Dave Pickles - CTO at The Trade Desk

(Joel Beasley at 00:00:00) Hello, my friends. Today we are talking to Dave Pickles, the CTO and co-founder of The Trade Desk. And we discuss the balance between privacy and connecting consumers with the products they want, how to practice servant leadership for your team, and how screening for defensiveness can produce a much more collaborative organization. All of this right here, right now on the Modern CTO Podcast. Here we go.

(Joel Beasley at 00:00:25) This is the Modern CTO Podcast. Hello, Dave.

(Dave Pickles at 00:00:38) Hello.

(Joel Beasley at 00:00:39) How you doing, my friend?

(Dave Pickles at 00:00:41) I'm good. I'm good. How are you?

(Joel Beasley at 00:00:42) Dude, life is good. I actually made bread today.

(Dave Pickles at 00:00:46) You sound like you're an introvert.

(Joel Beasley at 00:00:49) I do. No. Oh, because that's what they're doing. They're hiding out making bread.

(Dave Pickles at 00:00:54) If you're like, "Well, I love this pandemic thing. I don't have to leave the house. I bake."

(Joel Beasley at 00:01:00) No. I'm actually at the office. It just used to be a therapist office, so it kind of has a homey feel. But yeah, I've never made bread before. And today was my first time making bread because I was trying to do the gluten-free thing, and all the gluten-free breads I thought tasted horrible at the store. And so I heard through a friend of a friend about this stuff called Miracle Mix, and so I bought some and gave it a shot. And I was like, "Whoa. It is the best thing ever." You wouldn't even know it's gluten-free.

(Dave Pickles at 00:01:30) That's awesome. We do a lot of baking, especially during pandemic too, but we use tons of gluten.

(Joel Beasley at 00:01:37) Oh, you know, what started me on it was, and by the way, this is the podcast. We're just going to talk and hang out. That's cool.

(Dave Pickles at 00:01:44) Sounds great.

(Joel Beasley at 00:01:44) So my dad and my stepmom, they own a clinic called Peaks of Health. They help people achieve their best health, and they're physicians. And they've had this business for ten years, and they started asking me questions about their data. Like, "You know, how do we find—we hear personas, we hear all this stuff. You know, how do we do that with our people?" And so I started—I downloaded copies of their QuickBooks data. I started reviewing their transactions for fun. And I learned about their business, and then they do all these blood tests and stuff to find out people's sensitivities to—everyone has some sensitivity to everything you eat. It's just where it's at on the spectrum. And so I took these blood tests, and I was like, "Oh, man. Half my diet, I'm highly reactive to," and it suggests I stop eating those foods. So I did, and I just got a lot more energy.

(Dave Pickles at 00:02:32) That's awesome. I had a really bad back injury, and a lot of chiropractors and people like that told me, "Oh, first thing you do is get rid of gluten, get rid of dairy." And so I did it, and I noticed no difference whatsoever. So I got my individualized test results as well.

(Joel Beasley at 00:02:48) Did it say you were reactive, though?

(Dave Pickles at 00:02:50) No. We didn't do any testing. I'd be interested, actually. This whole pandemic is absolutely killing me as a data scientist. So if there was some more data certainty that I could go latch onto, it'd probably help with my anxiety.

(Joel Beasley at 00:03:06) Let's talk about this. What do you notice? What gets at you as a data scientist about this?

(Dave Pickles at 00:03:12) There's garbage in, garbage out. The data being collected about caseloads is awful. Like, in Ventura County, everybody freaks out on every Monday because the caseload went up. But that's only because they do all the tests on Monday for basically, you know, 11:59 on Friday through the weekend. And so there's just this perfect little spike in the data. And, you know, if you know about how data works, you're like, "Oh, okay. This is—I need to do some normalization or something like that, or I need to zoom out a little bit." But, you know, it's things like that. And just, what is the data? What is the test rate? How do you do the controls? And all the conclusions that are being drawn, I would never stand behind. If I was responsible for this data, I would be saying, "We don't know much."

(Joel Beasley at 00:03:58) Yeah. We're still looking for our process.

(Dave Pickles at 00:04:00) Live in fear.

(Joel Beasley at 00:04:03) I'm just not that type of person. Right? I'm just not a fearful, fear-driven individual, but I've trained myself to be somewhat optimistic. And when I'm not optimistic, I'm pragmatic or, you know, just try to be, like, first principles approach. But fear doesn't—I guess what I've learned is that I make decisions from the state that I want results in. So, like, I'll make decisions when I'm focused, in a good mood, or when I'm neutral. But when I'm angry or distracted or fearful, I don't make decisions because those never end up well.

(Dave Pickles at 00:04:39) No. That's really true. Actually, one of a huge part of sort of my reputation in my career is that I am just positive. And I'm like, "I've—we can do it. We can take the hill." And if things go wrong, it's like, "Oh, well, let's try the next thing." You know, with little pain. Like, there's this famous meeting—famous in my world—meeting at Microsoft where they had this thing called the Ship Room, where everybody goes to Ship Room from the whole division. Because, you know, you have to try to ship things all at once at Microsoft for some reason nobody understands, circa ten years ago. Everything's changed now, I'm sure. But there's the most blamey, backstabby meeting in the whole company. Because if you're the one holding up the ship, everybody looks at you and goes, "Oh, dude. Like, you know, no good reviews for you." And, you know, something came up in that meeting that I'd done wrong, and I basically said, "Oh, let me go fix that." And I got up and left. And everybody's like, "What just happened? He didn't trick anybody else. He just went and fixed it. That was weird." Yeah. I always think of that as—I noticed a change in how everybody looked at me after that. Then it was like, "Oh, that's the way to be."

(Joel Beasley at 00:05:51) Yeah. People respect discipline, and they respect people who have a high bar of quality for their work. And that leads you, or lends you, perfectly to founding a company. Right?

(Dave Pickles at 00:06:01) Yeah. Caring about quality is super important. Yeah. The founding of the company has most to do with the fact that I had an opportunity in front of me that I felt like I understood better than anybody else in the world. It wasn't so much that I wanted to have a company or anything like that. I like doing work. I like helping people. But it was like, "Hey. I think this thing's happening that's going to change the whole world, and I've been thinking about it two years longer than anyone else."

(Joel Beasley at 00:06:31) What was it?

(Dave Pickles at 00:06:32) Well, that was this idea that we're going to reshape online ads to look like the financial market.

