You have to think long term.
So we think in five to seven year terms and try not to worry if we're off by a year or two on the valuation.
Like that's a risk that I'm willing to take.
We turned it down based on price when I was at GA.
They figured out a huge market, hiding in plain sight a sales model that totally worked fast, product velocity, all the things that we look for big markets where they're the leader business model that was exceptional in scale, long room to run.
So that's a very painful one.
The benefit of single trigger puller model as opposed to committee decision making is it's the ultimate measure of conviction.
So if that individual has conviction and gets feedback from the partnership and maybe the feedback is constructive or negative and still wants to make the investment, that's conviction.
Today we're replaying a conversation from 20VC with Harry Stebbings, featuring A16Z general partner David George from our growth team.
David shares how he thinks about breakout growth investing, why edge comes from non-consensus views on market size, how to underwrite upside in competitive markets and what separates pull companies from push companies.
He also dives into unit economics.
Deciding when to double down and how single trigger decision-making shapes investment conviction.
They round out the conversation with SPACs the rise of crossover investors and how David manages pressure and competition over the long arc of an investing career.
Welcome back to 20VC with me, Harry Stebbings.
And what an episode we have in store for you today.
I just love doing this one.
It's one where the chat really completely went off piste and we didn't stick to the schedule at all, but always a sign of a great conversation.
So I'm thrilled to welcome David George, general partner at Andreessen Horrors, where he leads our growth investing practice.
Since joining in 2019, David has invested in the likes of Clubhouse Coinbase Databricks Figma Instacart, Robinhood and TripActions, to name a few.
David also sits on the board of Current, Greenlight, and WorkRise.
And prior to Andreessen, David spent seven years growth investing at General Atlantic, where he invested in the likes of Airbnb CrowdStrike, Opendoor Slack and Uber again just naming a you.
I'd also want to say a huge thank you to ariel at trip actions and ali at databricks, in particular, some amazing questions, suggestions from them, and i so appreciated that.
But without further ado, i'm now so excited to hand over to david george, general partner at andreessen horowitz.
You have now arrived at your destination, David.
It's such a joy to have you on the show today, my friend.
I do want to start by saying a huge thank you to Angela Strange on your team and Ali at Databricks for some brilliant question suggestions.
But thank you so much for joining me today, David.
Hey, thanks for having me, Harry.
I've listened to the pod a bunch and love what you've done with it.
Well, thank you so much.
I still can't believe I actually get paid to talk to people like you.
So it's a pretty crazy life.
But I do want to start with some context.
So tell me, how did you make your way into the world of venture and come to be leading Andreessen's growth fund today?
Yeah.
So look, I grew up in Kentucky, very far away from the world of finance and technology.
I was very fortunate.
I had an awesome upbringing.
My parents I had an older brother went to college at Notre Dame thinking I was going to be a lawyer.
Thankfully, I went into the finance industry out of school instead.
Eventually, I moved to San Francisco in 2008, where I started to first encounter the world of tech.
And I joined General Atlantic about 10 years ago.
At GA, I had the chance to invest in some amazing companies, spent about seven years there.
Companies like CrowdStrike Uber Airbnb Opendoor, Slack and then some others that are a little bit lesser known but really awesome.
Companies like Benevity and Seismic.
I came over to A16Z to start and run our new growth fund about two and a half years ago.
Initially got to know some of the folks at the firm just from some overlapping investments that we had.
So I think I knew Alex Rampell the best because of an open door.
And fast forward to now we're investing, had the chance to build a great team inside the firm and work with some awesome founders and companies here as well.
I mean, it's been an incredible time since joining Andre.
So do you have to touch on the GA time there?
Because GA is such a prominent kind of figure in the industry, but it's also a very different one.
So I have to ask, you know, what were your biggest takeaways from your time with GA?
And how do you think it impacted your mentality?
Yeah, look, GA is an amazing place.
I'm really grateful to the team there.
You know, one of the foundational frameworks that I developed coming out of it was what makes a breakout or successful growth investment or sort of pattern recognition of what I look for.
In almost all cases for me that has come down to, you know, a great founder, and then a theory on busting through consensus view on total addressable market, or TAM.
