Benjamin J. Shestakofsky on How Venture Capital Shapes Work, Innovation, and Inequality
01:12:53 Share

I've got a lot of grandbabies. Like, a lot, a lot. And when it comes to finding a gift for each of them, you know, it could add up. But this year, while I was making my way through Walmart, I realized I don't have to spend a lot to get the gifts they'll love. An OPI Mini Manny set! I'm gonna do so much nail arts. Oh yeah! A Lego set! My own a wall of water bottle! Ooh, and that's just half of them. Shop great gifts they're sure to love for $25 and under at Walmart.

This episode is brought to you by AWS. Amazon Q Business is the generative AI assistant that can securely understand your business data, summarize results, and streamline tasks. Learn what Amazon Q Business can do for you at aws.com slash learn more.

This episode is brought to you by Indeed. We're driven by the search for better, but when it comes to hiring, the best way to search for a candidate isn't to search at all. Don't search, match with Indeed. Use Indeed for scheduling, screening, and messaging so you can connect with candidates faster. Listeners of this show will get a $75 sponsored job credit to get your jobs more visibility at indeed.com slash MBO. Terms and conditions apply.

Welcome to the New Books Network. Welcome to Peoples and Things, where we explore human life with technology. I'm Lee Vinsel. Hey folks. Hey brothers and sisters of the P&T. How's it going? This episode is one of those I see as deeply intertwined with my long-running interest in methods.

The kinds of work we do to produce strong humanistic and social scientific knowledge. The ways in which we collect data in order to make reasonable inferences and generalizations, to make inferences and generalizations that are responsible.

Now, one way to set up this episode in particular is to say, okay, in some sectors of the U.S. economy, particularly around digital technology and other high-tech enterprises, venture capital is important. I mean, on one level, it's incredible how much attention venture capital gets from academics, given that VC funds less than 1%, maybe something closer to 0.5% of all new enterprises founded in the United States.

But that doesn't mean it's not important in some spaces. It's certainly worth studying deeply. And venture capital, like some other institutions today, is obsessed with what is called scaling or quickly growing organizations or technologies in ways where growing revenue outpaces growing costs. But how does venture capital and its ideology of scaling actually affect things on the ground in organizations?

Well, I think the guest in this episode, Ben Shestakovsky, assistant professor of sociology at the University of Pennsylvania, has written absolutely the best book we have on this topic. It's called Behind the Startup, How Venture Capital Shapes Work, Innovation and Inequality. For me, it's a great book.

The most incredible thing about this book is the deep access Ben got to the organization he writes about. Because as you'll hear in our conversation, he actually ended up working there as an employee.

But because Ben did great data gathering during his time there, both as a sociologist and as an employee, that access translates into a deep empirical understanding and picture of how venture capital shaped changes and experiences in this startup.

And it's for that reason, I would argue, that Ben gives us the best treatment so far of how venture capital and scaling affect both leaders and workers in organizations. God, I love good work when I see it. And Ben Shestakovsky's Behind the Startup is one of my favorites in recent years. And that's why I can say to you in all honesty, hey, get excited. ♪

Ben, thanks so much for taking the time to talk to me today. Thanks for the invitation, Lee. Behind the Startup is a neat book. If you're talking to strangers about it, maybe non-academic strangers, if that is everything that happens in your life, what do you tell them it's about? What were you trying to do with it? I think I would probably just start kind of with how I set up the book, which is that, you know, there's a lot of talk these days about

sort of the bad things that the tech industry has done. We went from, you know, a time of utopian thinking to a time of paranoid thinking. And, you know, and that we have a lot of stories about what's going on here. Like, why does innovation seem to create social problems? And some of the stories are about these founders who are, you know, kind of greedy and or just totally nutty.

Yeah. You know, and those are... Altman, Musk, Thiel, all these guys, right? Totally. Yeah. I mean, even Zuckerberg back in the day. Even Zuckerberg. People used to love to talk about this guy. Totally. And those are really entertaining stories. But, you know, to me, I think they only sort of scratch the surface. And then we also have all these stories about the technology itself, right? There's literally...

But in terms of both journalistic output and academic output, there's just thousands and thousands of articles and or books about, you know, AI, machine learning, big data, metrics, you know, this sort of sharing economy. Yeah, all of it sort of like this new wave of technologies. And, you know, often people are locating all sorts of social problems in the technologies themselves.

But what I did for this book is I spent 19 months doing participant observation research in a tech startup, an early stage startup, which basically just means I was sort of studying this company as I was working alongside the people who inhabited it. And what really stood out to me was not, you know, totally normal people who ran the company.

Not some sort of shady, you know, behind the scenes algorithms that were wreaking havoc on people's lives. But what really stood out to me just watching how things work day to day was how the venture capital funding model really strongly influenced everything that happened inside this company. And the need to...

to seek funding, to get more funding just to keep this company alive, put tremendous pressure on the managers of this firm and basically created a ton of organizational problems that they had to solve. And what I found was that they tended to solve these problems by combining

their technologies with low wage human labor. So I write about this team of about 200 people in the Philippines that did sort of, uh, very routinized information processing work to make the algorithms do what they were supposed to do or to help kind of push them along or, or help, help people innovate faster, help the engineers innovate faster. And then this team of 10, uh,

customer support agents working from home in the Las Vegas area and sort of their role in keeping users of this company's product engaged in spite of the fact that it was often changing and sometimes changing really for the worse for users because

venture capital investors might say, we want more revenue from you, which means you got to charge your users more money, which people actually often don't like. So the book is sort of ultimately about

This funding model shapes everything that happens inside of startups and really how it also reproduces the inequalities that the tech industry is known for, where the lion's share of the rewards that venture-backed companies produce are funneled just to a very tiny, tiny sliver of stakeholders at the very top. And of course, so much of the risk that is involved in

starting a company like this gets pushed down towards the people at the bottom of the chain. That's sort of in a nutshell what the book is about.