(Joel Beasley at 00:06:37) Oh, tell me more.

(Dave Pickles at 00:06:38) So when I went to Microsoft, I didn't know anything about ads. I knew about building scalable, reliable software systems. I've been working on telephone systems. I, you know, voice over IP, six nines reliability, crazy stuff like that. We, you know, rented space in amazing cages with physical security and Halon and all this stuff. That was my world. And I was like, "Advertising. Okay. So how does this work?" And as I looked at it, you know, it was just really, really inefficient the way everything worked. You know, you sort of give some money to Yahoo and hope for the best was kind of how advertising worked at that time. My co-founder tells a story about—he gave $100,000 to Yahoo with certain targeting criteria, and the campaign worked great. And he made a ton of money because he was doing arbitrage. Right? So if you sell more units than you had to pay for, then you actually pocket the money. And then he did the same buy with the same targeting criteria with $500,000, and it bombed and it wiped him out, way beyond anything that he made on the first one. And he's like, "Why? Why did that happen? How come, you know, Yahoo screwed me?" But Yahoo didn't screw him. Yahoo just didn't have any control over what was happening. They hadn't done the work to understand exactly what the properties of this commodity were that you were buying. They had some controls, but it wasn't really well-defined what it was. And so what we did is we said, "Well, let's go create a way to have an exchange where you understand all the properties of the commodity so you can have price discovery." So it's the very foundation of a financial market. It's like the New York Stock Exchange or the Chicago Mercantile Exchange for online ads. That's what I built at Microsoft. And then about the second I was done with that, I was like, "Oh, I think this is going to be really powerful because now you can be super specific about the kind of ads you want to go after." And, you know, my brain's just spinning on all the things that you could do as a buyer in this world to create advantage for yourself. And I was like, "Okay. So if you look at the financial markets," Jeff and I were talking about this together, "if you look at the financial markets, the New York Stock Exchange is not really the best business in the financial market. Goldman Sachs is." And so we started saying, "Oh, I think all these corollaries are right. You know, I think that, you know, there's differences too, and it's human behavior. It's still going to be a little harder to quantify than oil or, you know, pork or whatever. Like, you know, it's slightly different." And then we also evolved pretty quickly to say, "Well, let's not build Goldman because Goldman's a company that uses their own data to create benefit on their own behalf. They're playing on their own behalf and they're taking every advantage they can. Instead, what if we built the Bloomberg terminal?" So you build a system where you allow all of the buyers in the whole world to have all the rich metadata and make good decisions. And that's the idea that occurred to us when we founded the company was all of that. And I was like, "I don't think anybody else is thinking about this in this way right now. There's all these other companies in the space, but if we go build fast and if we build a modern, lean company, I think we can really win." And that was how we started.

(Joel Beasley at 00:09:49) That's amazing. I want you to bring it home for me a little bit. So I'm a small business owner. We run ads, and, you know, on different platforms, but I can't connect. And I get the concept of what you're saying. Like, they're two isolated topics in my head. Like, I know what my experience is like when I'm running ads, and then I know what you're kind of explaining about this concept because I have a background in financial software. So I understand what you mean when you're saying, "Bloomberg terminal." And so, like, help me connect those. Like, why am I not running into your product?

(Dave Pickles at 00:10:18) Mostly because it's difficult to use. So for the same reason, you don't have a Bloomberg terminal in the corner of your office. You work at E-Trade or you get on Fidelity.

(Joel Beasley at 00:10:28) I think, uh, Robinhood. That company is awesome.

(Dave Pickles at 00:10:34) Yeah. There's a bunch of them that make it more accessible and that also, you know, we have high minimums. And so we just—we sit deeper in the market than the consumer typically does or even a small business owner does. And that's really a strategic decision on our part in wanting to have a smaller company. You know, I wanted to have a one or 2,000-person software company, not a 200,000-person ad agency.

(Joel Beasley at 00:11:02) So what are they doing on the platform? Like, I get this—you're—I'm not connecting the dots and I really want to understand this because I like marketing. What are they actually doing, like, in that Bloomberg terminal of yours?

(Dave Pickles at 00:11:15) So the main thing, the primary use case would be you have your business, you have data from your website, you know who your best customers are, you also hopefully know who your second best customers are, and so on. You know, you've got a good idea of the lifetime revenue of all of your customers because you've done your work. You can then take those and you can put them in our platform, and we can say, "Okay. You've got 400 amazing customers. Let's go find everyone else in the world that acts substantially like them in terms of browsing behaviors." So it goes to the same kinds of sites, browses on the same kinds of days. Whatever. Like, you can, you know, you can do unsupervised machine learning and just go say, "What are all the other consumers that are likely great customers of yours?" And then—

(Joel Beasley at 00:11:58) That's brilliant.

(Dave Pickles at 00:11:59) Create a stack rank of them and just go put budget into, you know, getting in front of them and seeing whether—then, next step, see whether or not your creative causes them to come to your site, you know, and so on and so forth. So you create a series of tests to go find new customers and bring them in. And then once you've brought them to your website, then you have that direct relationship with them and you can do your thing on your site. That's not our part. The stuff you do on your site, it's up to you. That's MarTech in our world. We're AdTech. There's a little bit of a line there just to try to keep us from trying to build everything.

(Joel Beasley at 00:12:32) Which is important to specialize. It seems like you've taken the custom audiences or lookalike audiences on steroids. Like, it's like you've extracted that single feature into a whole more advanced business.

(Dave Pickles at 00:12:45) Yeah. And that's the easiest to understand example of what we do, but it's really very general. So you can construct any experiment you want to to go buy media, see how it worked, and then also make sure that you're paying the right price for every impression. That's one of the key differentiators for our platform is we made it so that—we call it expressiveness. So you can pay a specific price for every impression if you want to. That one's worth more than that one's worth less than that one, which is difficult to do technically when you're talking about—we look at 11 million ad requests a second. And for your brand, if you wanted to spend a lot of money, for you, $50,000 a month, I'd need to pick you, like, four of those. So getting from 11 million to four for every advertiser that wants to work on your platform is just a technical challenge like I've never seen. The implications for just how much RAM you need, how much compute you need to be able to do all that. Just to be able to listen to 11 million ad requests a second requires a lot of sockets, a lot of servers, a lot of network. It's just all of that is a very, very fun challenge to solve. But so, just that basic thing that we do has a million applications for a marketer.

(Joel Beasley at 00:13:56) And where are those requests coming from that you're listening to?

