Ep. 233: The Arch Effect: Tech-Driven Transparency for Alternative Investments with Ryan Eisenman

Come on in and sit back and relax. You’re listening to Episode 233 of the WealthTech Today podcast. I’m your host, Craig Iskowitz, founder of Ezra Group Consulting. This podcast features interviews, news and analysis on the trends and best practices, all about wealth management technology.
 
My guest for this episode was Ryan Eisenman, CEO and co founder of Arch.co. Arch.co is a platform for managing alternative investments. It was launched by co founders Ryan Eisenman, Joel Stein and Jason Trigg. And I spoke with Ryan about a couple of things I thought were very interesting about the platform. Number one was pain points that financial advisors have with investing in alternatives. We spoke about their schedule of investment, PE look through tool, which I thought was very, very useful for firms that invest in PE funds, their AI powered portfolio analysis tool and their capital workflow engine. All great ideas and great technology that are being brought to advisors by Arch.co.
 
But before we get started, let’s talk about tech stacks. At Ezra Group, we’ve seen tech stacks of hundreds of RIAs and let me tell you, most of them are loaded down with tech debt. So you shouldn’t feel too bad about yours. But let’s face it tech debt is like a giant anchor, holding back your business growth. If you want to free your firm for exponential growth, you should run, not walk to our website EzraGroup.com and fill out the Contact Us form. Our experienced team can evaluate your current tech ecosystem, deliver targeted recommendations, optimize your existing systems and operations or run an RFP and help you implement new software to take your firm to the next level. You can take advantage of our free consultation offer by going to EzraGroup.com.

Topics Mentioned

  • Leveraging Technology to Streamline Private Investment Management
  • Enhancing Decision-Making with Data Insights
  • Innovations in Document Management and Workflow Automation
  • Addressing Challenges and Ensuring Data Accuracy in Private Investment Management

Episode Transcript

Craig: All right. I’m excited to introduce our next guest. It’s Ryan Eisenman, co-founder and CEO of Arch. Ryan, welcome, man.

Ryan: Thank you. It’s great to be here. Craig, thanks for having me.

Craig: I’m glad we could squeeze you in. I know you’ve got a very busy schedule there, as do I. Where are you calling in from?

Ryan: I’m calling in from Flatiron in Manhattan, New York City.

Craig: We’re so close. I’m just across the way in New Jersey here. I’m glad that you could get into the program here in this dreary weather we’re having in the Northeast. But on the podcast, everything is happy and sunny all the time. So could you please give us a 30-second elevator pitch for Arch?

Ryan: Yes. Arch is a digital platform for managing private and alternative investments. We help our 225 clients—who are family offices, investment advisors, RIAs, private banks, and institutions—collect all their information across all their different private investments, visualize it and standardize it on a single platform, and then produce insights off of that information so that they can make stronger decisions.

Craig: Alternative investments are such a big trend now. We’re seeing so many platforms and marketplaces come out. Let’s talk about some of the interesting technology you guys have built that, I think, seems to differentiate you from some of the other platforms. One of the things you have is a schedule of investment look-through tool. Why is that important and how did you build it?

Ryan: Arch is built on the idea that today most asset managers—the Blackstones, the KKRs, and the Sequoias of the world—all report in PDFs. You get unstructured data that is coming to you in a document that you need to read. There’s a lot of information in these documents that is helpful for investors to understand.

Ryan: A lot of these funds send their schedules of investment, saying: We have invested in Stripe, SpaceX, Rippling, and other types of private companies. When you own a fund investment, you don’t just own the fund; you own all the underlying investments within the fund. Then there’s this element: If you invest in multiple funds, you may overlap. You may, without knowing it, have a very large position in something like SpaceX, and that may inform future decisions you want to make. There could be a correlation between the funds you want to be aware of.

Ryan: As you think about managing a large, complex portfolio, understanding the component pieces can shape future decisions and help you make decisions on how to manage your existing portfolio. We’ve been investing in this tool for the last year or so, taking it more publicly to market soon. That allows investors to automatically know—if a fund reports its schedule of investments, which most do—here’s how many individual positions you own across your investments.

Leveraging Technology to Streamline Private Investment Management

Craig: And what is the underlying technology behind that?

Ryan: There are a lot of overlapping things. The most basic bits: We built our software on AWS, so it’s scalable and secure. Then we essentially go out to hundreds of different portals on a daily basis and pull down all the documents that they post there. We pull information that’s sent to us via email. When all those documents and emails go through our system, they’re categorized and read. Data gets extracted off of those documents.