Did you, so did you think you were going to write about VC initially when you started doing this research or did that really kind of come out of, you know, like being there? Yeah, certainly not. I thought, you know, when I went in this, this field work is, um, in some ways I think nothing has changed in the industry. And in some ways I think, uh, you know, this field work was way back from 2012 and 2013. Um, way back.

Well, yeah, this was a different bubble, you know, this was before the gen AI bubble. This was a platform bubble. Yeah. You got it. Um, so, uh, but this was also before people were so focused on, you know, the purported harms coming from the tech lash had not come to be exactly, exactly. Um, and so, uh,

You know, honestly, at this time, my mind was not on financing at all. My mind was on something that people were talking about back then, which was like, oh, what is this new startup workplace? You remember like the Google model that everyone was trying to emulate at the time? And to some extent, everyone's still emulating, but...

Apparently, it's been scaled back in a lot of places, but, you know, it's like all these free meals, you know, all the perks, the ways that what really fascinated me was like, you've got all these young people. I mean, really just mostly guys right in their 20s. And they are.

sort of working, but also this is sort of a social scene for them. Not only, not only as work kind of a place to both work and play, but then all of the other startups around them, just geographically, you know, on their block in the South, the market neighborhood in San Francisco, um,

You know, there's just there's parties going on all the time. People are lunching. People are drinking. You know, I was really interested before I got there and sort of this social scene and sort of how startups blur work life boundaries to create sort of an environment of overwork, you know, where.

It's not necessarily seen or experienced as overworked because people kind of feel like they're having fun and they're socializing and things like that.

Now, when I got there, I pretty quickly realized they had just raised the first round of VC funding. They're trying to plow that money into hiring software engineers mostly. And the engineering team was just basically spending all their time on recruiting and interviewing and, you know, discussing job candidates. And there was very little like

going on at all. And so I came in expecting to find overwork. I'm not really seeing it there. But what I am seeing is that as all this hiring is going on in San Francisco, there's actually tons of people working behind the scenes offsite from their homes in the Philippines and Las Vegas to kind of keep this behemoth moving and keep

keep the user base growing and keep the platform functioning and keep users engaged. And so I sort of started to shift my attention away from, you know, this idea of overwork and how this kind of social scene and workplace culture are implicated towards really just trying to grapple with, like, what's everyone doing here? All of this, there really was

Hardly any at the time. There was probably a little in maybe like the HCI literature about some of this behind the scenes, you know, it was then often called click work, you know, Mechanical Turk was where a lot of this work was being sourced. But there was very little research on it at the time or relatively little.

And I was sort of just trying to wrap my head around what is going on here? What are all these people doing? Why are they doing what they're doing when they're doing it? And why isn't this just what I thought a tech company would be, which is a bunch of engineers sitting around making grinding and eating pizza? Exactly. Exactly. Yeah.

And so, and then as you started to see, you know, think about what you were seeing, then the incentive structures, the kind of the financial ecosystem started to become clearer to you? Yeah, I think first, first, what I started writing about, and this was sort of the first major article to come out of the project, and also my dissertation, was

at that point I was really just focused on the relationship between work and technology and looking at, uh, again, like sort of new stuff at that time, algorithms. Oh yeah. Oh yeah. You know, like, like what's going on here with these algorithms and how and why is it that these seemingly automated systems actually rely on these very, uh,

and distributed workforces in order to function. And you, I'm sure, remember that this was also a time when, I mean, it hasn't really gone away, but this...

literature and debates on the future of work was really booming. Oh, yeah. Right? It was this idea that like, okay, well, now that machine learning is here, and now that the algorithms are here, are they going to take all our jobs? Yeah. The second machine age and all that stuff was rise of the robots, all that kind of stuff. Exactly. Exactly. And so I didn't really...

find a lot of these arguments particularly credible. And I wanted to sort of use my empirical material to try and, you know, make an argument about, well, at least in contexts like this, contexts that are changing very quickly, because they're startups, right? They're sort of

They don't have an established product, they don't really have an established market and they just have to constantly pivot and adapt to try to figure out like, what are we doing and who might pay us for it.

Monday.com is the one centralized platform for everything work-related. And with Monday.com, work is just easier. Monday.com, for whatever you run.

Go to monday.com to learn more. Hey, I'm Ryan Reynolds. Recently, I asked Mint Mobile's legal team if big wireless companies are allowed to raise prices due to inflation. They said yes. And then when I asked if raising prices technically violates those onerous two-year contracts, they said, what the f*** are you talking about, you insane Hollywood a**hole?

So to recap, we're cutting the price of Mint Unlimited from $30 a month to just $15 a month. Give it a try at mintmobile.com slash switch. $45 upfront for three months plus taxes and fees. Promote it for new customers for a limited time. Unlimited more than 40 gigabytes per month. Slows. Full terms at mintmobile.com.

So, you know, in that kind of setting, I was trying to theorize, like, this is actually not the kind of setting where you're going to see algorithms just automate away everything. In fact, for algorithms to work for organizations like this to accomplish their goals, it actually makes a lot more sense for them to combine, you know, what algorithms are good at doing with what people are good at doing. And, yeah.