(Dave Pickles at 00:14:00) It's everything you can possibly imagine. It's every web publisher. It's every app. It's, increasingly, audio, connected television. The connected television boom is the story of the future of our company, really.

(Dave Pickles at 00:14:14) Everything we've done so far has been a warm-up for the big show, which is kind of doing our same thing and bringing personalization and strong results into the world of CTV, where previously in linear TV, it was very much spray and pray. You just would throw out ads to a national audience and say, "How'd it work? I don't know. Market share went up by a quarter point." It's really difficult to have much analytics in that world.

(Dave Pickles at 00:14:40) And so now with CTV, you can do things like manage frequency and make sure that you don't show the ad to the consumer too many times and annoy them. Simple things like that are just so powerful. And then you also bring some of this lookalike modeling concepts into putting ads into CTV. And what you get is the ability to prevent what happened in cable from happening in CTV. So what happened in cable is there wasn't very good targeting and there was a bunch of oversupply. And so you ended up having more and more ads laid into each hour of content over time to the point where it became unwatchable.

(Dave Pickles at 00:15:16) If you're used to streaming and you go back to cable, it's shocking what the ad load looks like. And if you let a company like ours do the targeting, we can pay a much stronger price per impression because we get the most out of every impression. We get a great result for the brand. And then the publisher only needs to show five or 10 ads an hour, say, something like that. And so then we can have a really awesome user experience for CTV that turns the flywheel even faster on getting everybody to move over into connected television and just enjoying the more modern world that we're living in, the golden age of media.

(Dave Pickles at 00:15:57) As long as people can get back to acting, the golden age of media will continue.

(Joel Beasley at 00:16:03) And I like the way that the future is headed where the advertisements aren't the way they were when I was growing up. Right? Because my daughter is three, and whenever there's a—my wife doesn't do premium of anything. I do premium of everything because I cannot stand ads. I'm premium on Pandora, premium on all my services because I don't want my attention interrupted.

(Joel Beasley at 00:16:27) And I like my ads to be embedded in the content I'm consuming creatively to where I don't even really know it's an ad because that's the best type of ad. Right? It's like, "What shirt is that guy wearing? I want that shirt." I should be able to click on that shirt and get it.

(Joel Beasley at 00:16:41) You know? But my daughter, she's watching Baby Shark or something and some ad for home cleaning services come up, she goes, "No. Oh, no. Oh, no." And it's like, that's exactly how I feel inside and that's the future. Nobody wants that interruptive type of advertising.

(Dave Pickles at 00:17:01) Yeah. It's like what are the noble pursuits of advertising? Right. It's connecting people with products that they're interested in. Right? That's the basic thing that we do and we drive the economy in a very real way. So it's helping brands find new customers, helping consumers discover new products that—it's more of a discovery process than a bombardment process when it's done right. And yeah, you're right. The more native they can be, the better. The more native they are, the harder they are to scale.

(Dave Pickles at 00:17:32) So what you're talking about, the amazing—imagine the amount of work it takes to put the shirt on the person and then have that in front of you. Right? You compare to the process of, "Well, let's throw a creative in front of you that we've modeled out that we think will kind of catch your eye and we'll keep it short." I remember when a certain brand was running ads in a game I was playing, and all of the other ads were 30 seconds and they ran a five-second spot. I'm like, "That's awesome. Thank you." And I remember your brand for that. That was cool. It's little things like that that can make a big difference in the relationship and the trust relationship between a consumer and a brand.

(Joel Beasley at 00:18:13) That's an important thing that I think about a lot because I feel that trust relationship and I feel when it's eroded, like when the people's automation goes haywire and you get the same, "Oh, hi. I'm your account manager" from seven different people or it's just like, "Oh, man. That sucks. They need to get their stuff together."

(Dave Pickles at 00:18:33) Or just even the shoes that follow you around the internet.

(Joel Beasley at 00:18:36) Oh, yeah.

(Dave Pickles at 00:18:37) That's a very basic use case for the kind of technology that we have and one that we discourage pretty actively. The over-frequency to try to go get somebody who—it was the shoes were sitting in your shopping cart. You weren't probably gonna come back and buy them. I wonder how many times you made people not come back and buy them with that kind of an ad because they're like, "Forget it. I don't wanna be a part of this anymore. I'm gonna go buy it somewhere else."

(Dave Pickles at 00:19:00) You have to be thoughtful when you're putting together an ad campaign. Because if you're just looking at the data and you're letting a machine make all the decisions, it's gonna do what it's gonna do. Right? It's gonna say, "Oh, well, we tested it and the optimal frequency is 600," or whatever.

(Joel Beasley at 00:19:18) Yeah. I believe there's somebody out there that has had those results, and they pursued that yet.

(Dave Pickles at 00:19:23) The outliers on those curves are dramatic. It's really crazy.

(Joel Beasley at 00:19:28) They purchased so the ad will go away.

(Dave Pickles at 00:19:31) Well, hopefully, you don't do that. I mean, so there is—if you've seen the icon in the top corner of all those ads, that's one of the things that I help pioneer to make sure that there is a way to escape and to say, "I don't want this ad" or "I don't want any ads." So that should all be at your fingertips if you're getting ads from any reputable company now.

(Joel Beasley at 00:19:49) Is that that ad choice thing that I'm using?

(Dave Pickles at 00:19:51) Ad choice. Yeah.

(Joel Beasley at 00:19:52) Yeah. That's pretty cool.

(Dave Pickles at 00:19:53) Yeah. We went to a lot of work to make that work and then Safari broke it. We created a way to go set first-party cookies in every domain that has anything to do with advertising to say, "This person says they don't want ads." And if somebody says they don't want ads, I would rather not waste my money putting ads in front of them. So why wouldn't I let them say that?

(Dave Pickles at 00:20:16) And now it no longer works in Safari, which we're trying to resolve.

(Joel Beasley at 00:20:20) So being in this ad tech world, what are your thoughts on—do you have deep thoughts on data privacy?

(Dave Pickles at 00:20:27) Yeah. Totally. So I—that's what I spend probably most of my time on at this point is trying to help Google avoid making mistakes with their browser. And so I think that there is no real tension between your actual privacy and the kinds of advertising that we want to do, that our brands wanna do that matters. So there was, for a little while, sort of an arms race towards "I'm gonna be more specific than the company that was in here a minute ago pitching."

