# VC10X - Using Data & Algorithms To Invest In The Best Teams - Collin West, Founder, Ensemble VC

An episode of [VC10X - Investing, Venture Capital, Asset Management, Family Office, Wealth Management](https://invisiblenetwork.ai/podcast/1632806986), 10 Jun 2025, 38 min.

Audio: https://anchor.fm/s/a453ddd4/podcast/play/103931270/https%3A%2F%2Fd3ctxlq1ktw2nl.cloudfront.net%2Fstaging%2F2025-5-10%2F401914601-44100-2-4bfea2981e1b5.mp3

Collin West is the Founder and Managing Partner of Ensemble VC, a venture capital firm he launched in 2020 to harness data science for discovering and supporting exceptional founding teams. Collin’s investment track record spans industry leaders like Zoom, Carta, Groww, ICON, Tally, and Sidecar Health. ⭐ Sponsored by Podcast10x - Podcasting agency for VCs - https://podcast10x.com Ensemble VC website - https://www.ensemble.vc/ Collin West on LinkedIn - https://www.linkedin.com/in/collinwest/ 🎙️ Hosted by Prashant Choubey Follow Prashant on X - https://x.com/ChoubeySahab Follow Prashant on LinkedIn - https://www.linkedin.com/in/choubeysahab In this episode we talk about - - How they leverage data & algorithms to find the startups with the best founding teams - What defines a good founding team for them? - The secret behind his great track record with multiple unicorn outcomes like Zoom, Carta, Groww among several others - How AI will change how VCs operate? - The new AI co-pilot for VCs that they've recently launched & lots more Timestamps: (00:00) Introduction (01:00) Collin's background and early entrepreneurial journey (02:59) Story of crossing the Arctic Ocean and lessons learned (05:11) How Ensemble VC uses data to identify and invest in startup teams (09:50) Discussion of successful investments like Zoom, Gro, and Carta (13:23) Addressing potential limitations of data-driven investment approach (15:19) How Ensemble VC wins access to the best deals (19:28) Detailed discussion about specific investments, including Zoom (23:14) Introduction of AI co-pilot for VCs (25:16) Traits of successful founders and team building (27:28) Current trends in venture capital and AI (29:35) How Ensemble VC supports portfolio companies post-investment (31:31) Biggest learning in startup investing (33:44) Potential of AI to preserve investment strategies (35:16) Advice for startup founders (36:19) Rapid fire round of questions about Ensemble VC's investment strategy For sponsorship or guest appearance requests, write to prashantchoubey3@gmail.com Subscribe to VC10X on Youtube, Spotify, Apple Podcasts.

## Transcript

[0:00] VCs lose money at an embarrassing rate.
[0:02] What other industry would we accept or even celebrate a
[0:05] failure rate of 80, sometimes 90%?
[0:09] So The Dirty little secret venture is the majority of these
[0:13] firms, Sean, are running on Excel sheets and relying on
[0:18] handshakes, vibe checks and physical networks.
[0:22] So instead of trying to wait for the best deals to come to us
[0:25] like, which is the old traditional way, we go out to
[0:29] the entrepreneurs to score best in our algorithm.
[0:33] We have a tendency to lionize founders as these like mythical
[0:37] creatures. And like, the reality is they're
[0:38] not doing it all in the near term.
[0:41] I actually could see solo GPS at early stage be able to be able
[0:48] to manage a billion dollar fund. That's a legitimate possibility
[0:51] with new efficiencies in AI. Hey everyone.
[0:56] This is Prashanth and I'll be host of the VC10X podcast.
[0:59] And today we have Colin W with us.
[1:01] Colin is the founder and Managing Partner of Ensemble VC,
[1:05] a Winds Capital firm he launched in 2020 to harness data science
[1:09] for discovering and supporting exceptional founding teams.
[1:12] Collins investment track record spans industry leaders like
[1:15] Zoom, Carta Grow, Icon, Tally, and Sidecar Health.
[1:20] In this episode, we talk about how they leverage data and
[1:23] algorithms to find the startups with the best founding teams.
[1:26] What defines a good founding team for them?
[1:29] The secret behind the great track record with multiple
[1:33] Unicorn outcomes like Zoom, Carta, Grow, among several
[1:36] others, how AI can change how VCs operate, the new AI copilot
[1:43] that they have recently launched for VCs and lots more.
[1:46] So without wasting any time, let's dive straight in.
[1:48] This episode is sponsored by my own podcasting agency called
[1:52] Podcast NX, where I help venture capitalists, asset managers,
[1:56] allocators and family offices and starting their own podcast
[1:59] and building a strong brand. Podcast NX is designed in a way
[2:03] to work well with the busy schedule of investors, where we
[2:06] handle every aspect of podcast production so you can focus on
[2:10] investing. If you're looking to build a
[2:12] strong investor brand, starting a podcast should be right up
[2:16] your priority list. It's the only form of marketing
[2:19] where you can showcase your strong network.