That was sort of the water that I was swimming in, you know, as I kind of bit off the first chunk of this project was sort of this literature on the future of work and sort of, I was positioning myself in the camp of people who are trying to explain, you know, when and why jobs were actually not always going to disappear and why

We might expect new technologies very often to, you know, generate new kinds of work in and around the technologies. And so then the question becomes less, is the work going to disappear? And more, what is the quality of this work going to be? How will it be compensated? How can it be improved? Yep. So that's sort of where I was at at first. And then a couple things happened that sort of shifted my attention to,

the investment side. One was a conversation, sort of my last conversation about the dissertation with my dissertation supervisor, the great Michael Buravoy. And he sort of just mentioned offhand, you know, it strikes me that all of these changes in the organization that you've observed, each of which sort of

led to a new sort of restructuring of the relationship between work and technology in the firm.

He said, "It seems to me like these changes might have actually been related to this company's quest for venture capital." I said, "Huh, that's actually interesting." He's like, "Maybe venture capital isn't sort of just like the background of the story. Maybe it's actually part of the foreground, like a really important part of the foreground." So that one kind of stuck in my head. And then a year later, Mary Gray and Siddharth Suri

published ghost work, which is, you know, a tour de force about sort of the same topic that I had been writing about, about the often hidden labor that lies behind algorithmic systems. And I sort of just had this feeling when the book came out, which I loved, and, you know, I reviewed it immediately, and I loved it immediately. It's fantastic in a lot of ways. But I sort of

just had this sense almost immediately, like, I can't write the book that I thought I was going to write anymore. Like, I think my book just got written. And so then what Michael had said to me sort of burrowed deeper into my brain. And I started to think more deeply about that. And once I started writing it, it was just so clear that he was right. And that, you know, that this really was a story that,

about startups and about how they're funded that I think just hadn't been told before. Obviously, it's not that no one ever talks about venture capital and what it does to startups, but what I haven't seen is another treatment of startups that is deep, ethnographic, and really looks at day-to-day

how organizational activity is structured around this impetus to really to move fast and break things, which comes from the demands of venture capital.

Yeah. Well, I have to say, I'm glad we have the version we do because I think it's really good. And like you say, I think it's unique where I think that the point you were working on, I'm sure it felt and was fresh at the time, but at this point, it's almost become like a mantra that wherever we see algorithm and artificial intelligence or whatever, there's always human labor behind that. But the picture you paint here, it really does feel new and unique.

Congrats on the New York Times review. How did you find that? You got reviewed with two other books on corporations and corporate stuff. So did you just bump into it? Did someone send it to you? How did you find out? Yeah, actually, I mean, this was really...

I think really a beautiful moment for me, which is that actually my dissertation supervisor, Michael Burboy, sent me an email and said, hey, congratulations. I saw the book review in the New York Times. And I said, what book review in the New York Times? So, but yeah, it was really sweet. It was kind of a full circle moment. Yeah. Certainly not something I was ever expecting when I was working on this project. But, you know, it sort of went from,

in my mind, at least like a book for an academic audience. That's what it was at the beginning. And then I was working with my editor, Naomi Schneider at University of California Press. And she really thought that this could be sort of more of a crossover academic book and, you know, planted that seed and it ended up moving in that direction. And ultimately, I'm really happy that it did.

I never thought that I would be, you know, trying to write for an audience that's sort of both academic and I guess lay, but, but I think, I think for the most part it worked. And I know the in the review, the reviewer said that readers would have to wade through some academies and and I'm sure that's true. And, you know, I'm glad it's true. This is my first book. This is pre-tenure. I have, I,

You have some things to accomplish here. Yeah, I have an audience here in academia. But it's also been really gratifying to hear from people outside of academia that the book works for them and that they kind of understand the sociology of it. That it's not just sort of a...

It's not written to be like an expose of a company. It's not written to unveil like, oh, these terrible, greedy people.

It's really about social structure. I was going to say it's about social structure. It's about the systems that are in place that incentivize different kinds of behaviors and constrain other possibilities. Ultimately, it's been great to hear from some folks who are not academics that that really came through to them. That's great, man. Tell us a bit about this firm you call All Done. What kind of firm...

was it? And you know, what was it up to? Where's it at? And where's it at now? Yeah. So when I got there, they just raised their first round of VC funding and they were running a platform for local services. So basically if you want to hire like a plumber or a math tutor or a wedding photographer, like things that people can do for you, usually in person, um,

And, you know, it's not like you can go on Amazon and get those things and you can read reviews of them on Yelp or maybe you can find them on directories, but they're trying to kind of make a more direct connection between buyers and sellers of local services. And yeah, at the time I got there, they had their first round of funding. By the time I left, they had their second round of funding.

These days, they're up to like a $3 billion valuation. So they did hit the unicorn benchmark, as did many, many platforms at that time. You know, like this is a company that, you know, like we said, it was part of that bubble where...

it really felt to investors that the value of these things was just going up and up. And of course, also a product of the low interest rate environment. There was so much money sloshing around during the last decade, people looking for returns in that low interest rate environment. And there's just a ton of money going into VC. And it really, I think, drove up

the valuation of a lot of these companies and All Done was no exception there. And so it's still around and successful and all these kinds of things. And tell us a bit,

I really loved your methodological appendix. I'm probably going to teach it down the road because I thought it was really great, especially for grad students. I think it's going to be helpful. But can you tell us a bit about how you got access? And also, you have a fascinating story because you actually left grad school for, what, a year to work there full time, right? So tell us a bit about that. Yeah.