(Dave Pickles at 00:20:59) Not only do I wanna know if you went to Walmart, I wanna know what side of the street you walked down. Right? That stuff is insane and that stuff leads to trust violations for brands. So if Mercedes has spent—Mercedes says they take about 10 years to build a customer beginning to end. I can ruin that in one minute by freaking somebody out with something creepy.

(Dave Pickles at 00:21:21) And so that never had any commercial value anyway. And so what we've been trying to do is build these opt-outs. Let's go build frameworks for GDPR compliance, the European regulations that are along these lines. Basically saying, "Can you guys be just a little less specific and do things that are scaled and impactful?" That, you know, if you look at the data that's being used to target you, at least in our platform, it's incredibly innocuous.

(Dave Pickles at 00:21:48) If you looked at Facebook eight years ago, you might be a little concerned with what they were doing with your data. But they too have done a lot of work to say, "Let's go create a ton of transparency." And then because we have transparency, let's go curtail some of the things that I don't think were done strategically, but more just done by some small team with a project somewhere that were like, "Oh, that's a bad idea." So there's been an evolution towards transparency and towards being benign with what we do with this data because that's what's best for everyone and anything else puts all of this at risk. It puts journalism at risk and the funding that we're able to provide those newspapers.

(Dave Pickles at 00:22:33) So we have a lot of responsibility to do the right thing.

(Joel Beasley at 00:22:36) I found something new last week. It's called—it came through as an ad, and it was called 1440. I don't know what the—I think if you could Google it, it'll come up. But it's this concept of unbiased news. So they have a story on their very simple website. You can sign up, register to get the newsletter. It's a daily thing, and they just put the stories, but it's—they remove the bias from them. And it's just written like journalism. It's amazing when you're reading it because it's like a glass of fresh water when you're in the desert. It's like, where has this type of content been?

(Joel Beasley at 00:23:15) And it's a free newsletter, and I'm only a week into it. Right? But it's so interesting how they saw this gap in the market of, like, every news piece has just become the talking head's opinion, and it's like, where is the news being reported? And so I really am enjoying it so far, but it's amazing about the clarity that it can bring you to actually just get a set of facts or get a set of the best knowledge we have today that's void of jabs from either side of the political system.

(Dave Pickles at 00:23:51) I mean, journalism is just super important. And even if you don't like what somebody's saying or you think they're biased, supporting that and making it possible is the foundation of our entire system. So when it becomes hard for a newspaper to make money and they start laying off reporters, that's a bad thing. So that is something that we talk about a lot is how do we make sure that—it's hard for all of the newspapers in the country to collect online subscriptions. It's hard for everyone to get people to subscribe because there are certain people that want things for free, like you and your wife.

(Dave Pickles at 00:24:28) That's a very real thing. I remember reading a statistic that in Spotify—so the Spotify free, Spotify premium, Spotify free is very aggressively de-featured compared to the premium version. It's almost for me, it's really difficult to even use it versus Spotify premium to me is one of my absolute favorite products ever made. Right? The chasm between those is enormous, but something like 80% of their users opt for free.

(Joel Beasley at 00:25:03) Yeah.

(Dave Pickles at 00:25:04) And if you think about that from a newspaper's point of view, if they can ad-fund effectively and not have a ton of ad load that distracts from the user experience and doesn't have strangely biased ads laid down—there's some challenges you have to solve, but it makes it so that they can fund their operation much more efficiently than they can asking consumers to sign up for yet another subscription. There's a huge amount of subscription fatigue out there. And so now, if you think, "Oh, well, so Netflix and some of these other services, your Hulus, those are a couple extra subscriptions you could say that might be taking away some people's willingness to subscribe to the Wall Street Journal."

(Joel Beasley at 00:25:46) And also, you know, they're not really writing stuff people wanna read because there are individuals out there that are making six figures, and they just host their own thing that they're writing about. You know, there's—I can't think of any of the names right now, but I followed a couple people that do that. They write their own updates, and they do their own journalism, and they're making money. It's just that the model shifted and the ones that are going out of business, they just weren't creative enough to keep up. It was like an institution that was put in place like a chain or something, and they just didn't keep up with the creativity and the market shift. It happens every day.

(Dave Pickles at 00:26:20) Yeah. I mean, that's fair, but there are a lot of journalists out there doing a lot of hard and good work that work for newspapers still, I would say.

(Joel Beasley at 00:26:27) Oh, yeah. No. For sure. There definitely are. I was just in my mind when I was talking about this, I was thinking about my local newspaper, you know.

(Dave Pickles at 00:26:35) Oh, okay.

(Joel Beasley at 00:26:35) And I was just—I have no interest in reading the content that they put out. It's just for me, it's not that interesting.

(Dave Pickles at 00:26:43) I don't know. So our local paper here is actually interesting to me at least because there are things happening here that are never going to be reported anywhere else. There is no outlet to talk about the controversy about the dam that we have up above Ojai and how the city down by the ocean has sued the city up on the hill because they got sued by the federal government, the Coastal Commission, to say that you're responsible for making sure that the fish can get upstream, and then they go sue the people up there to say, "Well, it's your dam that's causing the problems." Where would I ever hear about that?

(Joel Beasley at 00:27:20) That's true.

(Dave Pickles at 00:27:20) There is no forum for that. And that's fairly interesting. And then also this pandemic stuff, I'm like, "How do I get more localized information?" That's the most important thing to me is, like, "What's going on here? Where did our cases come from?" And there's no—Bing is kinda trying to roll some stuff up, but Bing is linking out to all of the local newspapers that are actually going and doing more in-depth reporting.

(Joel Beasley at 00:27:44) Yeah. So I was just talking bad about my newspaper. In all fairness, you know, I try to be as transparent as possible. I haven't read it in a long time. But when I was trying to get into a news source, because it was just something—in my mid-twenties, like 10 years ago, I was kinda interested in getting involved more.

(Joel Beasley at 00:28:04) And I just remember I went through a couple local papers, and they seem to all be just rehashing of what's in the national news. And I was like—or the front several stories were that. And had I opened it up and it would have been like, "Oh, you know, Bradenton and Sarasota are rivals," did A, B, and C, I would have probably been way more hooked. I also live in a smaller town. I mean, you're out in—

(Dave Pickles at 00:28:30) Yeah. Ventura. It's about an hour outside of LA.

(Joel Beasley at 00:28:33) Okay. So why did you choose that place for your headquarters?

(Dave Pickles at 00:28:37) I mean, so because it's where I live. Yeah. So this is just where I wanted to be. So it's sort of the other way around. I built my company where I am.