Ryan: That’s where we can start to differentiate, because once we have information in our system, we can send it downstream to your reporting and accounting system. We can display that information in ways that make it pragmatic and useful to you. We can send you a daily email summary of everything we’ve collected on your behalf. And we can summarize information and give you insights that are otherwise hard to obtain.

Craig: Can you give me an example of an insight that you can obtain through this?

Ryan: Yes. One of the recent innovations is probably our most popular product release over the last three to six months. Managers every quarter will send some kind of quarterly report or investor update. Like DRLP: “In the fourth quarter, we invested in this. Our fund is up 10%. Here are some of the other key macro-micro trends that are happening in our portfolio and in the world.” If you’re an LP in a lot of funds, you might be getting multiple of those documents all at the same time during the Q4 or Q1 reporting period. Instead of having to go to the portal and get the document or instead of Arch delivering the document to you seamlessly, we now use AI to summarize those documents. We’ll give you in your daily Arch Digest newsletter something like: “Here’s a paragraph that summarizes each of those investor letters and each of those quarterly reports.”

Ryan: Now I, as the investor or as the advisor, can very easily stay on top of my funds, get the key information as it’s being reported in real-time, and then easily be able to dive in and read more or dig into the whole letter if I want to, but at least be baseline knowledgeable on what’s happening.

Ryan: This is compared to the prior paradigm, where we learned from a lot of clients we were talking to that they often wouldn’t even read these letters just because they had to go to some third-party portal to get them. We’re looking to continually solve the workflow issues that keep people from receiving them and understanding their data.

Enhancing Decision-Making with Data Insights

Craig: I imagine there’s some kind of natural language processing in the back there as the rules engine is parsing through all these letters and pulling out the critical details. How long did it take to build that?

Ryan: I don’t have an exact timeline for you, but we essentially built our own prompt engineering on top of some of the large language models that are available today. Other companies have created these complex technologies that are simple in application and in the ability to use them. And a lot of the thesis around Arch is that we’re going to do the hard work for you of getting all your information, structuring it, and displaying it in a way that’s useful. Then we can leverage an LLM on top of your information in order to give you insights. And that’s a paradigm that we can continue to build on top of—leveraging simple tools on top of your data to make it understandable and useful.

Craig: Which LLM did you use or did you use multiple?

Ryan: We used a few different ones.

Craig: And it’s federated?

Ryan: That question may be beyond my pay grade, so I don’t have an answer to that one.

Craig: You have a layer on top that takes the best of those different language models and combines them.

Ryan: I believe so. But that I can get more information on.

Craig: No worries. You say you built your own prompt engineering. For people who maybe don’t know, what is that?

Ryan: If you were just to take a document and put it into an open AI or one of the other popular LLMs, you would get some result out of it. But if you, on top of these LLMs, dictate and create a prompt like, “Here’s exactly what we want you to pull out, and here’s what we want you to look for,” then you can identify key details or key things within those letters that are important.

Ryan: There’s a lot of specificity you want to drive towards, like: “I want to know the buys and sells in the portfolio. I want to know how much the portfolio is up. I want to know what the commentary is on the macro or what’s happening in the market at large. And I want to know if there are any personnel changes at a fund.” Those might be some of the top things that I care about.

Ryan: We’re about to come out with a second version of this, an updated version of our AI for investor letters. This is going to call out specific sections in even more detail and allow our clients to customize: Do I want the shortest version or do I want the longer version? And there’s a lot more that we can continue to do as we build out that feature set.

Craig: That sounds excellent. I’d like to see how that works at a later date. We’ll have to schedule that—a demo—because that does sound interesting.

Ryan: We can get you set up. We’ll get you set up on Arch.

Craig: Excellent. With the schedule of investment PE look-through tool, your portfolio analysis tool, all this rolls up into your platform.

Ryan: Yes.

Craig: What is the platform built on and how do clients access it?

Innovations in Document Management and Workflow Automation

Ryan: They access it mostly through a web app today. You can do it on your phone or your computer. We put multi-factor authentication in. But essentially, you go to your Chrome web browser or any web browser and access the platform. That’s the main place that you access. We send you the email, which is probably our most popular way of interacting with this data, just because it is convenient. It’s in your inbox—everything you need to know and nothing more on a daily basis.

Ryan: A big part of the Arch story is that this isn’t just for a single user. You have different user roles. You can say: “I’m an advisor. I need access, but my client needs access to a subset of the data.” And the client CPA needs access to all the tax documents—the K-1s and 1099s”—that Arch collects for them. That’s a big part of our user access: Being able to customize the permissions based on the user role and what people need to see.

Craig: That sounds important for some of our enterprise clients.

