Read below for a full transcript of the conversation.
Sean Park - Great, I think we're live now. All right, well. Welcome everyone and thank you for joining us today. My name is Sean Park, I'm a managing director at First Republic Bank. I'm in the Venture Capital Services Group. We're excited to be presenting today our topic, "Fund Portfolio Construction Models for the Real World." The goal of today's webinar is to provide an overview of fund or portfolio allocation models and provide some real-world insights that we hope will be informative and help you think about some of the common issues you'll be facing when constructing a model as well as some general best practices. The Q and A is live. We encourage you to post all your questions throughout the webinar. We'll try to get to them in real time, and if not, we'll reserve some time at the end of the webinar, and we'll try to get to all your questions. All right, now let's get to it. So let me take a minute to introduce the panel. Joining us today Anubhav Srivastava, sorry, Anubhav, he's the founder of Tactyc, Mike Palank, GP of MaC Venture Capital, and Taylor Davidson, founder, and CEO of Foresight. I'd like to take a minute now to have the panel introduce themselves. Anubhav, do you want to go first? You're on mute.
Anubhav Srivastava - Thank you, thank you for having us, Sean. My name is Anubhav, I'm the founder of Tactyc. For those unfamiliar, Tactyc is platform that makes it easy for managers to construct portfolios, manage portfolios, to scenario planning and strategizing their portfolio models. I personally came from an investment background, and then I spent five years at a venture capital fund. So, looking forward to talking about portfolio construction here.
Sean - Great. Mike, Taylor, please introduce yourselves.
Mike Palank - I'm Mike Palank, I'm in Los Angeles. I'm one of the founding GPs of MaC Venture Capital. We were an early investor in Anubhav's company Tactyc. Been on a really amazing portfolio construction journey with him over the last several years. Mac is a seed focused VC firm. We typically lead our co-lead deals at pre-seed or seed across a wide-ranging number of industries.
Sean - Great, thanks. Taylor?
Taylor Davidson - So, I'm Taylor, I run a company called Foresight, building financial projection, and models for startups as well as funds. I'm a former VC and I've worked with hundreds of managers on constructing portfolio construction models and modeling their funds.
Sean - All right, awesome. Well, thanks everyone. Let's jump in with you first Taylor. So, what is a fund or portfolio construction model and why is it important?
Taylor - I think of a portfolio construction model as a quantitative representation of an investment strategy, an investment thesis. It's kind of a measure of an idea in terms of like how you're going to invest the capital you raise from investors, and it's a rationalization of your investment approach. And it kind of comes down to the idea of like, how you're taking an overall set of committed capital and how you would invest that into new and follow-on. And then usually an expectation in terms of like, how that capital gets allocated into like a sample individual investment. So, you'll understand how that kind of gets to deployed over time as well. It's kind of like combination of, you know, overall allocation as well as usually like a capital deployment as well.
Sean - Great, so Mike, you know, having, or being a GP at a venture fund, you know, do you agree with Taylor's description? I'd love to hear your thoughts on why it's important, how you use the models in your fundraising or wherever you use it.
Mike - Sure, I like to say, you wouldn't go out and build a house or a building without a blueprint. It would be sort of asinine to go out and try to do that. And I think in exactly the same way, it's foolish to go out and start to deploy capital out of a fund, any type of fund, without a construction model, which is your blueprint for how you plan to operate and deploy capital. And yeah, as you're building a house, you may want to change dimensions of a bathroom or this or that, and obviously changing the dimensions of a bathroom may mean you have to make your living room a little bit smaller, and in the same way, if you adjust your construction strategy, it does affect other things that, you know, impact or flow into the construction model. But I think, you know, having one at the outset, even before that, we'll talk about this probably in more detail, but my opinion is, you need to figure out as an investor what are your, you know, unique superhuman strength? You know, what can you bring that no one else can and let that sort of dictate your strategy and let your construction model actually dictate how much money you should raise. I think most managers pick a number and sort of craft their construction strategy off that, and we actually do that too, but I think it's really interesting that you even think about fund sizing as a result of a thoughtful construction model that's driven at its core by what you're really good at, and the differentiating features you have as an investor. And I'm excited to talk more about that today, and then also about using that model on an ongoing basis. It's important to construct the blueprint, but then, you know, you don't build a blueprint and throw it away as you start building the house, you go back to it every day and make sure that, you know, you are on target with it, and if you are off wide you need to adjust, and I think that's important too with the model.
Sean - All right, we'll get into that in a little bit, but I always think it's interesting to see, you know, different sort of levels of sophistication, let's say with respect to fund managers, I think the model is really telling, right? Sometimes when you look at a model, you can really tell if a GP or fund manager really knows their stuff because the model will tell that. All right, great. Anubhav, do you want to walk us through, you know, maybe display a model, walk us through the key points, the highlights there as we think about, you know, fund modeling.
Anubhav - Sure. So, I'm going to show a model on our web app, the Tactyc web app. And I just want to preface by saying, the tool is less important. We show a model and we let you do modeling on the web. You could use a spreadsheet that Taylor has made. You could build this yourself in a spreadsheet, and it's the process and the journey that's far more important than what tool you're using here. So, what you're looking at here is, for example, a $150 million fund that we've modeled. And in our case, this model will tell us what our expected follow-on reserves might be, how much investible capital we might have, number of deals we expect to do, and our expected returns. These are all calculated metrics. And this is important to note. A lot of folks sometimes come in saying, "I want to follow-on reserves of 45%." Well, that may not always work with, you know, in the context of the rest of your strategies. So, in this model, we're calculating all these metrics and then there's various charts on this dashboard that give you more detail. How much investible capital do you have? Net of fees and expenses. How is that capital going to be used across the different rounds of investments? This one is going to be doing seed, series A and series B deals. How many deals are we going to be doing? How many deals will we do over time? How many follow-on investments will we do over time? What does our capital call schedule look like? And as Taylor mentioned, what does our capital deployment schedule look like? So, the model will calculate all of this for you to inform the GPs if this is in line with that strategy. Similarly, on the performance section, we actually project out of future fund valuation over time, as your investments are expected to move from unrealized to realize, or what is a power lock for your fund? And this is, again, something that, in our model is an output as opposed to being an input. RVPI, DPI, TVPIs, all of the key metrics and LP might care about, are they forecasted to be over time? And then finally, things like on a direct investment basis, what is your initial check strategy? How much are you reserving per round? What valuations are you assuming you'll be entering at? What round sizes are you going to be assuming you're entering at. Things like dilution per deal, what does a profile look like? Say for a series A investment, as that investment grows up, how will your ownership change? So, all of these things are something a good model, a good construction model should be able to answer if constructed correctly. Finally, there's -
Sean - Oh, sorry Anubhav, carry on.