[2:21] In other words, access to limited partners and founders
[2:25] and grow that network further. And contrary to what most people
[2:29] hear, it's still very early to start a podcast.
[2:32] There are at least 10X more newsletters and thousand X more
[2:35] blogs than there are podcasts in this world.
[2:38] You can learn more about our podcasting services at Podcast
[2:42] ten-x.com. I'll make sure to add the link
[2:44] in the description below. Enjoy this episode.
[2:47] Hey Colin, so good to have you on the VC10X podcast.
[2:50] How are you doing? Great.
[2:51] Thanks for having me. I really appreciate it.
[2:53] Yeah, it's a pleasure having you on.
[2:54] And to start things off, can we first have your story and how
[2:58] you started investing? Sure, absolutely.
[3:00] So I was born and raised in Dallas.
[3:02] Father owned a car wash where I worked, you know, days,
[3:06] weekends, nights, and learned a lot about kind of the local
[3:10] entrepreneurial journey. After going to SMU in Dallas for
[3:15] my undergrad, I started two companies, bootstrapped and sold
[3:18] them, and then went to Business School.
[3:21] While at Business School, I did this kind of wild expedition
[3:24] where I led the first team ever to cross the Arctic Ocean.
[3:28] And so breaking Guinness Book rowing about 41 days non-stop
[3:32] unsupported through the biggest storm seas in Arctic history.
[3:37] So there are a few lessons from that.
[3:38] One is is it makes it makes startup investing seem a lot
[3:43] less dangerous, which is certainly true.
[3:46] We almost died a lot. It was really dangerous.
[3:48] And and then the second thing is how important a great team is.
[3:53] If you have a really great team all literally rowing in the same
[3:56] direction, you can do something that's never been done before.
[3:58] So I kind of took that insight with me after crossing the
[4:03] Arctic and joined the pioneering first data-driven venture fund,
[4:08] Correlation Ventures, where I was a junior team member.
[4:11] Learned a ton from the team there on the initial building
[4:15] blocks of a data-driven strategy.
[4:17] And back in 2012, you know, to go spout that you're going to
[4:20] use data and venture was pretty wild at the time.
[4:23] So we were literally sometimes getting laughed out of rooms by
[4:27] LP's and pitch meetings. But nonetheless, Correlation is
[4:31] done fantastic. I'm still an advisor there and
[4:33] really excited and thankful for their mentorship.
[4:36] And then around 20/16/2017, I started seeing a lot more data
[4:40] on management teams come online. And so the core thesis kind of
[4:45] harking back to it being all about the team, which I learned
[4:49] crossing about crossing the Arctic Ocean was maybe there was
[4:52] a data-driven strategy where you could start to track at scale
[4:57] full startup teams and find companies that are talent
[5:00] magnets and go find those companies before everyone else.
[5:06] And so this is how I got into startup investing and now what
[5:09] I'm doing leading ensemble VC. Absolutely.
[5:12] That's quite interesting. And it seems you have a strong
[5:15] focus on data and and founding teams especially, right.
[5:19] So how do you judge that and how do you try to track that data in
[5:23] practice and try to find out those gems?
[5:26] So how does that work in practice?
[5:29] Yes, our investment thesis really starts with conventional
[5:34] VC is really kind of a 20th century invention.
[5:39] It really doesn't scale up for 21st century.
[5:41] And you know the ecosystem entrepreneur entrepreneurs is
[5:45] now so large and around the world venture really started
[5:49] kind of in this post war America where the first wave of VCs kind
[5:53] of came to California a couple decades after that.
[5:57] The system really worked well to create a lot of wealth for those
[6:01] first couple generations of VCs when tech, when tech was really
[6:06] hyper concentrated in what was, was now become Silicon Valley.
[6:10] But this year, think about this, 250,000 tech startups around the
[6:16] world will try to raise money. About half of those will be in
[6:19] the US. There's about more than 27
[6:22] million people qualified to help run these companies.
[6:27] It's just impossible for anyone to assess all these companies.
[6:31] And even if you could, you just be meeting the founder.
[6:34] And, you know, I was a founder myself and I was not the most
[6:37] impressive person on my team. My job was to recruit people
[6:41] better than me. So you're just missing me
[6:43] meeting me. You're missing a lot.
[6:45] And so you certainly can't meet every founder.
[6:47] You definitely can't meet every person on every team to help
[6:51] kind of figure out which teams are more likely to go into hyper
[6:53] growth. And so as a result, there's this
[6:56] huge problem. VCs lose money at an
[6:59] embarrassing rate, I hate to say it.
[7:02] And what other industry would we accept or even celebrate a
[7:06] failure rate of 80, sometimes 90%?