(Dave Pickles at 00:28:47) If there is a strategic side to it, it's that I wanted to be in my life and in my company removed from Silicon Valley culture a little bit, or even Silicon Beach culture. You know, if you're in any of these sort of tech hubs, I feel like there's a lot of turnover and there's a bit of a mercenary culture where it's totally okay to work somewhere for eight months and go to the next place, and everybody's totally fine with that. And for this thing that we're building, this is really esoteric. Like, it takes me eight months to get you productive if I hire you as an engineer. And if people are hopping, it's going to be really damaging.

(Dave Pickles at 00:29:24) And then on top of that, I just have a really strong network of people here that I trust. And so I felt like we could get our critical mass without having a whole lot of churn while we went through the most critical stage of the company, which is like ideation, getting to market, and gaining profitability. Like, that's to me, in my worldview, those are the big milestones. And I thought we could do that sort of with mostly our local crew. And then I had one of my local people move to Boulder, Colorado during the first-ish year.

(Dave Pickles at 00:29:57) And, you know, sort of like, what does this mean for my career? And I was like, well, it means we have a Boulder office now. And so we opened the Boulder office with just a few people, but that caused us to get good at working remote. And so after that, it made it so that, you know, we could have engineering teams wherever we want. We got good enough at it that we now do—we have a San Jose office.

(Dave Pickles at 00:30:17) We have a San Francisco office. We, you know, there's some great people there we want to hire. We also have Seattle. We're in New York. We're in Sydney.

(Dave Pickles at 00:30:23) We're in London. Like, so we're sort of like creeping around the world. We're in Hong Kong now with an engineering team. We're in Shanghai with an engineering team. And so that's been—it's all been a really, I guess, happy accident in a way.

(Dave Pickles at 00:30:38) Like, I can't say I perfectly saw that strategy from the beginning. But it's been really good to go into a bunch of smaller markets, build awesome teams, and then get good at remote is, I think, a really good strategy instead of, you know, going and posting up in Foster City and hoping to peel off Facebook employees is a little more of an uphill battle, right, as a startup that nobody knows. Like, for a while, no matter how cool your company is, nobody knows who you are. And so recruiting is difficult.

(Joel Beasley at 00:31:10) How did you pick the name?

(Dave Pickles at 00:31:11) So we batted around, and it was like, it was Ad Trade Desk, Ad Data Trade Desk. We just really had this concept of trade desk as being the key thing that we want to do, and then we called it The Trade Desk. Sort of like, it's not a trading desk. It's the trade desk. Like, you know, I don't know.

(Dave Pickles at 00:31:28) A little bit of a line in the sand, I guess. But yeah, it's just what worked. And then you do that before you even found the company. You kind of have a vision, you pick a name, and then lots of companies change their name later on. And we talked about it in like year four.

(Dave Pickles at 00:31:46) Why don't we do it? You know, because there's this—there are these things at agencies called trading desks. There's trading desks in the financial market. It's a little confusing. Why don't we do a new name?

(Dave Pickles at 00:31:55) But I just think like, enough—by then, we had enough of a brand build, enough people knew who we were. And it's like, it's a step backwards, and I don't really see the benefit. And so we just decided to stay the course. I think a lot of company renames are pretty lateral. Right?

(Joel Beasley at 00:32:11) Well, it's like, let's go and give us and find that certain color palette and yeah.

(Dave Pickles at 00:32:17) Okay. But so we just sort of ran with it. It just is, and now it's just us. It's TTD to us more than it even is The Trade Desk most of the time. It's just us.

(Joel Beasley at 00:32:27) So did you have any like specific names that you were kind of leaning towards, or do you just like completely put that out of your head?

(Dave Pickles at 00:32:34) We came up with a big list and nothing worked for me, and so I didn't keep them anywhere in here.

(Joel Beasley at 00:32:42) So you seem like a pretty disciplined individual. Would you say that's true?

(Dave Pickles at 00:32:47) I'd say I'm driven.

(Joel Beasley at 00:32:51) Driven?

(Dave Pickles at 00:32:52) I like—yeah. Like, once I decide I want something to happen, I tend to get pretty focused on making it happen. You could call it obsessive. It's like a sort of kinder, gentler obsessive.

(Joel Beasley at 00:33:02) Yeah. Obsessive got a bad rap, and so we had to find like a different way to say it.

(Dave Pickles at 00:33:09) Well, the obsessive can be crippling for people, so I don't want to use that word lightly. But, you know, I think a lot of engineers have a little bit of that in them.

(Joel Beasley at 00:33:17) The advanced focus.

(Dave Pickles at 00:33:19) Yeah. Okay. Yeah. Like, once we get our mind on something, it's sometimes hard to have us think about anything else.

(Joel Beasley at 00:33:27) And honestly, that's what drives success. I mean, your ability to focus and have, you know, tunnel vision and achieve the result is—I don't know. I think it's an important—it's gotten me, you know, pretty far in life.

(Dave Pickles at 00:33:41) Yeah. Yeah. It's super important, and so is long-term thinking. And that's another thing that I would use to describe myself, as I usually try to think about like how it's all going to play out, and then let's go in that direction. And then you can kind of suffer the wounds along the way a lot easier if like you're not just thinking about the thing right in front of you.

(Dave Pickles at 00:33:59) And so, you know, it's a marathon, not a sprint if you're going to build something big. And you need to be okay with failure. You need to actually kind of like have a good relationship with failure. And just if you fail, you learn. Keep moving.

(Dave Pickles at 00:34:11) And, you know, those things go together to make it so that you're kind of like using every day to create momentum towards a goal. And if you can just do that consistently over time, you get where you're going.

(Joel Beasley at 00:34:25) Do you have any little things that you know, heuristics or ideas or frameworks that help you? For me, I'll give you an example. Like, I started—I just finished about two years ago, I finished my first five-year plan. And so like I'm into my second five-year plan ever. But before that, I had like no plans or I had like a monthly goal or I had like a wish list.

(Joel Beasley at 00:34:47) And then I, you know, I'd heard some—I got into the personal development space and heard people talking about this. And I was like, alright. Well, let me create a five-year plan where I can do like the James Clear Atomic Habits and like say you, you know, be very intentional about the habits I have and how I spend my time throughout my week and how that aligns to where I'm trying to go in five years. And it is very rewarding.