Anubhav - There's LP returns which is obviously based on your waterfall and then GP returns. And those are basically the headline outputs. So, the first question is, how do we build this? What goes into it? What are drivers and what are calculations? Everything as I said here is a calculation. Let's talk about the drivers to the model. So, in our app, we take all managers through a fund construction wizard, where we ask them questions about the areas that they want to be thinking about. Some of these are things that are coming directly from the LPA. For example, in the general section, a manager comes in and says, "What is the total committed capital? How much GP commitment they might have? Is this a rolling fund or not? So, what is a timing of your commitments and how are they expected to come over time?" There's another section which is sector profiles. And now this is where we ask managers to think about what is the average round sizes and valuations that you expect for the sector to be investing in. And this is not just related to the rounds that you are investing in, but your companies are going to grow up. So, you will need to figure out what valuations they might be achieving three years from now. And what will your ownership look like? What will your fair market value look like? Now, one thing I'll point out is our model uses the concept of graduation rates and exit rates. Very simply put, graduation rate is a percentage of companies that are expected to graduate from one round to the next, and exit rate is a percentage of companies that are expected to be exiting after reaching around. This is what determines the power law curve for your fund. How many companies are going to be exiting and at what valuations they might be exiting at, which is a third field, which is the valuation that these rounds are going to be exiting at. So, we take all of this data, and then we ask you for allocations. So, allocations are where defining the types of deals you're going to be doing. This is seed, series A and series B deals. And for each of these profiles, we ask you to think about how much capital do you want to put into this type of investment? What is the initial check size, and what does your follow-on strategy look like? Out here, for example I've modeled a seed investment with a $1.4 million entry ticket size, and I want to maintain an ownership of 10% until the V round. So, we use all of this data, all of the data that you're putting in terms of check sizes, follow-on strategy, and that's what used to build up to a follow-on reserve that you saw on the dashboard. The last few things I'll point out, and we'll get into these into more detail, you obviously-
Sean - Anubhav, can I ask one question?
Anubhav - Yeah.
Sean - You mentioned constraints and you know; you can't look to get 20% and 50 companies if you're raising $100 million fund and have 50% for follow-on. How does your system so sort of tell you if you violated a constraint, you know, given the strategies that you're inputting?
Anubhav - Exactly, so that's the benefit of using a bottoms-up approach where you're saying for each deal, this is what I expect to do. So, we know exactly the total amount of capital that might be allocated, and we'll never reach beyond that capital. Contrast, that will be the approach where someone says, "I want a 45% reserve." Well, if you don't have enough graduations, you may never end up using that reserve. So, we never end up in a situation where you're having leftover capital, or you don't have enough capital. Last couple of things I'll point out, managers obviously want to think about your management fees and there's different fee basis that you can use, obviously committed capital is the most common approach, but you might have different fee tiers over time, and you want to think about the expenses, expenses that fall out outside of fees. Things like legal expenses, formation costs, software licensing, those are things you might want to think about as well. I won't belabor the point. There's exit recycling. We will talk about that as well. That's something you can turn on as well, as you want to define the waterfall. Are you doing a European waterfall or an American waterfall? So, all of these factors taken together will build up this model. And it's important to think about what constraints are and what are not, what are drivers, and what are not. And sometimes it's best to enter this phase with as open minded a construction process as possible, and let the model tell you what your reserve should be or how many deals you should be doing as opposed to you putting that as an input.
Sean - Great, thanks Anubhav. Well, you can see, I mean, there's a lot of detail here in this model. There's a lot of things here, and this kind of goes back to, I think Mike's point about, you know, you can use these things post fundraise and track, you know, how you're tracking past your strategy, but let me ask you based on that, based on looking at Anubhav's presentation of Tactyc. Mike, you know, what do you think of the tradeoff detail versus simplicity? Do you have any thoughts on that?
Mike - I do, I when we started our first fund, I wasn't incredibly well versed in construction strategy and sort of my thought, and I think a lot of people in the industry think this way as you. Like I said, you sort of pick a fund size of what you think you can raise around number, and then I think the general rule of thought is you pick a follow-on reserve anywhere from, you know, nothing. You know, I think the average is 40 to 60%. And the number is just sort of picked arbitrarily. And to be honest, that's kind of what we did. We just thought, you know, 40% for fund one follow-on, and then what we would say was we were going to put that 40% into our top 25% performing companies. And if you think about it, that really doesn't make sense. And we were sort of just taking this non modeled approach to it. Like, how can you say you're going to put your allocation in your top 25%? You know, what if you're only halfway through allocating into your initial investments? How do you know if that first or second company going back to market is in your top 25 when you haven't even invested in the last half of your companies? Like you don't know. So that's not the way to think about, you know, your follow-on strategy. You have to think more about it at like, you know, graduation rate. You know, we do 50 company portfolios. It's a 50% series C to A graduation rate, that means we'll have 25 companies raising series A rounds.