[7:09] And So what I'd like to say is let's just admit that the old
[7:14] way of investing, which a lot of times starts with these inbound
[7:18] inquiries from desperate founders or vibe checks at a
[7:22] coffee shop with founders, it just isn't analytical and it's
[7:25] just not scalable. And it actually totally wastes a
[7:29] ton of time for all parties, you know, and then loses a lot of
[7:33] investor money. So I think This is why 6 times
[7:35] as much money goes into buyouts. The top decile for buyouts is
[7:39] nowhere as interesting as top decile for venture, but buyouts
[7:44] rarely lose money. And So what we need is we need
[7:47] to build a system of investing into startups, which is much
[7:51] less risky so that more money can flow into the space, more
[7:54] companies can be founded and funded, and we can get to the
[7:57] future faster. And so that's what we believe
[8:00] we've done. We've found a better way to
[8:02] improve loss ratios, obviously juice returns as a result, but
[8:06] all along the way actually supporting truly groundbreaking
[8:10] founders, not use these derivative, you know, photo
[8:13] sharing app or something like that.
[8:14] And the way we do that is with truly AI native BC.
[8:18] So how you know, in Ensemble 10, we're about 10 people.
[8:23] Half of us are data scientists. The team scrapes public
[8:27] databases, the Internet, LinkedIn X college and corporate
[8:31] biographies, SEC filings, any bunch of other sources.
[8:35] We track over 400 data points on every individual person at a
[8:39] startup. That helps us rank the startups
[8:42] and even those individuals such as, you know, even first time
[8:45] founders according to their likelihood of commercial
[8:48] success. So instead of trying to wait for
[8:51] the best deals to come to us like which is the old
[8:53] traditional way, we go out to the entrepreneurs who score best
[8:58] in our algorithm and it's worked really well.
[9:00] So about half of our investments historically become unicorns.
[9:05] We were early investors in companies like Zoom Video Grow,
[9:08] which is the Robin Hood of India, Carta, Cycar, Health
[9:11] Icon, Seronic Chaos, and and we just think that that's going to
[9:18] continue to go that way because the thesis truly is all about
[9:22] the team. So this algorithmic approach is
[9:25] exactly the opposite of what most VCs do.
[9:30] Most VCs try to build a platform and market themselves to get
[9:33] deal flow flowing to them. We don't even have a classical
[9:37] web form asking entrepreneurs to send us their pitch deck.
[9:40] I don't I don't really want an inbound e-mail from founders.
[9:43] We approach the founders that we want to invest in and so as a
[9:48] result, we just look very different as a team.
[9:50] That's quite interesting. So it's a very proactive
[9:53] approach to investing and you're actively seeking out the best
[9:57] teams and then investing in them.
[9:59] So you mentioned a few investments.
[10:01] I know there's Zoom in there, there's grow in there.
[10:03] Carta and several others are really successful start-ups.
[10:08] So what all these investments also sourced through the system
[10:12] that you're mentioning or what? There may be different
[10:16] mechanisms for that. Yeah, exactly.
[10:18] So we're, so our strategy is to use the data platform to
[10:21] objectively find the best teams. And if those teams score well in
[10:27] our algorithms, then it's our job to basically say no to as
[10:30] many of those as possible through our diligence process.
[10:33] But we don't do investments in companies that our data-driven
[10:37] platform believes are not strong teams.
[10:40] OK, so, so even the algorithm is sort of influenced by some
[10:45] thought, right? So and someone, a human possibly
[10:49] has coded that, right? So what do you think makes a
[10:52] great team? What are the traits of that?
[10:54] What does the algorithm look for?
[10:56] Yeah, that's a good question. So for us, we're really looking
[10:59] at four major different buckets of information.
[11:04] And what's interesting is when humans look at a start up, we're
[11:08] generally pretty correct on the things that correlate with
[11:11] success. However, we're just our brains
[11:16] just can't handle that much data and we kind of treat everything
[11:18] as a binary one or zero rather than a weighted factor.
[11:23] And so the major buckets are industry experience, track
[11:28] record of success, networks and educations.
[11:31] And so our platform has really built this ranking, mappings,
[11:37] mapping and rankings platform that's really developed help us
[11:40] have pattern recognition across the startup ecosystem.
[11:44] So one primary truth does emerge over and over and it's that the
[11:49] best teams are much more likely to win.
[11:52] And so this is true of any sector.
[11:54] This is from defense to SAS, to robotics, AI, space tech, you
[11:59] know, all places that we've invested in.
[12:01] And it's all based off of the team.
[12:03] And so the real question really becomes is, and the question
[12:07] that you really need to lean into as an investor is can you
[12:10] identify the best teams early and when they're really still
[12:14] forming? And so our agentic model really
[12:17] helps them. So like to give you an example
[12:19] of this, one of our highest value products that our data
[12:22] science team produces is a real time talent mapping.