So, you know, I just started this project as like a class project. I was taking Michael Burvoy's participant observation seminar. And one of the great things about that seminar is that as soon as you start the class, he's like, get out in the field. Like you got to find something. You got to do it. You're going to learn by doing this. You're not going to learn by reading other ethnographies. And so, yeah.

I knew that I wanted to study a startup. And maybe we can talk more about how I got there later, if you're interested. But I knew I wanted to study a startup. And I just started doing the thing that they tell you to do when you're a novice ethnographer, which is talk to everyone you know, you know, and see who might have connections.

I was really lucky, of course, that I was at grad school in Berkeley. And so I was right across the bay from San Francisco and Silicon Valley. And really just through reaching out to colleagues in grad school, I got a couple of contacts with people in startups and startups.

contact number two turned into this field work. I was just incredibly lucky. You were very lucky. You say as much in the appendix too. It's just, it's, it's just, it's incredible how, you know, serendipity, um,

And you were very quickly welcomed into the organization. They said they valued your input and stuff. I mean, it's an incredible. Yeah, well, I was terrified. I mean, so first, first I went in and I pitched them and I said, basically, I had had some experience working in a non-technical role before.

in a startup before grad school. And I said, Hey, like, I can, hopefully I can deliver some value to you. Right. And I can get some value in exchange, which would be data, you know, that I can gather from my research. And, you know, they had just raised their first round of funding, and they're growing really fast. And they actually really needed some extra hands.

And so they agreed to take me on. And I honestly just tried to do a good job because I was terrified of getting kicked out at any moment. And so I thought, if I genuinely am useful to these folks, then they'll want to have me around, hopefully. And I can get the data I need to ultimately, the plan was like, finish this class and write a master's thesis. And that was the idea.

But really quickly, yeah, like you mentioned, like I read about in the appendix, really quickly, folks were saying like, hey, we really like having you here. You're doing really good work. Like, we appreciate you, which was coming from grad school. Like, no one ever says any of those things to you. It's true, man. It's true. So it's like a little taken aback. But, you know, someone almost immediately said like, hey, if you...

are willing to come. I was just, I was an unpaid intern just coming in one day a week. And he's like, we could use more help. And if you come in two days a week, like we could pay you, they're going to pay me like 25 bucks an hour, which, you know, I was like, wow, okay. I'm a grad student. So that sounds, sounds like something I could use. And, uh, I didn't think that as also maybe something we can talk about later, I didn't think accepting payment would, uh, compromise my quote unquote objectivity in the field. So I,

I did that. And then after a few months of doing that, you know, a couple of the executives approached me and said, like, you know, if you're willing to work full time for like a year, at least, like we have a position that we would make for you. And I just figured, you know, I talked to advisors and things like that. And

I just figured I'm never going to get this access. That's true. No one's ever going to get this access again. And even though I hadn't even written my master's thesis yet. And I honestly, you know, like I said, the stuff that I thought I was studying wasn't what I really, what ended up being interesting, but I didn't really have time to dive into that literature and figure out what it was, but I just figured I, this is an opportunity that I need to take. And so I,

I withdrew from grad school for a year. So, you know, the idea is like you get to go and you get to come back, but you're not on, you know, you're not on the roster anymore. So, yeah, so I agreed to do that and they made a position for me and I just really became...

pretty deeply immersed in the inner workings of this organization in a way that would have been impossible if I had stayed sort of on the fringes. I really became a middle manager. I was interfacing continuously

with colleagues both in San Francisco and across the Philippines and the Las Vegas team as well. And that was sort of my- You were working on customer support, is that right? I mean, so that makes sense of why you were interfacing with those teams. Yeah. Is that right? Yeah. So I had basically two duties. One was customer support, which was there were like 20 email support agents in the Philippines and 10 in Las Vegas.

And then I also had this sort of more general role, which they called operations manager, which basically meant that if things changed in the product,

And folks in the Philippines and Las Vegas had to know about it. It was sort of my job to like write up the report about what was going on and try and like get them up to speed. And then also when they had information to send back to us about what was going on with them, that would be funneled through me. And then when, uh,

When anyone in San Francisco had like a special random project that they wanted to assign to the remote teams, they would, uh,

they would send them to me to then sort of like funnel them out. So I became sort of like the delegator in chief or the outsourcer in chief, if you will. What a fascinating vantage point into this whole thing, really. And, you know, it also, the story you tell kind of sets up where I was going next, which is, um,

Okay, I have to tell you, I was just looking on eBay where I go for all kinds of things I love, and there it was. That hologram trading card. One of the rarest. The last one I needed for my set. Shiny like the designer handbag of my dreams. One of a kind. eBay had it, and now everyone's asking, ooh, where'd you get your windshield wipers? eBay has all the parts that fit my car. No more annoying, just beautiful.

Whatever you love, find it on eBay. eBay. Things people love. This episode is brought to you by Allstate. Some people just know they could save hundreds on car insurance by checking Allstate first. Like you know to check the date of the big game first.

before you accidentally buy tickets on your 20th wedding anniversary and have to spend the next 20 years of your marriage making up for it. Yeah, checking first is smart. So check Allstate first for a quote that could save you hundreds. You're in good hands with Allstate. Savings vary. Terms apply. Allstate Fire and Casualty Insurance Company and Affiliates. Northbrook, Illinois. At least in the marketing material that the press sent to me, it might be in the introduction too, but one of the... I really like how...

Something put it that, you know, you're studying what venture capital does to startups, right?