What percentage of those 25 do we want to do? What percentage can we do? You know, Anubhav's model allows you to input sort of, you know, round dynamics. If the average A round, let's say is, you know, $20 million, and we look to get 10%, our pro rata will be 2 million in each of those series A rounds. And if 25 companies are raising A, that's $50 million if we do all of our series A pro rata follow-on, that's, you know, 50% of $100 million fund, there goes all your follow-on. And so like, it just helps you think about, well, man, if I do 100% of my series A follow-on, and the graduation rates, my portfolio reflect the market rates, that's all my follow-on. So, I probably can't do 100, maybe I can do 50. And you have to think of it more grandly like that. And then there's also social, you know, human dynamics, like you don't really want to tell a founder no as there, you know, being successful, getting a series A term sheet, a lot of our companies have done seed bridge rounds. It's hard to not support a company if they're making progress towards the A. So, I think it's good to think about it in a more granular way, which, you know, to be honest, I think we were less granular in fund one, I think we became granular towards the end of fund one, and we're going into fund two with a very, very thoughtful approach, you know, informed by Tactyc.
Sean - Yeah, I think that makes a lot of sense, and something I'll say is, you know, I might say use a simple model for perhaps your deck, but make sure you have that detailed model and you understand it, you know, sitting somewhere that, you know, behind maybe, you know, in a spreadsheet, or using Tactyc or something. I'll pose this one to you Taylor. We've got some questions coming in right now. So, it says, so the first question from Magnus, can you elaborate on your thoughts on applying system like Tactyc or modeling in a spreadsheet focusing on a first-time fund between 100 and 200 million Euros. Well, let's leave it at that. And then, you know, you're up Taylor, I'd love to hear your thoughts, you know. When do you feel is a good time to use a model, maybe when do you move from a spreadsheet to something like Tactyc or something like you have a much more developed model?
Taylor - Yeah, so the way I answer it is like, you know, I build models for managers, kind of look into model and run funds for years, you know, we always try to model like a lot of the same things that are in Tactyc. I think there is like a path that a person kind of goes through as they evolve their business, right? So, you know, when you're first start looking to like, think about what a fund would look like, I think a spreadsheet works fine. I think when you're in the fundraising process, I think a spreadsheet can work find the case as well. I do think though that like... Oh, I think like with all companies, I give the same advice to like startups as well looking to manage their business is that, you know, as you evolve and need to kind of really handle the company a lot more detail, well, that's when things become harder do in spreadsheets. And I think when... You know, I've been doing this for a number of years and, you know, years ago there wasn't really an option, right? We didn't really have like... You could either use spreadsheets, and you had to adapt and have a person kind of manage everything all the time, or you had had a big, really complicated financial package. It was really expensive. And there wasn't really any option that allowed you to kind of do things like a much, much easier on the web. And I think that as you kind of grow, and actually kind of manage or run a company, even some of your other questions there in terms of like managing a budget or forecast and comparing it to what's actually happening, the more details you get in terms of like knowing actual results, the harder it is to kind of manage that in active basis in excel. And, you know, one of the common things is like, people will say about spreadsheet models in general is like, hey, they build it and they don't come back and use it. Well, the reason they don't come back and use it is it's hard. It's hard to actually kind of do that manage on ongoing basis. And, you know, that's why we're using a way kind of manage that process becomes a lot easier, a lot more variable, but I'm sure Anubhav has a perspective there as well.
Anubhav - Yeah, I'm of the opinion, the tool doesn't matter, especially when it comes to construction, a spreadsheet can just, frankly, sometimes it's easier. It's easier to explain that. The place where something like Tactyc becomes more useful is after construction, which is, your fund has been deployed, you've made 10 investments, how am I doing? I said I was going to do 3 1/2 TVPI, am I tracking to that? If not, what do I need to do going forward to course correct my fund? That's where that feedback loop becomes really important. And that's where, when we built that app, by talking to about 50 to 60 really successful managers on what their data driven workflow looks like, and they all employ this feedback loop. They don't throw away the construction model, whether it's a spreadsheet or whatever, they don't throw that away after the fundraise, they actually keep coming back to it. They update it, they ask it questions, and then they get back answers on what they should be doing going forward. So that's where I think the web app becomes a little bit easier, maybe to work with, but candidly, I don't think there's any difference between a spreadsheet or a web app.
Taylor - Yeah. I wish these were... I used Taylor's model before and some of the other firms, I wish yours was around when I was a CFO Anubhav, because I'll tell you, it takes a bit of discipline. In the old world, what we used to do is, you know, go back to a bunch of disparate spreadsheets and have to update each of them, and then try and include you together. And it was so, you know... It was so, you know, time consuming and convoluted to try and take care of all that. It did take some discipline to want to go in there and do it, and like, you know, make sure everything was up to date, so you could see where you needed to cost correct, or where you're actually at in the life of fund. But anyway, that's great.
Mike - Sorry for being a biased investor, but I think you should use Tactyc very early on. And I'm not just saying. Here's why. I think what being a good venture manager means is first doing self-reflection. And I think Tactyc sort of is like a tool that helps you through that self-reflection process. And, you know, obviously do I want to invest in later stage growth companies? Do I want to invest in baby companies just being formed? Like for me, what's my desire? You know, it's a totally different skillset. You're playing with different companies; you're making different kinds of decisions. We chose early stage because we like to get in the weeds and be hand holders and help with sort of the creation, bring to market process. We also chose to be generalists. That was just a personal choice of going after a number of industries. And that to us meant we needed to have a larger portfolio. We were just doing, you know, biotech, maybe we could zone in and do 15 or 10 company portfolios, but being generalists, we felt like we needed, you know, to do a larger number of companies. And there was also looking at graduation rates, and you know, more swings of the bat, but, you know, going through that, that Tactyc onboarding manager with each kind of question helps you kind of think like, all right, why am I putting this answer in here? Like, you really got to think about like, how many companies do I want to do, you know? Do I want to get all my ownership upfront, or do I want to sort of try to maintain it over time? Like, you know, there's two ways of doing it. Do I want to do SPV so I don't need to have as much, you know, follow-on allocation or any. And so, each of those questions makes you sort of think about, who do I want to be as a manager? What am I good at? And as you answer those, you know, you ultimately get a portfolio model and a blueprint how to act.