[12:26] So if a high talent individual is going to leave your SpaceX,
[12:30] your Androl or Palantir or another high performing kind of
[12:34] tech mafia, they're obviously incurring significant risk,
[12:40] something they'd only be doing if the upside potential were
[12:44] enormous. And so our data platform
[12:45] actually in real time provides a technical, a view of technical
[12:49] talent flow, really anticipating and industry shifts before they
[12:54] materialize. In fact, this is like, This is
[12:57] why we started getting into defense maybe four or five years
[12:59] ago before folks realized defense was interesting and
[13:02] everyone was kind of turning their back on it.
[13:04] We started to see really high quality individuals flow into
[13:08] those spaces and we just follow the talent.
[13:11] And so as a result, we've built, you know, one of the best
[13:14] defense portfolios out there, not really truly based off of a
[13:20] thesis there, more based off of that's where the best people
[13:23] were going. Right.
[13:25] That's quite interesting. So one counter view to that
[13:29] might be wrong though, is that you might still be missing out
[13:33] on some outliers who might be taking one or two boxes of the
[13:36] four that you mentioned, but might not be taking some other
[13:40] boxes that would not show up in your good team's results, right.
[13:45] So how do you think about that? Yeah, that's a great question.
[13:48] So you know, what's interesting about venture is, I mean you
[13:52] pass on so many good companies as as a part of your job and
[13:57] that would be a false negative, right?
[14:00] That's a bummer when you pass on a good company, but what's worse
[14:04] is a false positive because then you've taken your LP's investing
[14:08] dollars and you put it into a company that lost money.
[14:11] And so for us, we think of everything as kind of Bayesian
[14:14] probabilities. There's so many companies in the
[14:16] long tail that even companies that have a .001% chance of
[14:20] success, there's going to be some huge successes in that
[14:23] bucket. But again, with this scalability
[14:26] problem, that venture has 250,000 companies.
[14:29] You can't meet them all. So how do you actually sort by
[14:32] an objective criteria that gives you a much shorter list of
[14:36] companies that are much more likely to be successful?
[14:38] That's what we think about. So if you think about it as a
[14:41] phishing analogy, it's not that you're gonna catch all of the
[14:45] big fish. It's are you in the right area
[14:48] of the right pond at the right time of the day where you're
[14:51] much more likely to catch a big fish and we just want to spend
[14:54] all of our time there. Absolutely.
[14:56] Yeah, that makes complete sense. But also these startups when
[15:01] they meet all these criteria that you mentioned are also very
[15:06] interesting to all the other VCs out there.
[15:08] And VC also is a very competitive field, right?
[15:10] So how do you make sure that you are the one who is getting in
[15:14] front of that deal and not just getting in front, but also
[15:16] getting the deal, winning that deal, right?
[15:19] So tell me about that. Perfect example, I'll use a case
[15:21] study for you, a company we invested in a company called
[15:25] Seronic, because what you're bringing up is access.
[15:29] So no matter if you have the best algorithm in the world, if
[15:32] you can't actually get access to the investment, it's all for
[15:34] naughty. And so Ceronics a really good
[15:36] example of this actually, to be frank, when I first was meeting
[15:41] Ceronic, I initially had my own biases whereas and So what
[15:45] Ceronic is autonomous boats for the Navy basically swarm boats,
[15:50] surface vessels have a lots of different capabilities.
[15:54] And initially I was showing showing up thinking all the
[15:58] classic investor biases. Oh, this is building a hardware
[16:00] product that's harder than software.
[16:03] Oh, there's only the military is the main buyer, and if the
[16:07] military doesn't buy, that's a huge risk.
[16:09] There's a bunch of reasons to say no to the company, but
[16:11] whenever we actually looked at the company, we could see that
[16:14] Dino, the founder, was exceptional.
[16:16] Great experience in the Navy being a part of the Navy Navy
[16:21] Seals, but then also below the surface.
[16:23] He had recruited this incredible team from Androl that you know,
[16:28] is one of the best in autonomy and defense.
[16:31] And then he'd also recruited the head of manufacturing from
[16:34] SpaceX, the guy who literally built the Raptor engine from I
[16:38] think it was something like one a month or to a couple a day.
[16:42] So he was really good at high speed manufacturing and they had
[16:45] really good autonomy team. What is seronic?
[16:48] It's high speed manufacturing of autonomous boats.
[16:51] And so we could see that the the team was really exceptional.
[16:55] So instead of in the classic venture approach where you have
[16:59] like an associate or some junior member of the team cold e-mail
[17:03] this founder asking for a time for them to show up to a Zune
[17:06] and you know, bend the knee and beg for money from me, which is
[17:11] something that the Serrano team is never going to do.
[17:15] They had plenty of options. And so, and so instead of that,
[17:20] we're not cold emailing. In fact, we go to our one of our
[17:23] portfolio companies and ask for an intro to the founder and they
[17:26] give, they gave the founder a high recommendation.
[17:29] And the first meeting again is not just a Zoom video.
[17:32] In fact, all the founders of my firm, our entire investment
[17:35] team, we went to their to their warehouse manufacturing facility
[17:42] for our first meeting. And so that looks very different
[17:45] than come join this Zoom and beg for money.