And I think one of the things, it does a lot of things, actually, I mean, that's what your whole book's about, right? But one of the things it leads is to this constant experimentation you talk about. And I mean, given the role that you have to talk to these teams and they're dealing with people who are pissed off about these changes, I think you were probably able to see that really well, like what the constant experimentation was and then the effects it was having both on workers and users, right?

Yeah, absolutely. So, yeah, so VC, right, wants like hockey stick growth curves. It wants companies that can increase their valuation really fast. And the way to do that is generally to get

whatever the key user metrics are to just keep popping, right? So whether that's like your user growth or revenue growth or different types of activity that users are engaged in, you know, if you can show that those numbers are going up and up, then...

then basically the folks who already invested in you are going to be happy and there's going to be folks lining up to invest in you in the future because they believe in the story too. And they think down the line, they'll be able to sell their stake to someone at a much higher price than they paid. So how do you keep getting numbers to go up and up and up? I mean, one of the things that I show in one of the chapters on the folks in San Francisco is that

Generally, you don't do this by making incremental improvements in the user experience. AB tests are not the way. Well, AB tests are the way, but you want to be AB testing the right stuff. You don't want to be AB testing something that...

that you don't foresee any reason, even if it makes your users happy, if you don't foresee any reason that it's going to make those numbers skyrocket, you put that aside and you focus on the things that could really increase activity at scale. So it could be things that are as little as like changing the color of a button. And it's like, oh, when we change this button from green to yellow,

it turns out that 9% more buyers who visit our platform end up submitting requests to sellers, right? Which ultimately has, you know, other second order effects on other metrics. And it's like, wow, we did this little thing and it really made a noticeable difference. And maybe it's...

AB testing the subject lines of the emails that sellers receive and oh, when we use this subject line, they're 15% more likely to open the email. So there's just, there's like an endless array of things that you can experiment with to get, try to get the numbers up. And sometimes they're relatively small like that. And maybe users don't even notice them.

But sometimes they're very significant and they have really important consequences for users. And, you know, like the biggest one that I talk about in the book is the pivot to a new payment model. And this is, you know, again, totally VC related because

When you're raising your first round, the big thing that you want to do is just show that you can bring a lot of people in. Like proof of concept, we have a product, people are using it, isn't that great? But to raise the second round investors, we're telling them, well, yeah, you...

That's great, but now you have to show that you can monetize this. Show us that you can make some real money off of this. They're not saying you have to be profitable, but they're saying, let's get that moving up into the right. Let's get that metric moving up into the right in a significant way.

And so, you know, when you want to bring a lot of users in for your Series A funding, what do you do? Well, you suppress prices, right? Or, you know, like look at Uber in its early years. Like you subsidize wages, you increase wages. You give people an offer they can't refuse, right? But then when it's time to raise subsequent rounds and revenue becomes higher,

of greater interest to investors, well then you have to jack up the prices. And so what I got to see firsthand from the vantage point of customer service was the consequences of experimenting with these payment models

and how, yeah, users were understandably just totally livid. They felt often manipulated, betrayed, like they had sort of invested in the platform under one set of circumstances and policies and fee structures.

And then they sort of had the rug pulled out from under them and they're being told, you know, we're going to increase your price dramatically, sometimes five to 10% or five to 10 X rather. And, you know, we hope you'll keep using the platform and we still think you'll find value in it.

But it's just a tough sell. And so what do the managers do? They get the phone support team to call thousands of sellers who are high value customers who are spending money on the platform. And they just have to sort of absorb this barrage of abuse and anger to keep as many sellers as possible on board in spite of this

massive shift in the payment model. So yeah, from my position, I really got to see the consequences of that rapid experimentation for both users and the workers who had to sort of manage the fallout of the experiments. I was thinking about this. There was a really fascinating point you made. I think it was in the conclusion, but

So often in kind of STSE related fields, I see kind of an obsession with design. It comes out of like the Langdon winner politics of design stuff, right? And I think it also comes with like STS's nearness to engineering and design in academia and kind of like too often taking the engineering perspective on things. And the only one answer, you know, where the school I'm from is like,

no, we really need to focus on use, right? This is like the David Edgerton move. Like let's focus on humans in organizations and at home actually use shit because that's how stuff happens. But I think that, you know, you're working in one of these spaces and you were kind of telling this story earlier about how there was all this stuff on like the gig economy and

the sharing economy, which was all about like so many focuses on riders and users and drivers and all this. And you're saying like, this is one of those cases where focusing on, you didn't use the word design, you use the word building. I really like that. And I would also say the business of building something was really the move that needed to be made. And I just thought that was really interesting about like how so often like the design focus, the right

antidote is the shift to use. But you were looking at a case where actually there's so much shit on the use, both of drivers and riders in the case of that the right thing was to building. And I just wondered what you thought about that observation. Yeah. No, I...

That was really important to me to bring that perspective into the conversation because, I mean, listen, I love the literature on platform work, for instance, you know, gig economy platform work. We've learned so much, so many fascinating, important things about the consequences of design, you know, of platform design of algorithm design for users. And, um,

As you mentioned, also, like, I mean, the use thing is also, of course, such an important corrective to all the deterministic arguments that are out there, you know, because it keeps us focused on contingency and conditionality. And all that stuff is so important. You know, I teach this stuff online.