Mike - And Sean, if I can just add one thing here.
Sean - Yeah.
Mike - Portfolio construction is actually a difficult model. I mean, Taylor is one of the world classes exports when it comes to doing that in spreadsheet. But in reality, if you're building this from the ground up, and if you're doing it right, it's not something that can be done in a week. It's going to take a lot more time than that. Sometimes it can take months. And the reality is, most managers out there shouldn't be spending, they're not spreadsheet modelers, they shouldn't be spending two months building a model. They have to go raise funds, they have to make investments, and so use the tools that are out there, whether it's Tactyc, whether it's a spreadsheet model, but try not to reinvent the wheel because this is a solved problem at the end day.
Sean - Yeah, now, I agree. For myself, I often see, you know, I talk to new managers perhaps or emerging managers and they'll ask me, "Well, do you have a template, or could you help me put it together?" And I always say, "You know, no, you really sort of need to do it yourself." Because I've seen so many models now, and so many of them have issues and it's like, you know, I don't want you to present that to a potential investor and look silly. So, you need to go through the exercise and these are a lot of things you all have just touched on. But we're getting a lot of questions about sort of, I mean, I'll throw them into the best practice category. You know, how do you calculate advance rates? Stuff like that. But so let me start with you Taylor. As you think about, you know, in your models and you put models together. I mean, let's start with that. I mean, how do you think about advance rates? Do you just leave it as a toggle or do you look at data, or do you allow the manager to base their advance rates on, you know, their historical, you know, what they've done in the past? Like, you know, do they have a 50% advance rate from seed to series A, et cetera? Do you just kind of model it to be flexible or? Let me let you jump in there.
Taylor - So, Mike brought the earlier point and one thing I want to kind of tease out of what he said was that, a model is like a reflection of a strategy, right? Decreating a strategy is often more dictated by the qualitative practicalities of like your own background and how you're going to manage your fund, how you're going to execute it, and that's not driven by like the numbers, but that's driven by like what you can practically do as a manager. Now you can build your fund in the marketplace. And so, the model, therefore, and oftentimes, becomes like a reflection of like what you're doing to make sure, like, does your strategy, can you actually execute the strategy, and does the numbers align with your thinking around it? And I think that was an important point to kind of want to pull out of that. And so, then it comes down to like, okay, well, given how it's reflecting strategy, how does my assumptions and numbers, how does it reflect like, what I think is going to happen in terms of how I'm going to execute the fund, and how does my expectations vary against what I see also happening in the market. And I see there's a number of kind of questions that are also on the Q and A about data and benchmarks from those things. And there are resources, there are publicly available resources you could use to find out some of this data out there, but using all this data is hard and challenging, right? And so oftentimes we have to then talk to other emerging managers and these other groups, or other areas to find people who know numbers as well. And it becomes like a big data sharing thing as well. Like finding quality people to kind of interact with people like First Republic and Sean, okay? You've used you a number of times to ask kind of questions about like what you expect. And so, I think what you have to do is you have to get an idea of the overall kind of marketplace with benchmarks as are applicable for you and your situation.
If you just model like the average fund, the average thing, like it's not going to happen, partially because like the average, there are so many different ways to execute and create a fund, right? You could be creating in Silicon Valley early stage. So, I think a lot of the general knowledge out there is targeted towards a specific area a type of manager is investing into, you know, Silicon Valley, early-stage tech, but that's not how all venture funds are created. There's a bunch of funds all over the world, late stage, growth stage, middle areas, studios, other areas. And so, there's such a wide variation, even in industry sectors as well. There's such a wide variation in strategies, it can become really hard to find good data if that specific thing you're doing. Now, I think a Anubhav is doing that. He can probably speak to some kind of market numbers out there, because he gets to see and interact with a lot of different funds, and kind of see their data around expectations, but it can be a really, really hard thing to find good data. And so oftentimes we have to build a model therefore that is flexible to use for a number of different assumptions and then use it to kind of test and see how it kind of reflects your strategy based upon that. Does that answer the question?
Sean - Yeah, that makes a lot of sense. I mean, you know, you need to be flexible with the model, but I'm going to pose it to Anubhav too, but make it a little bit more specific, like, I mean...
Anubhav - Yeah.
Sean - So if you were, let's say consulting, and anyone can jump in here, but if you were consulting for a manager and they say, "You know, this model's great, but" you know, and some of the questions that we're getting now are asking these same things, you know. What do you look at? I mean, what do you look at for exits? How do I model... You know, how do I model for the return of my, you know, initial investment? How do I think about advance rates? How do I think about reserves and things like this? Is there a source you go to? Do you go to say PitchBook or do you, you know, again, is there somewhere specific or do you just, I give you the model, you figure out how you tweak it?
