Private credit at an inflection point

Reflections from the Private Market Forum, May 2026

I attended the Private Market Forum in May’26 as someone who works in data consulting, not as a market investor or a GP. I am a program leader at Logic Pursuits, a consultancy that specializes in data strategy and implementation, and much of our work sits at the intersection of financial services and data infrastructure. I came to the forum expecting to hear about market trends and emerging technology. What I found instead was a room full of senior practitioners describing, in considerable detail, the exact same data problems I had spent the past year trying to help a client solve.

Private credit has grown to approximately $2.1 trillion in global assets under management, according to the IMF, roughly doubling over the past five years. That growth is well documented. What is less visible from the outside is that the operational infrastructure supporting this growth has not kept pace. The industry is running an increasingly large and complex asset class on a data foundation that was not built for this scale. The forum made clear that this is not a peripheral concern. It is the central challenge facing private credit right now.

The Standards Gap

Every panel at the forum returned to the same underlying issue. Private markets have no agreed data standards. No common reporting formats. No shared definitions for how key metrics should be calculated or how data should flow between GPs, fund administrators, LPs, and the platforms serving end investors. DTCC noted that Fund Serve processes around 83% of mutual fund transactions. That kind of efficiency is possible in public markets because the industry spent decades agreeing on how things work. Private markets have no equivalent foundation.

Courtney Scharff from MSCI is building private market indexes and taxonomy precisely because nothing like them exists today. In public markets, benchmarks are so embedded that we rarely think about them. In private credit, we are still working out how to categorize assets consistently, let alone benchmark performance across managers. The valuation panel, with voices from Blackstone, Pantheon, and 73 Strings, described sound valuation practice as resting on five elements: people, process, policy, governance, and data. Their point was unambiguous. Data sits at the base. Without reliable, consistently defined data, every layer above it is weaker than it appears.

What struck me listening to all of this was not that it was new. It was that it was familiar. The challenges being described at an industry level play out in exactly the same way inside individual firms, and anyone who has worked close to fund reporting will recognize them immediately.

The Project That Brought This into Focus

Through Logic Pursuits, I led a reporting automation engagement at a financial asset management firm whose investment reports had been built and maintained in Excel across multiple funds. The business case was straightforward: the existing process was manual, time-consuming, and carried real risk with each reporting cycle. Analysts were spending days producing and validating reports that, with the right infrastructure, could be generated reliably and efficiently from source data.

As program leader, the role was to bring the right people together and keep the work moving. That means coordinating the consulting team alongside the client’s fund accountants, finance leads, and portfolio teams. It also means engaging the Data Governance function early, something that proved to be the most consequential part of the engagement and, in my experience, the part most often underestimated at the outset.

When we began mapping the existing Excel reports across funds, what we found was consistent with what the forum described: no single source of truth, and no agreed standard for how key metrics were defined. The same performance measure was calculated differently depending on which fund you looked at. Some differences were in how income items were classified. Others were in the timing of when data was recognized, or which system’s figures took precedence when two sources disagreed. Each discrepancy had a history. A decision had been made at some point, embedded in a formula, and left there. Nobody had gone back to ask whether it was consistent with how the same decision had been made elsewhere.

Automation is the easy part. The hard part is getting an organization to agree on what its data means before anything reliable can be built on top of it.

Data Governance as the Critical Path

This is where any automation project of this kind has to shift from a technology exercise into a governance one. The Data Governance team needs to be at the center of it, working with data owners across the business to establish a single authoritative definition for every key attribute. But to do that work well, you first have to understand why the inconsistencies exist in the first place.

The reality is that most reporting databases and data structures in financial services have not been designed with a long-term view. They have been built in layers, over years, driven by immediate business needs and delivery deadlines. A new fund launches, and a report needs to be ready by quarter end, so someone builds it quickly, making sensible decisions under time pressure, without necessarily checking how those decisions align with what was built for the fund before it. A system is replaced, and the migration is done under pressure, with data mapped as closely as possible but not always identically. A team changes, and the person who understood why a particular calculation was done a certain way moves on, leaving behind a formula that everyone uses but nobody fully questions. These are not failures of process. They are the natural consequence of organizations that have been growing fast and prioritizing delivery over standardization. The result, accumulated over time, is a data estate full of definitions that were never wrong exactly, but were never agreed either. Each one made sense in its own context. None of them were designed

to sit alongside each other and produce a consistent answer. That is what the Data Governance process has to untangle, and it cannot be rushed. Engaging data owners directly, understanding the history behind each definition, and building consensus around a single standard requires the kind of patient, domain-led conversation that no technology can shortcut.

What that process produces, when done properly, is a governed data dictionary: a single reference point where each attribute has a clear, agreed definition, a confirmed owner, and a documented rationale for why it was standardized in that way. Once that exists, the automation can be built on solid ground. Every calculation is tied back to a definition that has been reviewed and approved by the people who understand what it represents. Anyone looking at the output, whether an analyst, a fund accountant, or a senior stakeholder, can trace the number back to its source and trust that it has been produced consistently across every fund. That is what transforms reporting from something people validate nervously about each cycle into something the organization can actually rely on.

That traceability is not a technical feature. It is a governance outcome. And it is exactly what the forum’s speakers were describing as missing from private credit at an industry level.

AI Needs This Foundation Too

There was significant discussion at the forum about AI and its potential in private credit, from operational efficiency to fund manager evaluation and investor matching. The enthusiasm is understandable. But every speaker who raised AI landed on the same point: it is only as good as the data underneath it. They described the right architecture as data flowing into a clean, governed platform before AI is applied to the output. The sequence is not optional.

This resonated directly with our project experience. The long-term case for what we were building was not just about automating the current reports. It was about creating a data foundation on which more sophisticated analytics could eventually run. You cannot shortcut that by deploying AI tools on top of ungoverned data. You will get faster answers that are no more trustworthy than what you started with. The foundation has to come first.

The Broader Lesson

The Private Market Forum made clear that the standards problem in private credit is not going to be solved by any single firm or any single technology. It requires the kind of industry-wide collaboration that took public markets decades to achieve. DTCC, MSCI, and others are working on pieces of it. But the momentum has to come from GPs, fund administrators, and LPs all recognizing that agreed standards serve everyone, and that the cost of not having them is being paid every day in manual effort, reporting risk, and delayed insight.

What this kind of engagement demonstrates, at a smaller scale than the industry-wide problem, is that the work is achievable when approached with the right structure. A clear project mandate, strong Data

Governance involvement, direct engagement with data owners, and the discipline to agree on definitions before building anything are what make the difference. Technology follows the governance, not the other way around.

For any firm in private credit still running material reporting processes in Excel, the question worth asking is not what automation tool to buy. It is whether the organization has agreed on what its data means. That agreement, documented and governed, is the asset that everything else depends on. Building it takes time and requires the right people in the room. But it is the work that lasts.

Note on figures

The $2.1 trillion AUM figure is from the IMF Global Financial Stability Report, Chapter 2: \u201CThe Rise and Risks of Private Credit\u201D, April 2024. www.imf.org/en/Publications/GFSR/Issues/2024/04/16/global-financial-stability-report-april-2024. The 83% mutual fund transaction processing figure was cited by DTCC at the Private Market Forum, May 2026. All other observations are based on the author\u2019s notes from the forum and direct experience as a program leader leading a reporting automation engagement through Logic Pursuits.

Sannish Kamat

Author

Follow us on