The consequence of that is a company that grows faster and or longer than expected.
So take my time here.
Take Roblox the prevailing view when we invested was hey, it's just a kid's game.
Well, we felt that they had a shot, and still do, at being something much bigger a co-experienced platform, something that's much broader than games, much bigger than kids.
One counterintuitive observation from my time at GA and my pattern recognition coming out of it is exceptional business models are in growth investing.
They're not actually, in my experience, what gives you edge in making great growth investments.
So we don't take risk on investing in anything, but In my opinion, this is not where you generate outsized returns.
You can make mistakes here, but it rarely surprises to the upside.
So despite that, I feel like it's where 90% of growth investors spend 90% of their time.
So, coming back to the TAM point total addressable market point I can give you a couple more examples on forming non-consensus views and things that made successful investments for me.
One of the flavors of this is the consensus view of total addressable market lags.
What's actually happening in the market?
You know, when I was a GA, the consensus defined the market wrong at AppDynamics.
There would be way more applications than historically.
And, you know, the view of the software to support that just lagged.
So you could see it actually by looking at individual forward-thinking companies, but not from looking at market research reports.
One of the more recent examples is Figma, which we're investors in from A16Z.
The simplistic way to look at their market is software for designers.
But the historical definition of design and designers is actually pretty dated.
Our view was that all front end engineers in the future will engage in design.
And this is 10x plus bigger in terms of market opportunity than just defining what they do in the design space.
So you know, in both cases it was a refined view of the addressable market that led to a conclusion that in the future that that will be much bigger than folks recognize today.
I have to unpack a couple of elements here before we move on.
I just want to dive in and touch base here.
You mentioned the element on business model and I do want to touch on that because when you look at some companies say a DoorDash of the world.
It takes a while for the beauty of the business model to mature into what it is.
So how do you think about having the mental plasticity to see what the business model can be versus what it is today when investing in growth?
Yeah, look, DoorDash is a great example of this.
You just led me right into a discussion of mistakes that I've made in the past, which could be like a 90-minute VC episode.
But DoorDash is a great example of this.
I missed a lot in DoorDash, not just the unit economics, the power of the market size, the localized network effects.
But I would say the big piece on the unit economics for DoorDash is you could see early signs of it in performance of really mature markets that they were in.
And you could see the localized network effects.
So specifically in the South Bay suburbs in the Bay Area unit economics were very good.
And they ran some experiments at the time that we were looking that demonstrated that they could actually get the unit economics much higher.
The power of their market position allowed them to maintain strong unit economics on the restaurant side.
This but it was early to see it.
On building consumer stickiness, like loyalty programming,
So you put Even though, you know, the trailing data around it didn't show it.
You mentioned unit economics quite a lot there.
My challenge is like, when do they become important?
I'm using this session as an advice and a learning moment for me.
But at the early stage.
Sure, your CACs may be super low, but you've got the most aligned customer base that you're marketing towards.
And they will generally get a lot more expensive over time as you saturate that audience.
But then also brand and word of mouth can come into play and they can reduce significantly.
So I guess my question is like when do unit economics become central, and how central are they to you?
Yeah, look, at the growth stage, they're very important.
The thing that we look for in unit economics is, you know, where are they now?
And then, as you scale up much larger, do you have a theory on why they're not going to get worse and hopefully that they get better?
Again though, I go back to the point of, like what's driven, you know, outsized returns.
For me, the unit economics end up being sort of table stakes, right.
Like you can get it wrong to the downside.
It's very rare that a company ends up with, you know, unit economics that are much greater than what you expected.
It's much more common that, hey, the growth of the company just exceeded all expectations.
So, you know, at growth stage, when we invest, we look very closely at them.
The challenge that I have and sorry, we did have a schedule.
I'm just completely, you know, going off schedule.
The challenge is like with the unit econ, is like when you have the proliferation of capital that we have into the markets and suddenly founders have these budgets which are just eye-watering.
Salona has raised a billion I mean a billion dollar raise, like their need to be worried about CAC optimization or unit economic efficiency is just less because they've got a billion dollars in some cases.