Anubhav - We help people through that. So let me just show this to you. In our model, we obviously have graduation rates or what you call advanced rates, and you can apply an industry profile. So, you can come and say, "I want to look at the advanced rates for AI or for e-commerce companies." And if you click on that, I'll just automatically update that. Our advanced rates and graduation rates are source from external data. So that's a NVCA, PitchBook, Omni, Crunchbase, that's where we're getting from. Having said that, that is all benchmarks. And what we've frequently seen is most managers override that because they may have, this may be their second fund and their first fund, they saw that their graduation rates were much higher than what the industry averages were. And so, they may make the case look we are actually better pickers. We probably have a better acumen in terms of picking the right deals, and so we think we might do slightly better than the market and that's their prerogative. If they believe in that and they can convince the LP of that, they're welcome to change the graduation rates. But as a conservative assumption, we start everyone with the medians from this external data source. The second thing that you touched upon is, a lot of fund models are, you know, in the mold of 20% of my companies are going to be returning five X, 80% will return zero X and 10% will return 20X. And that's just kind of like this brute force mechanism to project out fund performance. What we do is we try and help the manager think through, okay, if a company does exit at the series A itself, let's say it was an ACRA hire, what is a reasonable exit valuation for it? And so, a default assumption there is, well, let's just track it to the pre-money valuation of the next round. We would assume it's going to track the same valuation profile round over round. But this is, again, another area where the manager could say, "Well, I'm in the pharma sector, and the pharma sector series A evaluations at exit are significantly higher than series B pre-money evaluations." And so that's one way for them to update that information. The bigger point is, if you have market data, use that as a starting point. If you have your own track record data, you can use that to override the market data if your LPs can believe that.
Sean - Great. That's great. I mean, I remember just having to sort of clue to a lot of this together from a number of data sources and it was exactly the same experience for me. I put together the base model and then, you know, my boss, the GP would be like, "Well, no, you know, we are going to do better than that." And you know, you need be able to, you know, factor that into it too. But, you know, that's good stuff. All right, let's talk a little bit about reserves and I'll switch over to you Mike. You know, we sort touched on this. I'd love to hear, you know, your strategy, and strategies maybe you're seeing out there for reserves and, you know, why reserve?
Mike - Well, I'll start with, why reserve? And, you know, I think that the consensus answer is, it's sort of like legalized insider trading in a way, where you invest in these companies early, maybe you're on their board, you see how they're doing, you see customer pipelines, you see where things could be in a year or two, and they go raise a round and you get to go into this round, you know, with this amazing insider information, and you can, you know, invest more money in this company and kind of maintain your ownership stake, you know. In a lot of cases, people get prorata which is the right to invest an amount that allows you to maintain whatever ownership sake you had in the round prior. And it can be one of the great things of venture investing is that you get to invest capital with this sort of insider information, which is a really, you know, powerful thing. And yeah, as companies are doing really well, you know, rounds become competitive, you're already in if you have prorata, you know. In theory, don't have to fight to get allocation. You have the legal right to it. Now, in a lot of cases, a founder may try to push your prorata down to get other new great and investors in, but, you know, it's to maintain this ownership stake that in some cases has been hard fought. And so those are rationales, you know, to do prorata.
Sean - Do you ever get asked by, you know, founders? Do you reserve? Do you reserve for your investments?
Mike - Oh for sure. I think good founders are trained to ask that question because, I think, and we've seen this, you know, in a lot of cases, companies will have to raise, things happen, right, you know? A great plan is great until you get punched in the face. Like most of our companies have had to raise rounds in between their seed round and A round, in some cases, to take advantage of a hiring opportunity, some cases because revenue hasn't materialized as fast or product development hasn't happened as fast as originally anticipated, but things are moving in the right direction, and they just need a little bit of capital to buy them six more months to hit those series A KPIs. And you would hope as a founder that your lead investors will have the, you know, capacity because they have follow-on reserves to invest in those bridge rounds, and also into your A round, I mean, there's a thing called signaling risk where if your seed lead doesn't invest in your A round, you know, the other investors, new investors looking at you will think, well, this investor is inside information and they're not investing. What does he know that I don't know, or she know that I don't know? I'm not going to come in. And so there's a real signaling risk if a seed investor or, you know, prior round lead does not invest in the subsequent round. And if you don't have follow-on reserve, well, then maybe there's a easy answer for that. It's like, I just don't do it. But, you know, I think they want their lead investor in prior around, in most cases, to come back in.
Sean - Yeah, I mean, in a competitive field, I mean, it's got to be, you know, a bad form if you say, "We don't reserve."
Sean - Yeah, yeah because year is going to know that they're back out looking for new investors, you know, in six or something, but I'm curious too, and maybe I'll pose this to you Taylor. Over the last few years we've gone from sort of traditional sort of series A funds, and now there's, you know, pre-seed, seed funds, or a lot of smaller funds, the amount of emerging managers popping up everywhere, you know, is off the charts with respect to reserve strategies in your modeling for some of your clients, you know, what have you seen change? Have you seen, you know, things like, as Mike mentioned, those reserves or, you know? I'd love to hear what you've seen, what you see today, maybe the changes versus what it used to be.
Taylor - Well, I think there's a bit more flexibility around it. Like the way... When people ask me about reserve strategy, I always say, "What's your fund strategy?" Meaning, you know, most times when people raise a fund, their goal is to raise another fund, and so you're not looking to optimize, you know, one particular fund, but optimize, you know, your business in terms of like managing capital to deploy usually over multiple funds over a long period of time. And so there are, I think are more emerging ways for people to go to market through SPVs and other areas that makes it a little more flexible in terms of thinking our reserve strategy, and to be more practical about it, like the way that kind of speaks back to Mike's point is, you can create a fund which says, "Hey, I'm going to allocate 50% of my reserve for follow-ons into my investments." Or you can say, "I'm going to use all my fund to do new investments and I'm going to, you know, offer up my prorata or my potential follow-on opportunities to my LPs to do direct investments." So, we'll do SPVs kind of off of the fund outside of that. We'll all raise a second fund to do only kind of growth opportunity style investments. And you've seen this like over the last, you know, decade or so. You've seen a lot more funds raising, like what they even call opportunity funds around that, which is their, you know, ability to kind of invest kind of later, you know. In later stages, we've seen a lot more of it over the years. So, I think we've seen a lot more flexibility, and easier ways to kind of go to market in terms of raising capital from investors opens up a lot more room for people think creatively around how they think about the reserve strategy. At the end of the day, people are trying to like figure out, hey, what's the best way to build capital, best way to put a market to potential investors. It ultimately comes down to like, how somebody thinks about being an investor as their business in terms of how they want to think about that.