(Dave Pickles at 00:35:11) There was a time in my life—so when I was at Microsoft, actually, like I needed that job. And like I had a fresh mortgage and like, you know, I was out of work, and I'm in a small market. Like, it had to work. And so I got to the point where I was like breaking my day down into fifteen-minute segments. And at the end of fifteen minutes, I was like, did you use that as well as you could?

(Dave Pickles at 00:35:33) And you can't do that in the long term. Like, that's a level of intensity that I don't think is sustainable for almost anyone. But it is a good thought process to say like, on whatever grain I can handle without stressing myself out, is that the best thing I could have done with that time? It's just a great thing to think about and to course correct.

(Dave Pickles at 00:35:53) I think bigger than that though, like that was when I needed to really push and that is something that I still think about but much more macro. I like to think about what is the one important thing about solving any problem because, you know, when you look at a problem and if you come up with a solution, it's got like 20 facets to it. That's pretty fragile. Right? Then and if you're getting all wild and—

(Joel Beasley at 00:36:17) You're figuring it all out.

(Dave Pickles at 00:36:17) And some people, some engineers in particular love that kind of thing. Like, I've thought of everything, every corner case, I locked it all down and it's all figured out. Right? But I've found that those solutions tend not to work and that if you can step back and think about what's the one thing that's going to change this game and make sure I get that right, and then just don't even bother with the rest of it, and then move on to some important thing for some other component. So, you know, much more pulled back macro view to solutions, I think is way more effective.

(Dave Pickles at 00:36:46) And learning to let go of some details, having confidence that you got the right thing right, I think is what drives a lot of—it's what drives a lot of sort of agile methodologies and success of a lot of companies that adopt them.

(Joel Beasley at 00:36:58) So what does your team look like today?

(Dave Pickles at 00:37:02) Big. You know, before this, my biggest team was seven people. Now I think I've got about 400 reporting up into me. But what we've done starting with like when it became too big to be one team, right then, I said like I really think I believe in scrum or kanban or whatever you want to call it.

(Dave Pickles at 00:37:25) Like, the concept that you take your overall problem and you break it into pieces that are discrete, that can be, you know, executed on truly autonomously, and then you staff a bunch of people onto it, and you get out of their way. You know, give them their resources. It's a PM, it's a dev lead, and it's X devs, and just go. And you know what to do. And talk to me if you think I'm wrong about what we're doing or if you need clarification.

(Dave Pickles at 00:37:49) Like, that is the total engagement of senior staff with that team. I think it's super effective. And so we've tried to stay as close to that as possible. As we've grown, we've changed some things from that, but we typically have 12 to 20 scrum teams running at any given time. We have come back in and changed some reporting structures.

(Dave Pickles at 00:38:11) So about the time we got to 200 devs, we decided it was really important to make sure that the most junior developers were working on the same thing as their boss because otherwise, it's just so hard to grow. If you come straight out of college and your boss is managing a bunch of people that are working on 10 different projects, there's just a loss there that we needed to come back and get. But that's really the only change we've made other than, you know, changing what the teams are in a really organic way over time. So we've had like a monthly planning meeting for ten years where we come together, we decide what we're going to do, and we disperse and go after it. And it really works.

(Dave Pickles at 00:38:53) It's in a lot of ways a lot easier than the top-down lazy approach to engineering management, but you definitely have to be okay with a loss of control. You have to have a good relationship with failure because there will be mistakes.

(Joel Beasley at 00:39:08) How do you ensure that you always have like top talent coming up through your organization?

(Dave Pickles at 00:39:14) Well, so the way I always talk to people about our company was I want to—if you would be the CTO at some startup, that's who I want working for us as engineers. That—and so I want to create an environment where people like that can thrive. And so people like that thrive on autonomy and on impact, and that's the same thing that I want. Right? So now our incentives are all really well aligned.

(Dave Pickles at 00:39:39) And I need to—I need to go find you a challenge. I got somebody that's bored. Go find them a challenge. Great. Alright.

(Dave Pickles at 00:39:44) That's an amazing problem to have instead of like, I have people who I've given something to, and they're like, oh, I don't know how or I don't care enough. Right? So you optimize for those things. And so we really just—it's really a matchmaking process. Interviewing is matchmaking.

(Dave Pickles at 00:40:00) Right? We've constructed an interview process that if you enjoy it, you're one of our people. If you don't, it's very unlikely that you're going to be successful here.

(Joel Beasley at 00:40:12) Tell me a little bit. Yeah.

(Dave Pickles at 00:40:14) Yeah. So we do some basic things to do vetting, but then as soon as we're like, hey, this could be good, we get into a coding exercise that keeps getting published on the Internet. I want to change it.

(Dave Pickles at 00:40:26) But, you know, it's a coding exercise that doesn't have an obvious solution or there are several solutions and it's pros and cons between them. There's not like one gold star. And so you do your thing, you send it in. We say, do your best work. Take however long you want, send it to us so we can see what your code looks like and, you know, sort of like what tools you use.

(Dave Pickles at 00:40:48) You get a little sense from that. But then, unless it's just terrible and we end the process there, we get into a code review with a manager. And people who enjoy the code review process and when the engineering manager says, hey, did you think about doing it this other way? And they're like, oh, that's a great idea. I didn't think of that.

(Dave Pickles at 00:41:06) Like, that's something that says like, hey, you know, that's the kind of person that we want to work with. You're intellectually curious. You're open. And now I'm giving it away. Everybody knows how to ace the test.

(Joel Beasley at 00:41:18) If they're smart enough to track you down and listen to this episode, they already have some points.

(Dave Pickles at 00:41:23) That's right. That's right. I'll let it go. I'll let it go. You've earned it.

(Dave Pickles at 00:41:27) You know, I do think it's hard to fake though, like, because we have a really deep intellectual conversation. And then on the far side of that, we get into interviews. So it is harder for us to hire than I think a lot of other companies. With that, it's slower, but we've been able to find people that have had really long-term success with us and for us.

(Joel Beasley at 00:41:46) I like that process. I haven't heard that before. That's pretty unique. And, you know, to the people who don't fit that profile, they wouldn't even want to try to take your test because they wouldn't want to be in that world. They don't want to be in that environment.

(Joel Beasley at 00:42:00) It's people who are attracted to it, and they're like, that's me. I want to go there and take that test because I believe that I fit into that mold.

(Dave Pickles at 00:42:08) Yeah. Let me show them what I got. Yeah. It's like that's a first marker of what we're looking for is a little bit of fire.

