Proof of Work: AI Value Creation
00:00:00 · Stuart Willson
[00:00:00] S0: Welcome everyone. Most of the AI conversation in private equity has been about deal sourcing, diligence, and port co value creation. But there's another conversation and transformation happening inside portfolio teams. It's about information, specifically the gap between when something starts to go wrong inside a portfolio company and when the GP actually finds out. AI is starting to close that gap in ways that weren't possible two years ago. And that's what we're going to be talking about today. I'm joined today by two people who are actively building this infrastructure, one on the platform side and one on the custom build side. Let me introduce them. Yann is the founder, co-founder, and CEO of 73 Strings, a portfolio valuation and monitoring platform built specifically for alternative asset managers. 73 Strings is doing something distinctive. They're not just modernizing the reporting and valuation layer, they're rebuilding the intelligence layer on top of it with AI agents that handle ingestion, insight generation, and more. Joe's the co-founder of Soal Labs, where he builds custom AI infrastructure for private equity firms and private credit firms, NED resolution, governed pipelines, the unglamorous architecture that determines whether your monitoring output is accurate or just confident. Before we get into it, let me also introduce myself. I'm Stuart Willson. I run Pluris. We help private equity firms and many leading organizations evaluate and implement AI through a network of world-class applied AI practitioners. I spent a lot of time in conversations with GPs about where AI is actually creating an edge. And portfolio monitoring keeps coming up in nearly every conversation I've had this year. I'm excited to dig into this with Yann and Joe. So let's start. Monthly reviews are valuable, but the process is brutal. Collection, reconciliation, variance analysis, board decks, and then you start over. By the time anything shows up in the numbers, the problem's already 60 days old. Yann, when you hear that, the monthly sprint, the 60-day lag, what's actually the core problem? Is it data? Is it process? Some combination of the three. What are you seeing?
00:02:31 · Yann Magnan
[00:02:31] S1: Thank you, uh Stuart, and uh very happy to uh to be here with you uh today and and and with uh Joe as well. So short answer to your question is that I think the problem is all three, uh, to be quite frank. But it starts with uh the data problem and uh maybe I should say a data model problem. One of the uh key issues in the alternative investment fund industry is that as you as we all know, portfolio companies like to report in different formats at different cadence, uh different metrics. You'll have a CFO that uh will uh report revenue as revenue, another one that will report revenue as a net sales. And that creates a lot of uh of inconsistency. So starting point data is a problem, data model is a problem. Then uh the uh process is uh is a problem because the issue with the with the what we're seeing happening as we speak is that there's definitely an acceleration in terms of um how GPs and how their clients, LPEs, want to uh be able to look into into the um want to monitor the their assets. And and the uh only way that you can do that is to make sure that you accelerate the cadence that I was referring to uh previously. Um so the process to get there, the process to accelerate that cadence and to make sure that you get all the data in the right spot, under the right format, structured in the right way, is also a big problem. Of course, that's where uh technology can help, that's where platforms like us can help, and that's where firms like Soal Labs and Joe, they can help as well by helping with uh AI, you know, put in place the right uh the right processes. And and then the consequence of all of that is the bandwidth problem. What I mean by that is that if you don't solve the data problem, the data mobile problem in the first place, if you don't have a process that is correctly managed and potentially and hopefully with technology, then you have a big bandwidth issue because we are all seeing that the industry is growing really, really fast. There's more and more AUM that is flowing uh towards private capital uh companies, which means that they are investing more and more. And that's very relevant, actually. There's definitely a need for the uh economies around the world to be uh better supported and more supported by private capital. But that means that there's more and more coming.
00:04:36 · Yann Magnan, Stuart Willson
Um and if there's more and more coming, if you're not so glad the uh the starting point data model and uh data processes with technology, then you end up with a big problem that you have to throw more and more people into the problem. Yeah, that would be it. [00:04:50] S0: Uh Joe, do you want to build on that?
00:04:52 · Joe Zein
[00:04:52] S2: Yeah, I mean, what Yann is mentioning is is very, I think, accurate and on point, right? It's a problem of data first, but also a big problem of process. I think the key thing, and and yeah, I think Yann put it perfectly, is that for different companies, if you think about the process and what's the life cycle of the reporting, whether it's monthly, quarterly, the portfolio companies are usually sending their financials and the three statements and whatever data pack differ agreed on in the credit agreement to the GP, depending on the size of the GP and how they're organized. You know, most of them have portfolio teams who are, you know, receiving those, sending them to the appropriate deal teams, the appropriate deal team who has agreed to be looking at the financials in a certain way, go and input them in this in the portfolio monitoring system. Some portfolio monitoring systems are not that good, so they don't want to interact with their UIs, so they put them in Excel, and then the problem to upload them, they're manually cross-checking, opening the two tabs next to one another, putting the numbers, trying to remember what the mapping was. Like did we map those guys are splitting sales revenue and uh SaaS revenue. So, how do we, you know, do we add them? You know, someone leaves, you don't know the rules, that they're not documented, you forget a zero, you add a zero, you you forget a conversion, and then the data makes it into the system. Then the portfolio team is trying to put the report together and sees a chart that looks very weird with a big dip or a big increase. And then, you know, it's that not never-ending slippery slope and loop that they, you know, the portfolio team is going back to the deal team and is looking at the document and saying something is off, and then it just turns out that you know, for that specific company, there was some specific edge case, or someone forgot a zero, or someone added a zero, or someone mislabeled one of the labels. And so the process is essentially going from receiving the data, ingesting it, categorizing it, putting it in the system, running checks, and then putting the reports together, the aggregate reports together. Today, this is all very manual. This is all very tracked, uh, this is all tracked in spreadsheets where people are checking tick boxes and sending those spreadsheets.
00:07:03 · Joe Zein, Stuart Willson
And in all different stages of this pipeline, there's automation that can be that can be happening. Strings is doing a lot in terms of like the ingesting and the categorization, and then there's also some not even AI, but there's also some orchestration layer where something as simple as a task system, you know, could be drastically beneficial uh and and improve the process by a lot. [00:07:25] S0: In a in a best case, how long does that take today? And in a worst case, how long does that take? [00:07:33] S2: I think it's a function of how many portfolio companies you have, right? And how big your team is. As Yann said, the natural reaction and the natural solution has been throw more people in that portfolio monitoring team, so throw more orchestrators who will who will manage that process. But usually they try to, you know, the the those teams try to set up SLAs because they have to be reporting by a certain date, right? And I've seen processes, quarterly processes take longer, take you know, 45 days for it to be completed on a portfolio of 70 companies. So you can only imagine uh how that scales with bigger orgs. Yeah.
00:07:33 · Joe Zein, Stuart Willson
[00:07:33] S2: I think it's a function of how many portfolio companies you have, right? And how big your team is. As Yann said, the natural reaction and the natural solution has been throw more people in that portfolio monitoring team, so throw more orchestrators who will who will manage that process. But usually they try to, you know, the the those teams try to set up SLAs because they have to be reporting by a certain date, right? And I've seen processes, quarterly processes take longer, take you know, 45 days for it to be completed on a portfolio of 70 companies. So you can only imagine uh how that scales with bigger orgs. Yeah. [00:08:09] S0: And is the uh is the problem the cadence? Uh and I'm gonna guess the answer to this, or is it the right cadence with the wrong infrastructure? I mean, we when you I guess when you think about you know what this would look like in an optimal world, I guess it would be real-time reporting. But tell me a little bit about like what your thoughts are on both the cadence and the infrastructure today, and you know, where do you see it going over the next a few years given the technological developments that you both are deploying into these companies?