How do you think about and advise founders on how close they should pay attention to unit econ when they have this proliferation and capital support?
Yeah, look, I think it's it's a great question.
And it's it's a function of market conditions, competitive environment.
So you know, there are instances where I'll tell founders that it would make sense to relax their criteria and maybe spend a little bit more in the name of growth if it's a hyper competitive market and it ends up being a super sticky customer base over time.
But you mentioned Salonis and the company a little bit.
They're in an incredible market position.
I think their business model is fantastic.
I would view their future as very elevated levels of high growth.
And there's not some looming intense competitive threat that makes them think that they should aggressively go burn a bunch of cash just to grow faster.
In our universe, in each of those sectors, that makes it sort of the right size for us.
I think one for me that I always find striking is that capital concentration on a per company basis.
How do you think about, bluntly, how to get as much cash into the winners as possible?
And what does that reinvestment decision making look like?
For like, do we really fucking double down?
Or do we let the capital market support it and we play a nice role?
We've invested multiple times in many of our companies.
So Coinbase, we invested three times, Roblox twice, Databricks three times, Stripe four times, TripActions four times.
Every time we assess one of those new investments, we do it with fresh eyes.
And so we call it re-underwriting.
We mentioned the upside there and the multiple expectation or multiple desire.
You know, the challenges are so much cash.
The prices are so high.
It's just much harder to do those multiples with the entry prices that we're paying.
I had Leila from Capital G on the show.
She said the prices are 2x what they were a couple of years ago on average entry price for her.
I'm interested like how do you think about your own price sensitivity today?
Given where we are today, multiples don't matter.
Talking a little bit, it touched on this and other things.
I'll just talk to you about our process and then I can address the valuation points.
So we first start, we assess the company, we assess the market, we assess founder, all independent evaluation.
And so if those check out, then we spend a lot of time on valuation and scenarios and making sure that we see our way to target returns.
So the best thing that we can do is invest in great companies that are growing very fast, because those afford you more degrees of freedom on valuation and you know i talked about the upside scenarios like they're the ones that are more likely to deliver upside scenarios.
So One of the frameworks that we use and talk about a lot and this is relevant for evaluation because it speaks to the flavor of companies that we tend to matchmake with is we look for what we call Glen Gary, Glen Ross market structures.
So the famous movie, this is like independent evaluation.
So what that means is there's a scenario.
Have you seen the movie?
I haven't.
So explain it for me.
Okay.
All right.
It's sort of like a boiler room sales old school movie.
So there's a scene where Alec Baldwin is presenting to his team their monthly sales competition.
It's his famous line where he says, okay, here's the prizes.
First place gets a Cadillac.
We actually think most, many or most tech markets play out in market cap in a similar way, where the leader captures the vast majority of the market cap creation.
You know, if you're not the leader, it's going to be a challenge situation.
So we look for those kinds of market structures.
You know this is very well known and well covered in.
You know consumer land companies like Google and Facebook that have clear network effects.
But surprisingly you can see it actually in a lot of industries that don't have network effects but play out in similar ways.
So in b2b, you know salesforce, workday service.
Now they command almost all the market cap in their respective markets.
So the way we approach the valuation question is you know, if we can get that point right and the company wants to work with us, you know, more often than not we can reach an agreement on valuation.
Now, you know, the biggest point is you know, in a market where valuations are higher than they used to be, you have to think long term.
So we think in five to seven year terms And try not to worry if we're off by a year or two on the valuation.
That's a risk that I'm willing to take.
If we underwrite something five years and it takes us seven years, I'm okay with that.
The last piece is just tech.
Markets are bigger than ever and there's going to be a lot of market cap creation.
And so if we're long-term oriented enough, we should be okay.
Tech's about a quarter of US market cap and that's just going to grow fast.
Can I ask one?
There's always a case where everyone's turned down a company based on price.
What company have you turned down based on price that keeps you up at night?
Oh my goodness.
The most painful one is probably Qualtrics, which, I mean, look, it's killer founders.
And yeah, I remain friends with the guys.
We turned it down based on price when I was at GA.