(Joel Beasley at 00:42:17) Yeah.

(Dave Pickles at 00:42:18) But then it's really—I think the biggest thing that we get out of that is we screen defensiveness. I think defensiveness is the worst characteristic you can have as an employee, but particularly as an engineer, when what we're trying to do is facilitate an open exchange of ideas and everybody building on each other's ideas and having that, that, you know, that great flywheel. If you have somebody that's like, you know, oh, my way is the only way and you're wrong and they're dug in. It just has a much bigger effect on the team than even just them performing badly.

(Joel Beasley at 00:42:53) Yeah. That—it's like that defensiveness is not good, but I'm trying to think about like how they grow. I used to be like not a great person. And so I come from being like not an awesome person, and then I heard this guy speak and I think it was a YouTube video or something, but he had said the most frustrating thing in life is expecting above average results without being an above average person. And when I heard that, you know how some things just stick in your mind?

(Joel Beasley at 00:43:24) That was like a slap in my face in my early twenties. I was like, I have these results that I want, but I'm not that person. I'm not taking those actions or living my life that way. And so I just changed everything about myself systematically, consistently over time and just improved in all the different areas. But yeah, I guess the movement I made was like from defensive to curious. You know?

(Joel Beasley at 00:43:50) Like, oh, why did they interpret it this way? Or like, I'm curious to know more about this. And yeah, that's I think that was a big transition for me too. And I like it because I am a curious person. I just didn't, I think even before, I didn't even know how to express it, you know.

(Joel Beasley at 00:44:07) So I guess I was defensive because I didn't understand myself. That's, there is something interesting. Yeah.

(Dave Pickles at 00:44:13) Your defensiveness was a bad habit that you didn't know was hurting you. Yes. Right? And so if you can get people to stop doing that, then you're doing everyone a huge favor because like, defensiveness is never productive.

(Dave Pickles at 00:44:26) You know, our VP Engineering talks a lot about growth mindset.

(Joel Beasley at 00:44:30) Mhmm.

(Dave Pickles at 00:44:30) Like, whenever you say, you know, that person can't do whatever, just put a yet on the end of it and whether that's you or them. And if you just think about it that way, it's like, woah. So what activity do I have to create to change the way I think about that or to help that person, you know, do better, you know, whatever I've identified as what I think their weakness is, and we'll see if they agree or not. Yeah.

(Dave Pickles at 00:44:54) If you can get into those habits of growth, it can be super powerful.

(Joel Beasley at 00:44:58) I heard that they're actually teaching growth mindset at some elementary and middle schools.

(Dave Pickles at 00:45:04) That's good.

(Joel Beasley at 00:45:05) Yeah. I heard about it. I'd asked someone. I was like, this is amazing. I love these concepts. I was like, how do we get this stuff to our kids early? Because I have some little ones and he had said, I think he was in California too, he said, yeah. My kids came home and asked me if I knew what the growth mindset was and they were in elementary school, so they're already learning it. I said, that is brilliant. Could you imagine what life would have been like if we were taught the growth mindset in elementary school?

(Dave Pickles at 00:45:30) Yeah. I mean, we all got there eventually. I think our generation is doing an okay job of getting there, but it can certainly help. They're also teaching coding in all those schools, which is amazing.

(Joel Beasley at 00:45:40) I know. I know. They have the coding on the tablets. They have like the coding playground that comes installed on some of the tablets, and it's just, it's so amazing. I guess that's why I'm also a freemium person is because the first several years of my programming career, I was only learning from free resources. And then I paid for a resource, and I was blown away by how much better it was paying for premium content. And I'm like, wow. You can learn a lot faster and move yourself forward quicker by going with the premium stuff.

(Dave Pickles at 00:46:09) You know, one of the reasons I work hard as an engineer is I'm just so angry that things don't work better. You know, when I grew up, it was like the year 2000, there's gonna be flying cars and, you know, you're gonna talk to your microwave and it's gonna do the right thing. And none of that is true. None of that is remotely true in 2020. And it's just so frustrating because I think those things can be true. And so I think it's great that we're teaching coding and engineering in schools because there is so much work to do if we're gonna achieve all those things that I think we all want and that I think can, you know, have huge impacts on society. Like, there's a lot of work to do. It's, you know, it's the 80/20 rule. Right?

(Dave Pickles at 00:46:53) We're not even to 20 yet on the 80/20 rule, and there's all that work left to do to make basic things just work together right. Like, you know, there's a couple of wins. I'll point to a couple of wins. USB. USB is a huge win on making things work together well. Like, the other day, my power went out, and I plugged my phone in with a USB to charge it, and I turned on its hotspot to power my Wi-Fi. That's how technology should work.

(Joel Beasley at 00:47:17) Yes.

(Dave Pickles at 00:47:17) Right? And it's so rare. I have so few examples of things like that where like, companies, different companies made things that really work together perfectly, like the way you'd expect right away. That, oh man, don't get me started.

(Joel Beasley at 00:47:31) That is true, especially when it comes to hardware.

(Dave Pickles at 00:47:34) It's true of everything. Oh man. If you get into my world, into platforms and making platforms try to work together, oh dude. It's not even, that's not even hard interoperability, and nobody bothers building the APIs to make it work.

(Joel Beasley at 00:47:47) Oh man. I think there's a, I think there's an Alexa-powered microwave. I wonder if that exists. And then maybe I'll get that to tell your wife about that and then you'll be, for your birthday, you'll be able to talk to your microwave.

(Dave Pickles at 00:47:59) Well, but how well does it work? Because, you know what? I just, I'm sure there's a microwave that will take some sort of voice instruction.

(Joel Beasley at 00:48:10) Imagine, like the future, where like the, you install like the fat and protein cartridges and it prints your food. You know, that would be pretty cool.

(Dave Pickles at 00:48:21) I'll take, I'll take heat a cup of water until it boils.

(Joel Beasley at 00:48:25) Yes. That's what you'll take? All right. I think Alexa could do that.

(Dave Pickles at 00:48:29) I don't think it can.

(Joel Beasley at 00:48:30) You don't think it can?

(Dave Pickles at 00:48:31) I think it could microwave it far past the point where it boils. I very much doubt that it has the right sensors to detect the boil.

(Joel Beasley at 00:48:38) Oh, but if you know the current temperature and you know, you know the current temperature through geolocation, right, because you know where the microwave is. Right? Well, then you'd have to, you'd have to be hooked up to like your Nest to know what the temperature is inside. You could probably get pretty close by knowing the temperature of the environment and then knowing how long you have to heat to make...