[17:49] Instead, let's go meet the rest of the team.
[17:51] Let's see what we can do to help.
[17:53] We already have insight on this company.
[17:55] I am not on endless sourcing calls with random companies all
[17:59] day because I have a target list of the best teams to go focus on
[18:02] and so I can just spend a lot more time with them. 80% of the
[18:05] companies we meet aren't even fundraising yet.
[18:08] We provide all sorts of resources in our products where
[18:11] we help our, when we're doing diligence, we help them find
[18:13] customers, leveraging our data platform.
[18:16] These types of things really help us help the company even
[18:18] before investing while parallel and parallel diligencing them.
[18:22] So, so that's how we win access. And so we, you know, we were
[18:27] super lucky to be invited into investing in the Series A and
[18:30] Saronic. And now you may have seen, you
[18:33] know, just 24 months later, they're now valued over $4
[18:35] billion. And it's and it's and it's an
[18:38] incredible business that we're just so grateful to be a part
[18:42] of. Right.
[18:43] Absolutely. I, I love this very targeted
[18:46] approach, kind of flipping the script on what founders have to
[18:50] do to get to VCs and you're trying to do the same for the
[18:54] best founders out there, the best founding teams out there,
[18:57] going out, reaching out and going out of the way to meet
[18:59] them. So that's, that really stands
[19:02] out and helps you win deals surely.
[19:05] And let's talk about some of the specific investments you've
[19:08] made, right? So Zoom is a very popular one.
[19:12] I would say that would that might be the highest valued
[19:15] company in your portfolio. Correct me if I'm wrong, but
[19:18] tell us about how you ended up investing in Zoom.
[19:22] At what point was that? How do you get in touch with the
[19:26] company? At what stage and how did you
[19:28] close the deal? Yeah, it's a good question.
[19:30] Yeah. So at this point, you know,
[19:32] we've had several unicorns. Zoom is certainly the highest
[19:36] value in market cap, however, it's not the highest in
[19:39] multiple, which is actually quite interesting.
[19:43] Zoom is a little bit later stage where typically Series A
[19:46] investors bell curve wise. Most of our dollars go in to you
[19:50] know 12 to $15 million rounds, sometimes earlier, sometimes
[19:54] later. We flexed up to do a little bit
[19:56] later stage in Zoom, but still that had incredible returns for
[20:00] us as you can imagine. Other interesting companies like
[20:04] Grow, which I think is a really fun one because this is company
[20:08] based in India where we found with the data system.
[20:13] I actually haven't even been to India yet, I'm ashamed to say.
[20:17] And this may be the best investment I ever do just kind
[20:20] of crazy. Yeah, yeah.
[20:22] And and so we did that at the Series A at a fantastic
[20:26] valuation and they just continue to grow.
[20:29] They're now India's largest investment brokerage.
[20:31] It's a, it's a, it's a very real, very big business that'll
[20:36] continue to be big in a short period of time.
[20:39] We're early in Carta. We, I mean, even just six months
[20:42] ago we invested in chaos, which is incredibly new technology
[20:47] around radar and defence and, and within six months now
[20:52] they're valued at $2 billion. And so the underlying truth is,
[20:58] is all of these just started with great teams and not just
[21:02] the founder. I think when people talk about
[21:04] the team, they tend to think it's the founder.
[21:06] It's not just the founder, it's the all-encompassing ensemble is
[21:09] our name, and we've just been really lucky to get to work with
[21:12] these folks. Absolutely, Yeah, Grow is a very
[21:17] big company here and I run some of My Portfolio in that nice in
[21:21] that app so. Yeah, please do.
[21:23] Anyway, any listeners in India use Grow?
[21:26] Yeah, over there. The founder is incredible.
[21:29] Absolutely. And maybe I might also put it
[21:33] that way that you might have might just have the golden touch
[21:36] when it comes to investing in startup that whatever you touch,
[21:40] you know turns to gold in a way and the valuation kind of
[21:43] skyrockets after you invest. How much would you believe that
[21:48] is true? Have you seen some of that
[21:49] happen that will be invested the next round or next six months or
[21:53] maybe one to two year, The company really takes on a
[21:56] different trajectory. Certainly things have gone well.
[21:59] I do think that it's more about better decisions and building a
[22:02] process rather than on individual deals.
[22:05] I think it's also very much a team thing.
[22:09] I mean half of our team is data scientists, engineers,
[22:12] mathematicians, physicists. I mean we, we're a product LED
[22:16] company where this is data is at the core of everything we do.
[22:20] Whenever we founded this company, we decided to make a
[22:22] big bet where we've spent half of our budget on this data
[22:26] system, which is incredibly high compared to typical VC.
[22:31] We're all in on this and the decisions we've made there have
[22:36] meant that as we get good at things, they're actually
[22:38] scalable in a way that we can repeat them.