Ryan: Yes. That’s why we started to do more work with Arch Banks, private banks, and enterprise clients that have potentially thousands of different users that would need access to subsets of this information.

Craig: Indeed. One other thing you described that you built is a capital call workflow engine. Can you talk about what workflow engine you built it on—a third-party engine or did you build it yourself? And what does this engine do exactly?

Ryan: Yes. This is something that we’re building pretty heavily in this space. We continually hear from clients that capital calls and distribution workflows are one of the most painful parts of what you’re doing today. We have a very basic workflow tool right now for completing capital calls. We’re working on a much more comprehensive version 2 of this. If capital calls are painful to you today, please send us a note. We would love to learn your perspective on what’s painful and what you’re looking for in a solution.

Ryan: We find that a lot of people today are running their capital call workflows across their email inboxes; maybe they have a shared Outlook inbox. They might be exporting information to a sales force running different workflow steps. Then they might be uploading information into a performance reporting system or a GL. They’re working with different custodians and different banks. They need authorization letters.

Ryan: There are a lot of these discrete steps that can be centralized and managed much more effectively and efficiently. That’s what we’re building—tools for the advisor and tools for the client—and also ways to make it more comprehensive and secure because it is a big concern throughout the industry. “Will I get fraudulent capital call instructions or send capital calls the wrong way?” We want to be part of the solution to making payments within the private fund space and the advisor space much more efficient and secure.

Craig: Yes. It seems that way, from what I’ve heard from clients who are involved. There’s a lot of paperwork and they’re usually building multiple Excel spreadsheets.

Ryan: Yes, exactly. Paperwork is core to how we think about our impact on the market. We want to make it so that you don’t need a human in the loop to manage an alternative investment. Today, you need humans all over the place. You need a human to confirm the wired instructions, to work with the client, to make sure cash is available, and to set up the wire. There are all these different places where we could think about more efficient systems and more automation.

Craig: Do you help the advisors help their clients come up with the funds to meet the capital call?

Ryan: We don’t today. It’ll probably be part of this future build—being able to pull in funding sources to understand: If I’m funding from this bank account or this brokerage account, do I have cash—yes/no—or does the advisor need to go raise cash? But it’s not in the current state.

Craig: Maybe I should back up a second. Can you just describe what a capital call is for anybody who doesn’t know?

Ryan: Yes, of course. When you commit to any kind of private equity fund—we include venture in that definition—you have an upfront commitment. It might be $100,000 or it could be $100 million if you’re a large pension fund. And you don’t fund it on day one. They might call that $100,000 over three years. They’ll do it in increments as they make investments. Part of this is that they don’t need all the capital on day one because they need to go out and find the right investments to make. When they find the investments, they’ll say: “DRLP, we’re calling 10% of your capital. Can you please fund it in ten days?” Ten days is the typical window for funding a capital call. You get that PDF. It’s probably posted to your portal. You pull it down. Then you need to go to your bank or your advisor and do a wire fund. There are a lot of different steps.

Ryan: There’s also this other part around making sure you have cash available for these capital calls. Cash flow forecasting is a big topic that we’re talking to a lot of our clients about. That’s another area where, if this is interesting to you, please let us know to talk to you about cash flow forecasting.

Ryan: But you’ll have, over the life of a fund, possibly 12 capital calls. It could be three or four a year. Some are more, some are less. Some will use capital calls or lines of credit where they’ll take out a loan against the LP interest in order to fund capital calls. It helps with returns. It also helps with not having to call capital all the time. There are some ways that funds deal with this to make it easier for LPs, but typically, you have quite a few of these capital calls.

Ryan: Talking to a large family office this morning, they said they have capital calls almost every day and almost every week across their 150 or so funds. These are quite frequent occurrences for family offices and RIAs.

Craig: Two terms we should also explain to listeners are the difference between a GP and an LP.

Ryan: The general partner is the person who runs the fund. You can think about it as the fund manager. It is the person at Sequoia who is making investments, raising capital, and orchestrating the fund’s activities. The LP is the limited partner. It’s the other person or institution that supplies capital to the fund. Once they make that commitment to invest, they typically have no interaction with or control over what investments the GP would make. That’s why it’s limited. Typically, LPs pay some kind of management fee (maybe it’s 2%) and some kind of carry (maybe 20% of the profits) to the GP for all their hard work.

Craig: It is important to reward hard work. Moving on to other areas of your platform. As you mentioned, there are lots of PDF files flying around and lots of documents in these types of private investments. You built your own document management functionality. Why did you do that? What did you build it on? And how is it working?