Sean - Yeah.
Mike - But this is where... I don’t know if we're going to get into this, but I mean, sorry to be the hype man for Tactyc, but like, this is probably one of the more powerful things of Tactyc is allowing an investor to sort of compare apples to apples, to apples, where I'm sitting here and I've got the opportunity, you know, to invest in a new company, follow-on into company A, or follow-on into company B, and let's say, I can only choose one. How do I decide what's the most optimal, you know, use of my capital? Do I do the new investment follow-on to A or B? And a lot of cases, people are just, you know, they're guessing, they're going with their gut, which sometimes works, but I love it with Tactyc. You can actually go in and put really kind of like, you know, take the kind of calculus in your brain and really kind of put it on paper, which is like, well, how do I think this, how will this new company perform? Like, what are my probability weighted outcome scenarios? And then follow-on A follow-on B. Maybe they're a bit further along. So maybe like, you know, the new investment, and maybe there's an 80% chance it goes to zero, but follow-on A is a little bit further on. So maybe it's like 50% chance of zero, you know, 50%, it becomes a hundred million and follow-on, B is a little bit further along, and maybe 25% is going to go to zero. Maybe it's not a big of a market. So maybe 75% or 50%, it goes to 100 million, 25% goes to a billion. And when you sort of really like break out your thinking more grandly, it becomes possible to weigh, you know, or to evaluate those choices, apples, to apples, to apples, which is really powerful.
Sean - Yeah. I want to turn it over to you in a second, Anubhav. I think you've got something to show us, but I just wanted to follow on from Mike's point there. This is kind of when you move from, you know, the model to actually managing your fund. And I recall, you know, this would've been great, because I recall, you know, earlier in my career. You know, it wasn't so much at the beginning of the fund's life, but towards the end where we had to think about capital rationing, and we hadn't really thought through, you know, the amount of reserve, the amount of follow-ons that were going to happen for the existing portfolio and, you know, we were just putting bridges in and things like that at the time and paying, turn another card, as we'd say, then all of a sudden, it was like, wait, we only have enough, you know, commitments, we only have enough capital left to fund a certain amount of deals. And so then it became really interesting because, you know, it was like, you know, stack ranking your portfolio. And that was fun because there's many partners. So anyway, but... Anubhav, I'd love to see what you have, or hear your thoughts as well.
Anubhav - Well, you know, a reserve strategy could be an entire hour session on itself. I'll try to make this into a 45-second version.
Sean - Okay.
Anubhav - First of all, if you reserve too much, most managers sometimes forget that that reduces your TVPIs. If you reserve too much, follow-on dollars will always have a lower yield than your initial investment, if it's an upround, which in this market it is. So, you don't want to reserve too much because you're probably lowering your TVPI. On the other hand, if resort too little, that means you're going to be doing a lot more initial deals. Do you have the team to support that? Can you diligence that number of deals? So those are some of the qualitative factors that come into play. Having said that, so what we do in Tactyc is we rank your deals. All of your portfolio companies, we show a chart that says, "What are your expected returns on the follow-on investments? Or we show the follow-on MOIC, which is just a MOIC on just your follow-on dollars, and you rank all those companies. I'm on getting the details because this is not a demo and I don't want it to be, but the idea here is for every single company, you can define a base case, a downside case and an upside case, and we will actually track on a probability weighted basis. What is your following-on MOIC look like for that deal? Then that's a metric you can compare against all of your other deals across your portfolio. Sean, to your other point, what is the market doing?
I actually have data for us to look at. So, this is for Q3 2021. This is the follow-on reserve percentages on average across the funds that that we've looked at. And there's some interesting trends to look at here. So, if you look at the smallest, the less than 50 million and the greater than $200 million funds, they have the widest variance of follow-on strategies. They are investing as little as 22% in reserves all the way up to 60%, while the midsize funds, 10 to plus are up around 45 to 50%. So what does that mean? It means some of the smaller funds are actually willing to take the bet that they're not going to fall on as much. They're increasing their concentration on initial investments. That strategy does not seem to be the case for some of these mid-size funds. Now, this also highlights the fact that the point that Taylor made, that the reason that there's a 22% here or a 34% here is because guys might have another fund, an opportunity fund or an SPV fund that's not being captured here. We're seeing those on the rise a lot as well. So just so you know, I mean, it depends on how this performance is going to bear out. It's 10-year fund, so we don't, it's too early for us to say, but we are seeing this divergence of different strategies coming to play.
Sean - I'd say, you know, I see kind of the same thing. I mean, if you're a giant sand hill road, it's interesting. I was talking to the CFO the other day and he was mentioning that the reserve strategy is just nuts now because all the deals are getting preempted, and how do you reserve for, you know, a firm that's a unicorn within, you know, eight months of the initial check. So, there's definitely that, but then on the earlier stage side, it's definitely, you know, one of the things I see and I think is kind of an interesting approach. It's sort of, you know, you can definitely say you reserve and you have that fundable reserve for the seed through series A and you just kind of, you know, you can choose to back your winners, but you can also pull out into an SPV strategy too, and then offer that up, it's also in a reserve, you know, or an opportunity fund and that allows, you know... You know, that's a whole other discussion, but you know, your institutional investors or folks that wanted that particular deal, you know, they can still get in essence you're still reserving from the company's perspective. I don't know, Mike, is that something that you guys think about?