[22:42] Venture returns are notoriously hard to repeat and notoriously
[22:46] hard to build a process that's predictable.
[22:49] As a result, LP's, people that invest in VC get scared and have
[22:55] a hard time. You know, maybe you have a good
[22:56] fun, but then you have no idea what's next.
[22:58] How, how are you building this repeatable process?
[23:00] So I don't think I have a golden touch.
[23:01] I do think that we've made good decisions on building a process
[23:04] and we have an incredible team working on that every day.
[23:08] And and so I do believe that we're gonna be more repeatable
[23:12] than the typical human making decisions.
[23:15] Absolutely, Yeah, makes sense. Great answer.
[23:17] And talking about the new AI copilot for VCs that you've
[23:22] recently launched and how will that help VCs back future
[23:27] witness? Is it the same thing that we've
[23:28] been talking about all along or is it somewhat different?
[23:31] It's somewhat different. So, so think about how we use
[23:35] our data platform internally to help us identify these companies
[23:38] to invest in. It's really an AI first approach
[23:41] to de risking venture investing and being proactive by
[23:45] identifying the best teams. However, our fund focuses on a
[23:50] very specific subset of those companies that's series AUS
[23:54] mostly. It turns out many Sand Hill Rd.
[23:58] institutions want to use Unity, our platform as well.
[24:02] In fact, we now have VC firms like Draper Associates with Tim
[24:06] Draper, Boost VC, Adam Draper, Overmatch already using Unity
[24:12] and then happy to announce that Silicon Valley's preeminent law
[24:17] firm, Wilson Sassini, who who are incredible is also using
[24:22] Unity to assess which startup clients to to pursue.
[24:25] And so so it's turned into quite a scalable product.
[24:29] Absolutely. That's quite interesting.
[24:31] And Adam Draper has been on the podcast previously early on in
[24:36] the show. But yeah, that's great.
[24:39] And having seen so much of game film and several startups in
[24:44] your portfolio doing incredibly well, you know, having those
[24:48] credentials of being a good team is one thing and executing on
[24:52] that and actually getting the results is different, right?
[24:55] So how have you seen the best founders in your portfolio play
[24:59] out? What are the particular traits
[25:01] in them that you would like to highlight and that you try to
[25:04] look for when you meet the founders?
[25:05] You know, sourcing is one thing, but then meeting them and then
[25:09] identifying the human traits of them.
[25:11] How do you go about that and what are the things that you
[25:13] spotted in the now successful founders?
[25:16] Well, it's certainly important to spend a lot of time with
[25:18] those founders. You know, I think the
[25:21] traditional process, if you're, if you just don't have enough
[25:24] time with them, it's hard to get diligence, get references, do
[25:28] all the traditional things and do and not skip steps.
[25:31] I think one of our ethos is internally is never skip steps.
[25:37] So there's this grit and elect resilience that's certainly
[25:40] there. I love understanding the way
[25:44] people's minds work and spending time with them is really the
[25:47] only way to do that. I'm obviously going to also stay
[25:51] on message here, but it's very true.
[25:53] Great founders are really good at building strong teams and
[25:57] and, and so obviously we spent a lot of time objectively looking
[26:01] at the depth and the depth data platform, but then also from a
[26:04] diligence perspective, how we do this.
[26:06] And so like for example, Mark Zuckerberg, right, notably a
[26:11] college dropout, obviously not exactly a college burnout,
[26:14] right? He dropped out of Harvard
[26:17] because he has this like start up that's growing like a weed,
[26:20] but but he also didn't have much experience.
[26:25] And so if you're just looking at him in isolation, yeah, it would
[26:29] be hard to pick him to invest in.
[26:32] But you can see even early on, we can see this historically in
[26:36] our data platform, really incredible people were joining
[26:39] that did have work experience in these spaces, did have track
[26:42] records of successes, did have large networks and relevant
[26:46] educations. And as all those people are
[26:48] forming and coming together, you can, you can find a Facebook
[26:51] early, you can find a zoom video early, you can find a Saronic
[26:55] early. And so it's really about the
[26:58] team. It's not a specific founder
[27:00] playing hero, hero ball, doing everything.
[27:03] And I think we, we have a, we have a tendency to lionize
[27:06] founders as these like mythical creatures and like, the reality
[27:09] is, is they're not doing it all. And, and so there's a bunch of
[27:14] information there that we can track to help us figure out
[27:16] which teams are the best. And so anyways, great founders
[27:21] build great teams. Absolutely, 100%.
[27:25] What trends are you particularly excited about?
[27:28] Good question. 2. Significant trends have emerged
[27:32] within the last six months that I think are interesting.
[27:35] One, I think leading VCs are really betting their LP dollars.
[27:40] They're investing in AI entrepreneurs.