Ryan: Documents have been integral to us since day one. The first thing we built as a company was a solution for collecting all of your K-1s, storing them in a central document repository, and making them available simultaneously to the investor, the advisor, and their accountant. Then we started aggregating every private fund document from all the portals with the idea that most LPs and their advisors do not want to go to dozens or hundreds of different portals so we can give you one platform to find all your documents. Then we just continued to add to that data set and the document set of the documents we pulled, so now it’s all your private investment documents. We’re now also starting to store 1099s, pulling 1099s from banks and brokerages.

Ryan: What we hear frequently from our investment advisor clients in particular and also from our family clients is that they have multiple systems that they and their clients need to log into and they ideally want to get that into one system. They want one login for all their documents. There are also some private equity funds that, in and of themselves, have multiple portals you need to log into to get documents from that fund. At a minimum, we can move it so that all your private investment documents are in one place. But we can also move towards a paradigm where all your documents in general can be stored in a single environment that is flexible, searchable, and easy to understand so that you’re not scrambling across the internet trying to find something like, “Where’s my K-1?” or “Where’s my subscription document?” We can have it all in a very logical database lab model.

Addressing Challenges and Ensuring Data Accuracy in Private Investment Management

Craig: How does this connect to your workflow engine?

Ryan: Certain documents, when they’re received, automatically kick off workflows. When we get a capital call, then the user gets updated: “You have a capital call for $10,000 due in 10 days.” They get reminders on it. There’s a task management workflow page within the platform so you can track all the tasks in real time. It’s all neatly integrated together. Even to the point that as soon as I say, “Great, I paid that capital call, it’s done,” we’ll take that pending transaction, confirm the transaction, reconcile the data, update the value in Arch, and then push that new value to our downstream reporting and general ledger partners so that when cash moves to fund a capital call or fund an investment, the investment value also changes at the same time. It’s helpful for advisors so they don’t have this spiky loss of AUM when there’s money in motion. It makes the information more accurate for both them and for their clients.

Craig: That sounds awesome because I’ve seen that problem firsthand where the spikes hit. It’s like, “What is this?” And no one seems to remember why that happened.

Ryan: Yes. And it’s just like, “This was the three days when money left our Bank of America account and went to Blackstone. But then we didn’t get the statement for 45 days,” so it’s just missing. There’s that problem of missing money quite a bit.

Craig: And in part of your workflow engine, do you have a dashboard that says you’ve got 150 funds, and based on what we know about them, we’re expecting documents from them every quarter, here’s your list? This fund didn’t send you a form—it would flag that.

Ryan: Exactly. That’s a huge part of what we do. We do that for all the quarterly reporting documents, your financial statements, account statements, and investor letters. You can see what you’ve received, when you received them, when you expect to receive them, and what’s missing. We centralize that. The first thing we built on that side, though, is for K-1s because K-1 chasing season now goes from January to October. It’s a [inaudible] year that people are looking for their K-1s.

Craig: K-1 chasing season.

Ryan: Yes. We’ll make T-shirts. And we put it all on the dashboard so you can see it. If I’m expecting 70 K-1s, I’ve received 62 of them, and here are the 8 that are missing. Here’s what the fund manager told us—we can use the collective intelligence across the entire platform—when I’m expecting to get those K-1s. And then our tool automatically reaches out for missing K-1s to make that part of the process much easier.

Craig: Do you reach out via email?

Ryan: We do.

Craig: That’s excellent. That’s also super important because no one seems to know how many… I’m surprised your family offices know how many funds they have.

Ryan: For some of our clients, yes. Sometimes this is the first time people see their total committed capital, total contributed capital, unfunded commitments, and total value across all their investments. We can essentially aggregate up that view and show them: You have 100 fund investments and here’s how it looks.

Ryan: One of our clients runs a private equity firm and he was like: “I’m a great fiduciary at my day job, and I realized by being on Arch that I’ve made too many fund commitments and I’m a bad fiduciary for myself personally.” That’s helped him shape some of his behavior in terms of how he invests over time.

Craig: Have you seen anything crazy when you’ve come into contact with a new client, then you start organizing all their fund data and all the information, and they go, “We didn’t even know we had” X or “We have too much exposure to this”?

Ryan: Yes. There’s a lot of data we can help people understand that they don’t understand otherwise. One interesting one going back to the look-through tool recently is that we were showing a client their look-through for the first time, and she was saying: “Here are all the underlying portfolio companies.” They had one instance where they invested in a fund that invested in a company and then they directly invested in a company as well. Only upon seeing the look-through tool did they see that the fund had marked that particular portfolio company to zero. They were holding it at a half-a-million-dollar position. And reinvested. That gave them an insight on their portfolio that they otherwise would not have known. That’s the type of data that we want to help our clients illuminate.