They figured out a huge market, hiding in plain sight a sales model that totally worked fast, product velocity, all the things that we look for big markets where they're the leader business model that was exceptional in scale, long room to run.
So that's a very painful one.
I do love Ryan.
Just never played golf with him.
He's an absolute fiend.
I do have to ask you know.
You said there about kind of entry price no longer mattering, with that brilliantly kind of provocative title.
I was taught, you know, I'm from the old school of venture, temporal diversification mattered too.
And that was a core part of any portfolio.
Today with deployment cycles, it doesn't seem to matter either.
Do you think temporal diversification matters today?
Or do you think it's about adjusting to the game on the field and being in the moment?
Look, I think it's more about adjusting to the game on the field and being in the moment.
And then we don't invest in hundreds of companies.
And so our goal is not to be an index fund of the overall tech market.
We just need to continue to outperform and do our best.
And if we can see the path and get confident about the path to achieving our target returns, that's okay.
Diversification
If you have more of your funding reserves, you know that stretches the deployment cycle of your fund over five years instead of two years or whatever the number is.
That naturally provides some temporal diversification.
I have to ask, man.
We mentioned the price changes and the increasing price we've seen over the last especially year, but last few years.
You know, we've seen P hedge funds, crossover funds.
I mean, everyone and their mother investing in late stage and pre IPO.
I've got to ask, man, when you look at it.
How do you think about this massive proliferation of capital at the late stage, especially with players who seem to be playing a different game in terms of return expectations, willingness to pay 2x what we pay?
How do you think about this new landscape and how you win in it?
Yeah, look, it's a great question.
First, I think there's a little bit of a misperception in the market about some of those firms who have been more aggressive recently.
There done to adjust.
So one, diligence processes and fundraisers, you know, they've gotten faster to put it lightly.
What that causes is it makes investors narrow the focus area of what they can cover in diligence.
You have to have a prepared mind coming into things and you have to be really smart about where you spend your time.
So we always focus on the three or four things.
You know, I talked about business model being sort of a table stakes exercise for us.
You know, so that's typically, you know, one, maybe two of those things.
But, you know, the real decision making process will come around that view of future growth.
So we have to be prepared to make those judgments faster and assess those questions quicker.
I remember when I started at GA it would be like okay, a company's raising money and let's take two months and you turn over every single piece of minutia for a company.
It's probably more efficient and better overall for the market.
That that's not the way it's done anymore, because it's just inefficient all around.
On the competitive point, I think of us as playing the end game relative to some of the newer players.
So we move fast.
We pay fair market prices.
We can write very large checks and follow on into companies many times over the years.
That's all what some of the newer players are known for.
But because of the way that Mark and Ben have built the firm, we can also do a lot, And that is relevant whether the company is at series A or the series D in the growth phase.
So you know, we tend to matchmake with the companies who see us as being able to deliver more than just dollars.
You mentioned Mark and Ben there.
In terms of like investment decision making, how does that look like for you?
Yeah, it's a little more on the informal side.
We have a single trigger puller model, though.
We have discussions, and deep, robust discussions about investment decisions, but it's ultimately the call of the GP who's sponsoring the investment.
I serve as a generalized specialist.
I heard Paul Enright use this term recently.
It works for us.
And what that means in this case is I sit across all the industry groups.
So you know we can measure the relative excitement of a B2B software company versus a bio company or a crypto opportunity or consumer company.
And this is really helpful in sort of measuring relative excitement and attractiveness.
So the other piece on investment decision making and it goes back to the matchmaking point is you know, we try and prioritize companies where there's still some degree of company building that takes place.
So we can help companies given what Mark and Ben have built.
And a big part of our decision making process is we have the luxury of partnering with our early stage GPs and deal partners across the firm.
So we feel like we'll have unique insights into big trends or themes early from being a part of the early stage business.
You mentioned the single trigger element.
That I'm really interested because you know, at the end of the day, you've just got to have the courage to pull the trigger and go.
And it's a big, courageous moment to write that check and put your name to it.
I guess.
How did you approach that?
How did you mentally get over that with your first?
And would you have any advice?