I swam in those streams when I was doing the future of work stuff, you know, in many ways, like, no, we got to understand the context here. Um, but I think that the pro to me, some of the problems with that literature is like,

Practically, what do you do with it? Okay, so the algorithms are bad. So what do we need to do? We need to make the algorithms less harmful. And there's this entire...

cottage industry slash a mansion. I mean, there's so much academic research now that's about like algorithmic design and harms and how to make them, whether it's more fair or transparent or this and that. Right. But the thing that always gets me about this stuff is like, well, most of the algorithms and systems we're talking about are made inside of

capitalist organizations. And there's a reason they're designed the way that they're designed. It's because it's not because the people who make them are bad people.

I mean, maybe, you know. Some of them are. Some of them are, but that's not the point. That's not the point. That's not really the point. It's not about bad people. It's not that there's something inherently bad about like algorithms themselves or machine learning itself. The problem is that like these people are all basically, they're playing a game. Like they're struggling. They're doing the best that they can

with the conditions that they've been given to produce the outcomes that their organization is supposed to produce. And so, you know, that's why...

you see algorithms doing bad things. That's why you see founders doing bad things. If we don't pay attention, I think to like, what is shaping the build, right? What is influencing that moment of the build? I think that's where you have to intervene if you're really interested in changing technologies. You can't just say like,

Uber's got bad algorithms and so we need to, I don't know, outlaw this algorithm, ban that algorithm. They'll come up with some other shit. Exactly. But, you know, I'm really interested in just looking at, you

somewhat pie in the sky, but also totally grounded in objective reality that exists now. There are other ways to do this. There are other ways to invest in innovation that are not about making the wealthiest among us even wealthier. They're generally kind of models that are marginal to the tech industry, but

But they do exist. People are producing different kinds of outcomes with these models. And I think they sort of point to the fact that, you know, if we can find ways to shrink VC's influence and make more space for alternatives to VC, I think that's the best way to

to keep tech companies from doing bad stuff. It's to change the ownership structure and take investors and their influence further and further out of the picture. That's really nice. And the other thing, I mean, and it's kind of,

You kind of set up what I wanted to ask you next. And this is a little meta too, but you organized the book around a series of lags. And I wanted to talk to you about that. And one thing, this didn't, I mean, and maybe it never occurred to you. I don't know if you've ever had to read a lot of Thomas Hughes, the historian of systems and technology in your career. All right. So, right. I mean, he's real big in the history of technology in some STS lands, right? But Hughes had this concept of a reverse salient.

And it's basically so next week I'm delivering a talk on the sociology of problem selection in science and technology. Right. And so I've been thinking about Hughes because Hughes's concept of a reverse salience is basically like as you're developing a system, there's these places that lag from

So a salient is in a military where like the avant-garde is charged ahead and the reverse salient would be like where things have fallen behind. And Hughes's argument is that system builders and inventors, they're often, what happens is the, you know, you can't build the system up further without dealing with that thing that's holding you back, right? And so it's like kind of the squeaky wheel gets the grease or whatever that old phrase is, right? Yeah.

And so that's his notion of like system development. And it involves kind of identifying these reverse salience or having them forced upon you or something. And then they get your attention, they get work. So I was just thinking about that because you have like a series of lags, right? And we can talk, I mean, I think that you're saying the financial system or like, you know, the social structures that these folks are working in are what's highlighting what a lag is, right?

So I wanted to talk through, you have at least three lags. You have the valuation lag, the technical, well, you have technical drag and then organizational drag. So I wanted to kind of talk through that lag picture and then we can kind of play it out a bit. Yeah, that's, it doesn't surprise me to hear about

your source on that because I think there were, I think there are some other archaic sources on which this may have also initially emerged from. I'm remembering in my dissertation writing group many years ago that, you know, this lag term came up and I think there were other references that were bandied about too. I just, while looking into this this morning, I just discovered that I stole a term from the famous,

economic historian, Nathan Rosenberg this morning. So shit happens. Bad man. Okay. Um, yeah. So, uh, but yeah, I mean, so the way that, uh, the way that I conceptualize the lags and drags, you're right. It really does come from the way that VC shapes, uh,

basically the problems that startups encounter. And I use these terms lags and drags because they have this sort of, you know, connotation of

speed, velocity, temporality that are all just so important for companies that are, you know, seeking repeated rounds of venture funding and, you know, trying to get as big as possible, as fast as possible before someone else figures out how to do what they were trying to do quicker than them. And so the idea behind these lags and drags was basically just setting up

the way that VC and its demands systematically creates organizational problems. And kind of like you said, these are like things that are holding you back, that are preventing you from getting to that next stage that the company needs to be at in order to continue to increase its valuation. And so, you know, I start, it's not totally, you know, it's not totally intuitive, but I start with a drag.

I'm sorry. No, I start with a lag. See, this is the problem. Part of why it's not intuitive is lags and drags sound very similar. But I start with a lag, which is called valuation lag. And the idea there is simply kind of setting up like, what does this team in San Francisco have to do? Like, what do they do? Well, they're always trying to basically make... The company is valued at a certain level by investors, and they need to bridge the gap between...

the realities that the company is facing on the ground where, you know, things are generally always breaking and not working and not where they're supposed to be, not where anyone wants them to be. And they have to sort of

bridge that gap between reality and this imagined future of, you know, incredible, unbelievable success. And so they are in their efforts to constantly be engineering change in the platform. They are basically, you know, trying to catch the company up to what they and investors hope its actual value on the market can be.

This is a Reese's Peanut Butter Cups sound experiment. We're looking to find the perfect way to hear Reese's so you'll buy more of them. Here we go. Reese's. Reese's. Reese's. Reese's. Reese's. Hey, get out of here, you little stinker! Reese's. Reese's. Reese's. Peanut Butter Cups. That breathy one sounded very creepy, am I right?