Craig: Oh yes. That could be disastrous if you had that kind of mix-up between different funds and different investments in the same fund in different ways. You’re overexposed.

Ryan: Yes, exactly.

Craig: Moving on to another area: Private investment value for individuals. Is this the same as the AI portfolio analysis tool where you’re reading the investor letters or is this reading different statements for individual investors?

Ryan: I’m sorry, which one in particular?

Craig: The private investment value for individuals.

Ryan: What you may be referring to is that we have this value visual graph now.

Craig: Right. Yes, the chart you showed me. The visualization. That comes from the other data.

Ryan: Yes, exactly. Because we’re collecting all your documents and digitizing them, we can then allow you to see information in a visually interesting way. We can, for example, plot every increase or decrease in value for a particular investment relative to the statement that it came from. If you’re looking at your investments and you’re like, “Wow, there’s a huge spike in value” or “a huge decrease in value,” on that graph, you can click on the bubble. It’ll take you to the document. You can see exactly what increased and why and be able to easily dive into the information there. Essentially, it’s with the goal of putting data at your fingertips, so that it’s easy to understand and easy to get back to the source document to make that investigation process feasible and doable for our clients.

Craig: Across your platform, you’ve got the comprehensive archive—all your investment documents, all the history, all the investment updates, all the account statements, tax documents, capital call notices, etc.—all stored. All that rolls up since you’re scanning all that in, using natural language processing, and parsing it all out. You’ve got an aggregated portal where I can see everything in one place. Everything gets updated in one place and you alert them in case something isn’t getting updated that should be updated. What about finding errors? You’ve got what could be thousands of documents, but some of them might be wrong. I’ve seen issues where a report comes from a GP and there’s an error. We don’t realize it until it’s already down the line. Do you have any way to check that?

Ryan: Yes. We run hundreds of data checks on the information we receive to make sure that we’re processing it correctly and that the upstream information is correct. We can oftentimes spot things like: “This doesn’t make sense. The total contributed capital dropped.” Why is that important? If I am continually paying into an investment, the next statement should always show that I either have the same contribution or higher if I made another contribution. If it drops, that tells us that some accounting change or something isn’t being calculated correctly, so then we’ll investigate those errors.

Ryan: Another one that we see that’s important, especially for advisors, is making sure that all of your clients get the right tax documents and don’t get someone else’s tax document. It’s a huge issue for folks. We have gotten the wrong tax documents before for some very secretive people that we probably should not have gotten tax documents for. One thing we built for this upcoming tax season is EIN verification. What does that mean? It means that even after we receive a tax document before it goes through our system and gets released to the end client, it has to pass a validation check where the last four digits have to match the last four digits in our system to be able to be sent to the end client. That makes sure that every client only gets their tax documents and if there are any of those upstream errors, the buck stops with Arch.

Craig: That’s what I want to hear. If I’m a family office or ultra-high-net-worth RIA client, I want to hear that the buck stops with you. I don’t want to have to go hunting through different portals and systems and check with different vendors to get that information. I want it all in one place.

Ryan: Yes. That’s what we’re doing.

Craig: There’s one thing that I did find a little confusing. There are multiple firms with the word ‘arch’ out there. Can you tell people where they can find information about your firm, Arch?

Ryan: Yes. We are Arch.co. You can find us on Google, LinkedIn, or mildly on Twitter. We could arguably be a little bit more active there. If you have any questions on this note, you can contact us on our website. You can schedule time directly with our team. We’re excited to continue to solve problems in this space. If there are things that are adjacent to what we described today, that you’re struggling with as a firm, we want to be the company that ships the most features and the most functionality into the private fund space to make it as easy as possible to manage and understand your alternative investments. We want to be doing this for the next 25 years, if not more. Please drop us a line if we can ever be helpful to you.

Craig: That was one of the best summaries at the end of a podcast I’ve ever heard. Ryan, thanks so much for being here.

Ryan: Of course. Craig, thank you. It was good to chat with you.

SEARCH

ABOUT ME

The Wealth Tech Today blog is published by Craig Iskowitz, founder and CEO of Ezra Group, a boutique consulting firm that caters to banks, broker-dealers, RIA’s, asset managers and the leading vendors in the surrounding #fintech space. He can be reached at craig@ezragroupllc.com

SUBSCRIBE TO OUR NEWSLETTER VIA EMAIL

@CRAIGISKOWITZ

ARCHIVES

Archives