Mike - Well, we got to ask that, you know, we're at end of our second fundraise and so many of our potential LPs we pitched, we're asking about like, "Are you guys doing an opportunity fund?" I think they're just used to, I think they expected us to say yes, because I think everyone's raising as many funds as much capital as they can possibly do. Again, like we had a look at ourselves internally and what we were good at and wanted to do right now and, you know, manage and we thought, no, you know, not right now, we don't want to raise an opportunity fund. We just, you know, decided to sort of up, you know, our follow-on from 40 to 50, and that was actually informed by the model, but you see it a lot, and yeah, to lot of better said it, Sean, but like so many companies are going back. I mean, there is a lot of capital out there. So, I think founders feel like they can go back whenever. Like, opportunistically, they don't need it, but we should go get it. And so, there's a lot of companies going back, you know, quicker, I think either the math, like, you know, normally it's every 18 months you're fundraising is the rule of thumb. I think with our companies, so far in fund one, the average time back to market was something like 225 days across, you know, the first 50 companies we've done in our first three years. So, you know, the data sort of supports that these are happening, you know, these follow-on rounds more often, which just you got to think about how that affects your follow-on strategy.
Sean - Yeah. All right. Let's talk about some of the, I know common issues and mistakes that we see. I mean, we've all seen a few fund models and we've all probably, well, I know we've all made a few, you know, let's just say there's been learning along the way. But Taylor, you know, throughout some common issues you see, and you know, any advice you'd give folks as they think about, you know, putting together their fund models?
Taylor - Well, what I mean, reserve strategy is usually one of the big ones, right? And the way that that usually comes up is, people have an idea of like, what their initial check size are going to be, and they make an assumption of what their second check or the third check's going to be and does it... If you do the math, does it make sense, right? Not thinking about in terms of what prorata and their ability to actually get into those quality deals at the check size they're assuming. So, it's very common to like assume a check size strategy in terms of follow-ons that doesn't match what's actually possible to do, kind of given the market, given kind of what's rationally out there. That's usually like the biggest area that comes up. Another one big does it comes into like a capital deployment budgeting for fees, budgeting for fees over time.
Sean - Yeah.
Taylor - It's usually another big one. We haven't touched on recycling yet, but like thinking about how kind of model and recycling and how that-
Sean - Yeah, before we get to recycling, can you actually go back and explain, you know, why understanding, you know, let's call it partnership expenses or fund, you know, fund expenses, you know, why is that important? Why do we need to understand that? You know, who cares?
Taylor - Well, I mean, you know... All the money you raise for investors is not money you're actually going to invest in the company. Some is going to go towards management fees to support the fund, and some extra fees get charged to fund on top of that. For a first-time manager, they usually don't know. What fees get charged with the fund? What fees don't get charged with the fund? And so it's usually like... if it's your first time doing it, there's generally some confusion in terms of how those fees work and then how to budget from appropriate labor time. And the second part comes down, doesn't from an overall perspective. And the second part that comes in is how to actually think about those over time. because those fees are going to change expectations of those expenses to kind of budget for it, is going to change over time as you deploy capital. And so just thinking about the need to reserve capital to kind of operate the fund over the life of the fund is usually kind of one of those things that comes up there. Did I answer your question?
Sean - Yeah, it does. I mean, I'll just elaborate a little bit in that, you know. Over a 10-year life of a fund, if you just take into account management fees and fund expenses, you know, somewhere north of 20%, you know, at least 2% a year on average is not investible. So, if you are putting together a fund model and you're allocating the entire committed capital to investments, you know, that's a problem, because you don't have that much investible capital. So, I do see that more often than I'd like from, you know, new managers and that's part of just the learning around, you know, what's expense that the fund pays versus, you know, comes out the management company. Anubhav, do you see that much or do you hear of anyone, or get those questions around, what's a fund expense? How do you model management company expenses versus fund expenses?
Anubhav - Yes, what we've seen is folks tend to not spend enough time on the expenses. They actually spend a lot of time on the management fees, but expenses are not something that they usually build out a full budget for. Sometimes it's just a little bit of vague, and not as much thought has been put into it. And the reality is, I state up on it out, it reduces your investible capital. So that's actually just dollars that are never going to see the lighter day in terms of real investment. So, you want to take that time to do that. And then recycling, I don't know if you want to dive into recycling now, but that's what we see happening.
Sean - Let's go for it
Anubhav - Why don't you start us off, and let's touch on some recycling. We don't have a whole lot of time-
Anubhav - I know, I know.
Anubhav - Well, the one thing is, and I think both Taylor's model and Tactyc supports this is, there's multiple forms of recycling, and we see a lot of confusion in the market on exactly what is what. So for example, there's a concept of just straight exit proceed recycling, which means whenever you have an exit, you take some percentage of it, and that's applied to new investment. You don't charge fees on that. That just goes into new investments. And then there's a concept, which is frankly a misnomer called management fee recycling, which might lead one to believe that, hey, maybe the GPs are not owning their fees or maybe they're deferring their fees. That's not true. Management fee recycling is also exit recycling. It's just capped up to the level of management fees on to date. What we've seen in the market is in general. 20 to 30% of your committed capital is usually recycled. So that's where the LPs get back their fees as well as maybe a little bit of expenses, but that's usually the norm that we see. We're also seeing, and this may be, I don't know if this is helpful or not, but we ran some numbers around, how many of our clients were successful in raising a fund, and how many were not able to fully raise a fund. And by and large, the data is pretty clear that if you are having some sort of recycling, and you are aligned with your LPs, you have a higher chance of actually raising that fund successfully. So we're seeing that happen a lot more.
Sean - Do you recycle Mike, or can I ask you that?