I remember with mine, I was shitting myself.
And mine was for a million dollars.
Well, it's better to be honest, I mean.
Yeah, look, it's better to be honest, man.
Look, I totally agree.
And like the first time I did that, I still remember I mean it's not.
It was not a single trigger puller model at GA, but I remember my first investment there and I was panicked.
Like, am I making the right decision here?
I don't know.
I've grown to get comfortable with it over time.
I think the benefit of the single trigger and this may be obvious, but the benefit of single trigger puller model as opposed to committee decision-making is it's the ultimate individual has conviction and gets feedback from the partnership and maybe the feedback is constructive or negative and still wants to make the investment.
That's conviction, right.
And if there's a committee model sometimes, you know, one of the negative things that happens is they're selling.
And so because often a committee will look to the person and try and measure or gauge conviction and And so they're selling.
And so my experience.
I think it leads to less intellectually honest conversations and open conversations because you know there's some element of convincing that needs to happen, as opposed to personal conviction building by challenging and, you know, asking questions and exposing concerns and getting feedback on those.
This is a tough one, but do you think there is internal politics?
I mean, there is at most firms, but do you think there is internal politics in Andreessen?
Do you think there's anything that you or one can do to maybe prevent or minimize them?
It's going to sound like I'm bullshitting because it's like the politics are pretty.
Look if you're a single trigger puller model.
Lots of the politics come from shit like that.
Trying to convince people, trying to advocate your own ideas, trying to shit on other people's ideas because you know maybe there's a deal that gets done instead of you.
Why would that introduce politics if it's single trigger?
Because it's like, I like this deal and I have conviction.
So see you later.
Rather than like, I have to get you on side.
I'll vote for your deal if you vote for my deal.
Yeah, no, that's what I'm saying.
I'm saying the same thing.
I'm saying the single trigger for the model eliminates the politics right right, yeah.
And then the other thing that causes politics internally is like promotions path for people, and i think part of the benefit of the way that just the background of most of my partners being founders of companies is they've actually run companies like this is a typically a problem with investment firms.
It's like superstar investors end up running the firm whereas you know we, our firm, is run by people who have run companies.
Yeah, I do totally agree with you and see that one of many reasons I work alone well that, and you know, no one likes me, but other than that, it's totally cool.
Everybody loves you.
Man, I was fishing, total fishing, I stepped right into it anyway.
Now I do want to ask as well, though final couple, but it's like you know, everyone's got a SPAC now.
I've been tempted by a 20 VC SPAC.
I think that's so cool.
But anyway, everyone's got a SPAC.
How do you think about the rise of SPACs, the SPAC market, whether it's good, bad, an opportunity?
Look, I think SPACs are great for the company side.
They provide another form of liquidity and getting public for companies.
I counsel, I've talked to some of my company's founders about going down the SPAC route.
And the thing that I say to them always is it's a totally fine path to go down, as long as it's not seen as a milestone and you're just ready to be a public company after you're done with it.
So it's good from the company side.
I think on the issuer side it's valuable to the market if there's some uniqueness that comes along with it.
So if there's a value proposition that's unique, great.
It will be appealing to the right founders.
If it's just a financial vehicle, I think the proliferation of so many SPACs will make it a little more challenging to generate really attractive returns in that market.
Final one, you mentioned some of your companies.
I spoke to some of your companies before the show and one, Ali, who I love, by the way.
But Ali said to me, you got to ask this guy.
He's been super successful and he's achieved so much.
What drives you today?
And how do you think about your relationship with money?
On a plane last night and I was texting with Ali about something similar
Money is not what drives me.
I love learning new things every day.
I get the chance to work with the best founders in the world, building awesome companies just like you.
I love my job. paranoia of failure and I'm extremely competitive.
And so returns and generating returns are one element of the scoreboard for me.
And I love that competition and I want to be the best.
There's a mission side of our business too, which is generating great returns for our limited partners.
So these universities, nonprofits, it's the same as you.
They've entrusted us with their investments and I want to make sure I work as hard as I can to give them the best outcomes in the market so that they can use those dollars to good cause.
How do you prevent that fear paralyzing you?