[27:42] We've obviously seen that. However, it's this really
[27:46] interesting dissonance between that and VCs in this industry
[27:50] for the most part not really using AI themselves to improve
[27:53] performance across the three major functions of their own
[27:57] industry, sourcing, selecting and servicing the portfolio
[27:59] companies. So The Dirty little secret
[28:01] venture is the majority of these firms, Rashan, are running on
[28:06] Excel sheets and relying on handshakes, vibe checks and
[28:11] physical networks. This does not work at scale in a
[28:15] global ecosystem of entrepreneurs.
[28:16] So, so obviously we've decided to build a new model that
[28:20] circumvents that. And the second piece is as a
[28:22] result, VCs can now invest much more efficiently than the huge
[28:27] firms without requiring these giant teams.
[28:30] So like we're about 10 people, we have no desire to become a
[28:33] 500 person A16Z oligopoly firm. In fact, in the near term, I
[28:39] actually could see solo GPS at early stage be able to be able
[28:46] to manage a billion dollar fund. That's a legitimate possibility
[28:48] with new efficiencies in AI which we've never seen before.
[28:52] And and I think we're going to see a lot of big VCs quietly
[28:56] downsize and I think it's going to, there's going to be a rise
[28:59] in more solo GPS. Absolutely, yes, completely
[29:03] agree. And I think VCs try to back
[29:07] innovation but try to stay away from innovation when it comes to
[29:10] themselves. But I think the time is now.
[29:13] They have to either evolve or might have to, you know, size
[29:17] down their firms or something else, but they will have to
[29:20] evolve. But yeah, you, we have talked
[29:23] about how you use data to select source and select startups.
[29:27] And now let's talk about the servicing aspect.
[29:29] So once you've invested, how do you use data to help those
[29:33] companies get successful in what they're doing?
[29:36] Yeah, there's a few ways. I mean, there's obviously the
[29:37] traditional things that all BCS do.
[29:40] We are unique in that we have, if you think about my pie chart
[29:43] of time, instead of 9095% of my time being spent meeting random
[29:49] companies, I now can spend it actually helping my companies,
[29:51] doing board work, all these things.
[29:53] So I have more time to do the traditional aspects to venture
[29:57] investing, which is positive. But obviously as a product LED
[30:01] firm with products, we also layer in products that we've
[30:04] built for our companies. And so like one example is we
[30:07] have a product that helps identify the right investors for
[30:11] them to pitch, not just the specific firm, but the specific
[30:14] partners in those firms that are much more likely to invest in
[30:17] the, you know, this company's sector stage geography.
[30:20] And it's actually been quite accurate.
[30:21] We've had some really high hit rates on the investor
[30:24] introductions we've made. And for example, last quarter,
[30:27] we've had 7 new financings alone, some of that as a result
[30:31] of this product. Another is, is if you think
[30:33] about our product, as we use an algorithm to identify which
[30:37] companies are more likely to be successful in our database, we
[30:40] can also create new algorithms. And so we've built a go to
[30:44] market platform that built that you leverages a unique ICP ideal
[30:49] customer profile algorithm to identify companies in the and
[30:53] basically the ether that are much more likely to buy from
[30:57] your company based off of all sorts of filters you just can't
[31:01] find in traditional sales tools. And that's because you know, we
[31:03] spent all these millions of dollars across all these data
[31:06] sets building our own proprietary data.
[31:08] And so that's been super successful for many of our
[31:10] companies as well. So, so the key is, is a little
[31:14] bit more time, help them, help them find more capital and then
[31:18] help them find customers. And if you can do those three
[31:20] things while you're outperforming the dramatic
[31:24] majority of VCs. Yeah, that's great.
[31:28] What's been your biggest learning investing in start-ups
[31:30] over the years? That's a good one.
[31:35] So let's see. So so the same way autopilot
[31:40] helps reduce human errors in cars, AI really reduces human
[31:45] biases and investing. So a bias towards a founder like
[31:50] that looks like you or speaks the same mother tongue or went
[31:53] to the same college. You know, that 20th century, I
[31:56] like to call network driven approach to VC when Silicon
[31:59] Valley was a tighter community just doesn't scale for 21st
[32:03] century entrepreneurs. You know, key team team members
[32:07] are just as likely to come from top universities in hot startups
[32:11] in India, the United Emirates, Canada, Estonia and beyond.
[32:16] And so once you embrace this kind of AI approach to
[32:19] investing, you realize quickly how much VC is changing from
[32:23] that 20th century model. It's This is why I think one of
[32:26] our most enthusiastic users of Unity is Tim Draper, right, the
[32:31] third generation VC whose grandfather basically started
[32:35] modern BC investing in Silicon Valley, as well as his son Adam
[32:40] Draper, 4th generation investor. And Tim, I believe realizes that
[32:45] the way he started in this industry with endless meetings
[32:48] and networking and psychological profiling to assess
[32:52] entrepreneurial spirit of a founder is just no longer
[32:56] scalable given the sheer number of founders.
[32:57] So it's much better to amass a giant database of knowledge,
[33:01] including an archive of these nearly 30 million entrepreneurs
[33:05] and early team members that we track.