Mike - Yeah, we've modeled out 100% of management fee recycling. You know, there's timing issues obviously that come into play because you have to have distributions to have the capital to recycle, and they sort of have to time out as follow-on because it'll probably be follow-on rounds, most likely, at least for us since we're seed in our active periods between two and three years. By the time we start to see exits, usually we're done with the active period. It's just follow-on investing that we're doing. So, the timing sort of has to work out, but absolutely, you know, it was a pretty easy decision for us as we want to put the full size of the fund to work, you know, to make sure that we can get the highest return we can.
Sean - Yeah, I mean, I would say, you know, it's a very institutional limited partner concept. If you're raising, and somebody ask this question, you know, what's the sort of common size of a first-time fund, and you can't actually just put out some data on that on first time funds. It's around 10 and then there's another, you know, it's like a bill around 10 and 25 is what they had. But usually if it's friends and family that you're raising the first fund for, I'd say recycling is not something that you really have to deal with a lot. You don't have to kind of talk to your investors because they're generally sort of, you know, friends and family, they're not sophisticated institutional investors that want more shots on goal, but I definitely think that, you know, you need to understand what it is and know that, you know, because you can't invest, well, under a normal funds constraints, you can't invest 100% if you commit capital due to the fees. You need to be able to recycle. And there is a term in your partnership agreement that usually caps recycling into 120% of the funds. But you know, everyone has to look at their own partnership agreement to see where that's at, but it's all right. Common mistakes or issues that you see in fund models, Anubhav, you got anything there that-
Anubhav - Yeah. Well, one thing I see a lot is clients coming to us saying, "I want a portfolio of 30 companies that's going to give me a four X return, 50% follow-on reserves, and this is my check sizes." Sometimes that still sound possible, because sometimes what you've told me is actually a calculation. It's not meant to be an input. And this sort of goes back to what's an input, what's an output, but what we've seen as sometimes the managers may not be as open in their construction process. Sometimes the entire commitment size could be a variable. You could actually change different commitment sizes if you want to land at a specific zone in terms of TVPI, a number of deals. So, one thing I encourage managers is to really think about, what is absolutely a constraint. Is it number of deals? Because you don't have enough of a team. Is it check sizes? Is it because you want a certain ownership? What exactly is a constraint? And let all the other pieces fall out where they may, instead of trying to fit into a specific profile that you might be looking for. The second thing I've seen a lot is building a very deterministic model, and it's a spreadsheet of 30 investments, and this is exactly how my fund is going to work. Two of them are going to succeed, the rest of 28 are going to fail. That might work just for construction. But when you start to move into, if you want to start to do these other things, you want a construction tool that can be updated for real investments, and when a course corrective fund, that's not going to work. That's where you need a little bit more flexibility in your spreadsheet or in whatever tool you might be using. That can incorporate all of that, you know, that flexibility for actual data coming in. So the other thing is, don't build a very deterministic model because it's not going to serve well once your fund is launched.
Taylor - I just want to add one point to that. It is a common way that I see people to build models is that instead of using approach, say, hey, I have a certain amount of capital I want to invest, and here's my average company, here's how deploys over time, which is one way of handling it. In other way, say, here's all the specific companies I'm going to invest into, or the specific types of companies and certain times, and the fallen ranch checks I'm going to do, and here's when the X are going to happen, here's how much X is going to be. You can do that. And it's not about exercise to have to think through that, so actually like makes you think rationally about how things work. At the same time though, when you start changing assumptions or changing ideas, it can be hard to manage that process and making sure the model actually structurally works. It can be hard to create that deployment of capital, so it makes sense and matches up fees and matches up the total committed capital. It's a common error that happens when you try to kind of build a very per investment style structure, but also it introduces the idea, if you have to make a lot more assumptions, a lot more data input, creates a lot more room for errors, but also because, you know, we know that, you know, that the return structures over a lot of companies is very based upon power laws, and so there's going to be a couple big exits, a lot of zeros called big exits. They're going to happen at some point in time. And so, if you build a model like that, you know, when you make that specific assumption for when that specific big exit happen, so actually can have a big impact in swing or overall returns. And so, you can create a lot more variability into the actual kind of results, your forecast, because you're making much more, a large number of very specific assumptions. So, you can do it that way. It's not bad to know it, but it can be a lot more kind of difficult world to manage. It's something I see every month.
Sean - All right, thanks. Well, I think we are getting really close to the end here. I don’t know if we want to try and we've got maybe a minute to try and tackle some of these questions. I don't know if you can see some. Let's see, there are a lot of inputs for the manager, manager needs to figure out what valuation of fair market value of companies. I think we touched on that one already. There are a lot of sources that you could look at. Anubhav, do you want to hit that one quick?
Anubhav - Yeah. Yeah, basically it depends. What we do is we assume the exit valuations to be as a starting part of the same as a pre money valuation of the next round. You're welcome to change that depending on your sector profile, but that's the standard valuation profile we follow to come up with a performance.
Sean - All right, and I think Mike, you hit this one already. A lot of managers say you can't really understand portfolio construction until you mess up in a real-life scenario. How do you think about the discipline and modeling versus real life execution?
Mike - Yeah, well, I mean, that was a mistake, not a mistake, but something that we learned, I think was that, again, I mentioned in the beginning, we originally said we were going to invest in our top 20, 25% of companies, but in reality as life happens, you know, your companies come back to you. It is hard to say no. You know, it's not black or white in these early days of should you follow-on or not, and these become emotional decisions often. And the math and the models may tell you something that, hey, I probably shouldn't do this. There's higher MOIC on three other follow-on investments, but there's just something about this founder I think they're going to... You know, and so it's an imperfect practice. And so, I think that's just, you know, life is different than a model and things, you know, never quite match up.
Sean - All right, well, we're out of time now. We'll try and answer some of these questions offline. Taylor, Anubhav, Mike, really appreciate you being on the webinar today. It was really great question, very thoughtful and you know, for everyone on the webinar, thanks for joining. We'll see you the next time. Thanks everyone.