, not on running platforms and data. Northern Trust found that managers planning to outsource non-core activities has more than doubled, from 18% in 2024 to 39% in 2026. Two models are emerging from this trend: fully outsourced and hybrid.

Holland Mountain increasingly offers both; Seb Marwood has recently joined to lead this proposition and evolve our capabilities. Contact us to learn how we can help.

Early movers are winning big

Private wealth is now one of private equity’s biggest new sources of capital, and semi-liquid funds are the route in. Early movers are being rewarded: Carlyle’s private client evergreen inflows rose from approximately US$300 million in 2023 to US$3 billion in Q3 2025, and Blackstone’s private wealth fundraising grew 53% to US$43bn in 2025. PitchBook expects evergreen private equity assets to more than double to US$610bn by 2030. With 59% of managers pursuing wealth capital, the window is now.

Wealth needs its own operating model

Capturing that wealth is complex. Semi-liquid funds promise liquidity from illiquid assets, and 2026 redemption pressure showed how quickly that is tested. Getting structure, liquidity, valuation and distribution right needs specialist advice from the outset. Five areas change:

01

Distribution

Most firms reach wealth through private banks, platforms and IFAs, whose fees reduce management fee income and can concentrate redemptions.

02

Liquidity and valuation

Gates, liquid sleeves and credit lines must hold when redemptions exceed inflows, with timely NAVs that treat investors fairly.

03

Fundraising and investor experience

Both need specialist people and a service built for many more investors.

04

Onboarding

A candidate to outsource, so staffing for fast-rising volumes becomes the provider’s problem.

05

Reporting

Stays in-house: with the right technology and data, 10 investors or 10,000 make no difference.

Just because you can serve institutions does not mean you can serve wealth.

This is a growing topic among our client base, and Holland Mountain helps firms set up the operating model for wealth before launch. Contact us to learn more.

More firms are asking about ROI

More GPs are asking about the return on their operations, technology, and data investments. Budget, projects, and people are all finite, and initiatives now compete for them. In the past, firms often did what the loudest voice asked for. Now they need to choose the initiatives that give an actual return.

“Previously, people never even asked the question. Now there is a lot to do, and a finite amount of people and budget to do it.”

Not every AI use case brings returns

Every team has ideas for AI, and many will bring no return at all. Spend is high, yet many initiatives have not moved past the pilot. Firms need to know which use cases work, which do not, and which to keep.

ROI needs to apply across the board

A benefits case is no longer enough. A benefit to one team, often the investment team, may not give the firm a return. Firms now look across the whole operation for the most efficiency or tangible benefit. Many already have ROI, but they do not know how to describe it. We help firms work out where that return comes from and measure it. Contact us to learn more.

An evolving expectation

LPs do not go deep into a GP’s operating model. At a high level, though, firms should be able to show a process for change that weighs ROI and prioritization.

As AI spend grows, we expect this to become an evolving expectation.