[33:07] And I personally think that Tim could train our model on the
[33:12] types of deals he likes so well that he could maybe even invest
[33:15] after with that algorithm after he dies.
[33:19] Like we could actually continually have a Tim AI, which
[33:22] would be really fun. So data is all powerful, much
[33:25] more so than an individual's ability to pick winners.
[33:30] And it's it's super interesting that Tim and Adam, who are
[33:33] literally maybe the best to ever do it and precede certainly in D
[33:38] tech, are excited to leverage us as well as a collaboration.
[33:45] Yeah, yeah, I think that's a very interesting point.
[33:48] You mentioned how AI can also have that ability to make people
[33:52] immortal, at least in their idea and their thinking and how they
[33:56] did things right. And that's a very exciting
[33:58] thing. Let's say Steve Jobs trained an
[34:01] AI to how we did things at Apple.
[34:05] And maybe we'd see Apple in a different way to, to what it is
[34:09] today, right? Still doing very well.
[34:12] But maybe that would be so different on how he thought
[34:15] about things. And it would be so handy for Tim
[34:18] Cook to have that AI handy to consult Steve Jobs that OK, what
[34:22] do you think about this? And Steve would say, OK, I think
[34:26] we should do this or that, right?
[34:28] That's quite interesting. I think that's something that's
[34:31] going to happen, right? Exciting.
[34:34] And I think not just on the VC front, surely VC is going to
[34:37] change or has to change how it operates, but also on the start
[34:41] up front, I think a lot of things are going to change how
[34:43] start-ups operate, right? I think execution is going to
[34:46] get a lot faster because of AI. You will be able to do a lot of
[34:51] things just quicker and I trade quicker and make mistakes
[34:55] quicker. And the we see cycles of maybe
[34:58] 10 years might get shortened because of that because teams
[35:02] are just able to execute faster, right.
[35:04] So that idea is also out there. Awesome.
[35:08] With that, I'll I'll move to my last question, which is what's
[35:13] something you wish you learnt earlier in your investing career
[35:17] that you have learnt now? Yeah.
[35:19] Yeah, I always go back to this old Bill Bill Gates quote, which
[35:23] is most people overestimate what they can do in one year and
[35:27] underestimate what they can do in 10 years.
[35:30] And I think that's so true. I've seen that in my journey.
[35:34] Turns out if you stay focused on a mission for a decade, you
[35:37] know, 13 years now for me, what you dream could happen, what you
[35:41] pitch to investors for years and years, what you lost sleep on,
[35:45] what you continue to pursue in the face of so many doubters,
[35:49] eventually that decade does happen.
[35:52] And so I think to start up founders out there, if you think
[35:55] something is true, in my case, it was that VCs should have data
[35:59] informing their decisions, you know, and, and whatever your
[36:04] thing that you think is authentically true, just keep
[36:06] going. You're probably right.
[36:09] Absolutely great advice there. And with that, we'll move to the
[36:12] rapid fire around wherein I'll ask you 6 questions about the
[36:16] investing you're doing and you have to give 6 quick answers.
[36:19] Sounds perfect. Awesome.
[36:21] So the first one is what are the sectors and regions you invest
[36:24] in? All sectors we do avoid FDA,
[36:27] mostly US, but we'll up do up to 15% international.
[36:32] Got it. What's the typical stage of
[36:35] investment? The bell curve wise mostly
[36:37] Series A, but sometimes we do seed and sometimes we do a
[36:40] little bit later stage. Got it.
[36:44] Do you lead rounds? We'll lead, we'll Co lead, we'll
[36:47] syndicate. Not dogmatic about that most of
[36:51] the time. We find the company originated
[36:53] again at first though. Got it.
[36:57] What's the typical check size you put in?
[37:00] It can range anywhere from 2:00 to 7:00 and on our pre seed and
[37:06] seed bets around 200K. Got it.
[37:10] Where can founders get in touch in case there's a possible way
[37:13] to do that? Yeah, for sure.
[37:16] There's no web form as I talked about it, but you can always
[37:18] e-mail me. Colin Collin at ensemble dot VC.
[37:23] Awesome, last one, where can our listeners follow you and what
[37:26] ensemble is up to? Yep.
[37:28] Find us on Twitter, LinkedIn. Those are primary spots at
[37:32] Ensemble VC for both and then my personal is at Colin
[37:36] RWCOLLINRWEST, Come at me. Absolutely.
[37:43] I'll make sure to put all those links in the show notes below.
[37:46] Thanks so much for doing this, Colin, and I was looking forward
[37:49] to talking to you for so long. Finally, we get to do this and I
[37:53] wish you happy investing. Yeah.
[37:55] Thank you so much.

Shows are listed from Apple Podcasts' charts and directory, and Podcast Index where it is used. Every show and episode belongs to its publisher and plays from the publisher's own feed; this site hosts none of them.