This episode is brought to you by Peloton. If you're looking for flexible workouts, Peloton's got you covered. Summer runs or playoff season meditations. Whatever your vibe, Peloton has thousands of classes built to push you and has something there to adapt with you whether you need a challenge or rest. Peloton has everything you need whenever you need it. Find your push. Find your power. Peloton. Visit onepeloton.com.

But as they're trying to do this, they're constantly hitting up against these organizational problems, which I then call drags. So early on, and I'm not saying these particular drags are universal to every startup. Right. It could be different in different organizations. But I do think that the VC cycle is going to be systematically producing drags of some nature as startups work their way up

the funding cycle. And so, you know, I start again, series A, when the idea is that what you really have to do is get a bunch of users in here and show that lots of people want to use this product.

That's when they're facing what I call technical drag. Like I said, the engineers are really busy. They're not actually really doing much engineering at all because they're trying to recruit and interview new candidates. And yet the company is actually still continuing to work and to grow its user base by like 50% over just one quarter. And so I was trying to figure out like, well, what? How did that happen?

And so they overcame these technical limitations by plugging in these information processing workers in the Philippines and, you know, setting them up with processes where either they would like complete an algorithmic process that they sort of coded up in a messy way. And then they sent it over to folks in the Philippines to like use their judgment to actually finish up

a process that was, you know, that software had started. Or maybe the folks in the Philippines would do things like

do a lot of data mining on the internet to try and support an experiment that an engineer wanted to run. Basically, like, make the experiments go faster. They really provided sort of this behind-the-scenes computational infrastructure that, you know, that allowed innovation to proceed in spite of both, you know, pretty significant resource constraints and also just limitations of machine learning at the time.

So that's, you know, you got one drag there, you got another drag that then comes up as they start to seek their Series B. And again, the pivot to revenue generation. Well, at this point, when you're changing your business model, and you're trying to bring all these users along with you,

That's what I call trust drag, where users just feel like their trust is betrayed. Like, what do you mean I have to pay 10 times as much for the same thing? This is outrageous. And that's when you bring in those phone support workers to try and solve that problem of, you know, of users' trust in the platform being undermined. Yeah. And then finally, I talk about after that second round was raised, I talk about organizational drag.

And at that point, the big issue is that when you're an early stage startup, there's all these ways that you structure your company that are really specifically tailored to being an early stage firm that's really trying to adapt to unexpected contingencies. And everyone's wearing a lot of hats. Everyone knows what's going on across the whole company. And then now you've raised the second round and suddenly the VCs are saying,

okay, boys, it's time to professionalize. Right. Let's bring in the MBAs. Yeah, let's bring in the, you know, the CTO who was the CTO at this company everyone's heard of. Right. You know, it's time to turn this into what future investors will view as sort of like a credible business. Right. Yeah.

I think this is very normal, actually. Oh, totally. Yeah, yeah, yeah. It's very normal. It's really part of the process. And that's what the investors were telling the folks at Aldon is like, this is what we tell every firm in our portfolio when they've raised their Series B. This is what you should expect. Got it.

But generally what it does is it sort of creates this problem inside the organization that, again, is probably more or less universal to successful startups where folks who were in really early start to feel really disenchanted because they realize like, oh, this thing that I helped to build with my blood, sweat and tears. Yeah. This thing isn't really my thing anymore. Right.

And actually, you know, new things are being asked of me. Maybe I'm being asked to manage when I wasn't before. Yeah. Maybe I now have, I'm now reporting to someone who wasn't even here last month. Right, right, right. Who now suddenly is making way more than me. Yeah. I have to work on all these projects that, you know. It's all getting more boring. Yeah, totally. Totally. Yeah, the whole thing gets more boring. And so like,

sort of getting disenchanted across the organization. But of course, this ends up meaning different things at all done, depending on like which team people are on. And I talk really especially about the folks in Las Vegas who played such an important role in making that second fundraise possible because they made that pivot to a new business model happen. Yeah. But ultimately, the

management just decides like for a big company that team is not really cutting it anymore and we want to start anew with kind of a more professionalized team and so all these folks who were there at the beginning and sort of

uh, hoping and, and really, uh, led to expect that as the company grew, it would mean sort of better things for them actually sort of just end up, uh, uh, left behind. Yeah. So, uh, yeah. So anyway, that's, that's kind of, uh, it's really nice, man. It's a nice model. And I, you know, I think that, yeah, it's one of the,

I think you have a real kind of picture or, you know, theory model here. And it's just really nice. And it plays out nice structurally across the book. I wanted to, you know, I wanted to ask you a bit more about that alternatives thing that you were talking about earlier and what you talk about in the book's conclusion. But I also thought I would just connect it to, you know, the kind of what's next question I usually end with.

you know, what's next for you. And the reason I thought I would connect them is if I remember correctly from that nice beer we had together in Philly, one of the, one of the, uh, you know, projects that you were playing with at that point actually was like exploring alternatives in a deeper way. So yeah. Are you still thinking about that? Yeah. Yeah. So I am still, I'm still thinking about that. I've had some, done some early interviews at this point. Um, so I guess you'd call it like the pilots. Uh,

And yeah, I just I got really interested as I was writing the book in...

what you know again so if if i'm not in the camp that says like well the technology is bad and we need better algorithms yeah i am in the camp that says the finance is bad we need better financing yeah so um you know and i just started looking out there for like what are the other models for making tech and yeah you know you and i know there are a lot of interesting examples out there i love uh

my colleague Jessa Lengel's book, An Internet for the People, that's about Craigslist. And, you know, this is a company that's been more or less privately owned for 30 years, aside from kind of a little blip in the middle.