Because I too have this kind of relationship with fear.
The way I channel my fear, I channel it into working harder.
So if I have a fear at any moment that I'm not doing the right things, the way I try and compensate for that is I dive in, I work harder, I do something different, I reach in.
And the beauty of our business, same as yours, is there's an endless amount of work that you can do.
You can always get smarter about a company.
You can always get smarter about a trend or a theme.
You can always try and form a differentiated point of view on something.
You can reach out to people who you've learned from.
There's just an endless way that you can, you know, expend work time, and so when I feel like I'm not doing well or failing, I tend to, you know, go deeper into that stuff.
I do want to move into my favorite though David, which is a quick fire round.
So I say a short statement and then you give me your immediate thoughts.
Does that sound okay?
Let's do it.
Okay, so what's the favorite book and why, David?
Okay, I'm going to give you two.
One is like a business book that I think is the most important one, which is for the business that I'm in, which is Increasing Returns to Scale, by Brian Arthur.
The second one is my fun favorite book is Count of Monte Cristo.
Tell me, have you read Seven Powers by Hamilton Helmer?
I have not, but it is downloaded on my Kindle.
Honestly, what we do, it is amazing.
Tell me, what lie do rich people tell themselves most often?
I think it's overemphasizing their own work and underplaying the role of luck and some circumstances and other people who contribute to their success.
It's like form of like just world fallacy.
What is the single biggest challenge of your role with Andreessen today?
It's constantly evolving to changes in the competitive market.
So I mentioned this earlier.
Mark and Ben turned the industry on its head 12 years ago and people are constantly trying to figure out the next thing to turn on its head.
So our strategy and how we work with companies always has to change with that.
What do you know now that you wish you'd known when you started at Andreessen a couple of years ago?
I mean, Andrews is still pretty fresh for me.
I wish I had known earlier in my career and growth to spend more time thinking about what can go right as opposed to modeling or trying to predict what may go wrong.
This is a tough one.
What advice do you often give that you find hard to follow yourself?
It goes a little bit back to our fear point, but there are going to be ups and downs over the life of an investment in a partnership with a founder.
Don't get too high from the highs.
Don't get too low from the lows.
It's one thing on the investing side.
The same could be said for building a company, but probably times 100 in magnitude.
Final one, David, what's the most recent publicly announced investment?
And why did you say yes and get so excited?
So it was Loom.
So the frameworks that I love is I try and invest in companies that are pull companies, not push companies.
What that means is the market is pulling their product from them, as opposed to they're trying to push their product out to the market.
So they're an asynchronous video company.
It's viral.
It's spreading organically.
It's growing 10x year over year at scale.
They're building enterprise top-down sales onto bottom-up traction.
And Run by founders who are passionate.
They're domain experts, they know the product, they're building a brand and they're building a great business.
I've got to ask this one final question.
With pull and push businesses, often it's pull in the beginning and then it goes to push.
Do you get worried when it goes to push and how do you determine how long you have to run on the pull?
Yeah, this goes into the market work that we do.
I think you can form a pretty sophisticated point of view on this.
One.
If it's something that's very unique and they are the market leader, the pull will probably last much longer.
Secondly, it's part of our diligence.
We talk to non-customers probably more than we talk to customers of companies, and you can get a pretty good sense for how they're going to behave in the future based on those conversations, to try and predict it.
And then back to my table stakes business model comment.
You know, you just got to make sure that, as pushing that comes along with the pulling that the economics of that are going to make sense.
David, listen, I absolutely love this discussion.
Thank you so much for putting up with my completely wayward questions, but it was amazing and I so appreciate it.
Thanks for having me, man.
I mean, if you couldn't tell, I absolutely love that discussion.
I want to say a huge thank you to David for being so patient with my completely off-schedule questions.
He was just fantastic there, and as I said, just loved it.
Shows like that make me really appreciate what I do.
As always, I so so appreciate your support, and I can't wait to bring you a fantastic episode this coming Thursday.
And subscribe to our Substack at a16z.substack.com.
Thanks again for listening, and I'll see you in the next episode.
Thank you so much for having me.