But Craig Newmark, the founder, is really this dude with a web 1.0 ethos. And he believes in an open internet. He has barely changed this product over like 30 years, right? Right. He's not harvesting user data. He's not selling ads. He's not A-B testing. He's just kind of providing something that is, by the way, is a profitable company. Yeah.

Yeah. He makes a living. Totally. Totally. He's fine. He's doing fine. And he will tell you, I've seen him, Jessa brought him in to do a talk, you know, like a few years ago, he's doing totally fine. He's got a foundation. Like it's no problem for Craig. Um,

Right. You know, but he's done this thing where he's built this private company and he's able to balance the profit motive with other values. Yeah. And he's only charging you to use this platform if you're a landlord or you're an employer, you know. Right. Because he knows you've got some money.

So, you know, I'm not saying that it's obviously a lot harder to start a Craigslist now than it was in the mid-90s, right? There's so much more money and so much more acquisition and copycatting and... Well, the network effects are just hard to reproduce for something like that. But it does, to me, just sort of point to the fact that, like, there...

There is the possibility when you don't have external funding and external control, even if you want to have a for-profit company...

you can still do it in a way that maybe balances profits with other values. I think that, you know, proprietary capitalism, I think that that long, longer view look is also part of it. You know, like the DuPont family, DuPont was one of the early R&D labs. And, you know, now they're actually splitting up again. They're all fucked up now because things have changed in capitalism since that day. But I think

When it was a family-owned operation, it really was able to take a different perspective on a lot of these things. I think there's other kinds of forms of proprietary capitalism you could explore the history of in thinking about that as an alternative. Lay a couple others on us. Do you want to be collectives or cooperatives? Yeah. There's a really wonderful research community in this area right now.

Trevor Schultz is sort of one of the leading lights, but there are many really excellent scholars internationally that are doing work in this area. But yeah, the idea of a platform cooperative is simply that, like other cooperative businesses, it's a platform, but it's owned and operated by the workers who rely on it to find gigs. So, yeah.

There are examples of taxi cooperatives out there. One of the most famous examples is this house cleaning cooperative in New York called Up and Go. And the basic model is simply that, yeah, this platform connects buyers and sellers of whatever the service is, but...

It's not funneling the gains into investors' pockets. It's not funneling the gains into hiring a ton of engineers so you can build even more features that maybe nobody needs, but we'll get the numbers up.

Like 95% of the money is just going to go back to the workers and maybe 5% will go towards maintaining the platform and building new features or something like that. So, you know, it's an actually existing model that's working out there. Is it possible to sort of compete with big tech and big capital? You know, probably not in the same way. Scale on the level of big tech is just not going to happen probably for a co-op, but...

But we do see examples of, you know, successful cooperative businesses that serve local, you know, relatively delimited communities and are able to generate, you know, enough revenue to give people a better living than they would be getting through these, you

So, you know, like I said, there's a lot of really interesting research of people looking at this stuff all over the world. And I think this is such an important potential direction.

And do I remember there was another project you were playing footsie with? Are you-- Yeah, I'm playing footsie also with-- and actually have done a bunch of interviews already of just sort of the other side of this platform. Now I'm sort of betraying my initial statement. But the folks who use the All Done platform, I have obviously now a very robust story of the build side. Yeah, yeah.

But this is a population that I think is under studied. And so I am sort of making a foray into trying to get a better handle on the experiences of these local service providers and sort of like what works and what doesn't work for them. Because these platforms, you know, and everyone who's in the space knows this now, like,

There's not one platform story. There's not one algorithm story. There's different sectors. There's different platforms within a sector. That's right. Platforms themselves change.

so, so that's sort of the other thing that's, uh, on my, uh, stovetop at the moment is looking at the, the, the seller side of this platform. Nice. And Julie, Julie, your colleague, Julia Ticona, she's doing a care platform, right? Or care platforms. Yeah. Yeah. It's nice that people are broadening out in that way. Exactly. Yeah. Her work is really fantastic. And I know you had her on the podcast, uh,

but that's a really exciting project that I'm very much. She's she did some research on this a while ago, but I know it's going to, it's going to be a book soon and I'm sure it's going to be a tour de force. Yeah, man. Well, Ben, thanks so much for coming on, bud. This was a lot of fun and I'm really glad to highlight this book, which I enjoyed a lot and I think is real sharp. So thank you very much. Thanks for having me, Lee. Always great to chat. I hope you enjoyed this episode of our podcast.

You can reach us with questions, comments, and suggestions at LeeVinsel at gmail.com or by following me on Twitter at STS underscore news or on YouTube at People's Things. Our podcast is distributed by the New Books Network, the leading platform for academic podcasts, so that you can find us wherever you get your podcasts.

Peoples and things, like most things in this world, depends on the work of many people. I want to thank my brother Jake Vinsel for writing the music for the show. I want to thank my buddy Juliana Castro for designing the logos for the podcast. You can check out her work at julianacastro.co. Joe Fort is the producer for the podcast and Mandy Lamb is the production assistant.

This podcast and other Peoples and Things programming are produced in affiliation with Virginia Tech Publishing and supported by the Center for Humanities and the University Libraries at Virginia Tech. For information about other podcasts from Virginia Tech Publishing, visit publishing.vt.edu.

For the entire Peoples and Things team, I am Lee Vintzel. And most importantly, I want to thank you for listening. Thanks.