(Dave Pickles at 00:48:59) Altitude, mass of the cup, the dissolved solids in the water, like, there's variables.

(Joel Beasley at 00:49:04) There's variables.

(Dave Pickles at 00:49:06) What you really need is a thermal sensor.

(Joel Beasley at 00:49:08) Yes, you do.

(Dave Pickles at 00:49:09) It's the one important thing.

(Joel Beasley at 00:49:11) I was trying to get around that requirement.

(Dave Pickles at 00:49:16) That was a great example of what we're talking about earlier.

(Joel Beasley at 00:49:19) Yeah. You need, if you just had the thermal sensor, that solves everything. It's like the one thing. So you need an Alexa-powered microwave, thermal sensor, and then the software ability for her to connect that.

(Dave Pickles at 00:49:33) And then we'll build a machine learning model to understand the size of the object and the way that the heat radiates through the object because you're only gonna get a surface reading off your sensor.

(Joel Beasley at 00:49:44) What's your tolerance for how close the time has to be to boil? Because like, does it have to be persistently boiling for more than like three seconds? Like, what do you consider a full boil?

(Dave Pickles at 00:49:55) I mean, we're getting detailed into requirements now. Yeah. I would say I want a full rolling boil sufficient to make black tea without evaporation to the level that would ruin my cup of tea.

(Joel Beasley at 00:50:10) Oh, all right.

(Dave Pickles at 00:50:13) So I need like, I need like five milliliters max evaporation and a rolling boil.

(Joel Beasley at 00:50:17) There you go. We're doing it. Get the interns on it.

(Dave Pickles at 00:50:21) Ship it, Amazon.

(Joel Beasley at 00:50:23) Ship it right now. Stop the pandemic response. Let's get this Amazon microwave out there to the world. No. But it is gonna be a long time before all of our devices are connected seamlessly because now I'm sure there's like a smart fridge where you can get something connected and there's a small benefit, and I like to see it moving forward. But much similar to you, there is like this underlying drive slash frustration that I desire that I want. I want the future to come faster. I look forward to that. Personally, I hope that we reach, I hope that Neuralink-type technology allows us to persist our consciousness prior to my expiration. You know, I think that'd be pretty cool.

(Dave Pickles at 00:51:07) I don't know about that one.

(Joel Beasley at 00:51:09) You don't know about that one?

(Dave Pickles at 00:51:10) I just don't know if it's gonna get there. I think the way we started that conversation, we were talking about coding in schools. And so I think that is, that is the way to think about it makes me optimistic, which is I think there's a tremendous amount of work to do that is powerful. And we can teach all of the kids to do it. So we can have women and people of color in our software engineering workforce at much higher levels than we have today and that, you know, the strong wages and the sort of the flexibility of work from home and all the cool things that come with tech work can be super democratized, especially now that we're kind of like emptying out San Francisco in the pandemic and everybody's working from home. I think those can be super positive things for the whole world and just for, you know, sort of helping do our little part to improve equality.

(Joel Beasley at 00:52:03) Yeah. It's important. I wanna talk about a couple things about your like, one-on-ones and, you know, high potentials. Like, how do you approach one-on-ones?

(Dave Pickles at 00:52:14) I'm not the best at this. So like, that's why we have a VP Engineering who is. My one-on-ones tend to be conversational but focused on like the things that we care about. So like, number one, the one-on-one, I wanna connect as a person. I wanna like, have at least a moment to check in on people and know how they're doing. I think that's an important part of work that you can't ignore. You can't be all business all the time. You know, but then mostly what I wanna do is talk about principles and the things that we're trying to do at the highest level. What really are we trying to accomplish rather than, hey, you know, when's this gonna get done and anything like that. And now, especially, my job is to lead by example and to try to help all of our managers not be micromanagers, and instead to talk in principles to help reinforce the vision of what we're trying to do, and then, of course, to deal with whatever's going on in people's careers and make sure that I'm opening up opportunities. So I very much believe in servant leadership in engineering management.

(Dave Pickles at 00:53:25) Management is a different job. It's not a better job. It's not a job that's right for every person. And somebody once told me, like, one in ten engineers makes a good manager. I think that's overstated. Just because it's very different and you really do have to be focused on the success of the people around you more than yourself to be a great manager. You can be an okay manager without that, but there's no great manager that doesn't have that mindset.

(Joel Beasley at 00:53:50) Yes. I agree. I made the transition out of the desire to or out of the awareness that I can only write so much code and I want to make changes to this world. And if I can learn these skills of management so I can work with people to achieve a result greater than myself, then I can have a greater impact. And that to me was like, you know, I wrote code every day for 17 years. And, you know, up until the podcast got more popular last year, and I stopped writing every day for work. So it was like a big thing. It was a huge part of my identity, the fact that I was an engineer and wrote the software. And I felt like every step away from that was like losing part of that identity, so I had to find a way to like think about it and put a thought pattern behind it and that concept of I can accomplish more. Then I realized once I got into that perspective, I realized that is really what drove me to be an engineer in the first place.

(Joel Beasley at 00:54:55) That first time I wrote code that, like, caveman moment, they're like, oh my god. I did that. I actually impacted the screen. I did something. I made something happen. I performed a calculation. It was just so cool to me, and it was that feeling. And I said, wow. That's like this underlying driving factor, so I'm gonna tap into that. And then that's how I managed to make that transition and be like, maybe the one in 20. Right?

(Dave Pickles at 00:55:19) Yeah. I often say, like, to me, software languages are like hammers and nails. Like, I don't really care about hammers and nails. I wanna build houses. And you can't build a house by yourself, let alone a skyscraper.

(Joel Beasley at 00:55:32) Dude, that was amazing. That was like mic drop. I really like the way you described that. Yeah.

(Dave Pickles at 00:55:37) Somebody's asking me whether I like Python or Perl better, I'm like, you're asking me the wrong question.

(Joel Beasley at 00:55:44) Yeah. Do you like two stories or skyscrapers?

(Dave Pickles at 00:55:47) Yeah. Right. What are we building? Yeah.

(Joel Beasley at 00:55:50) That is so true. That is actually I really like that perspective. I'm gonna make a clip of that. That sounds good. Yeah.

(Joel Beasley at 00:55:57) And Dave, anything else we didn't cover today?

(Dave Pickles at 00:56:00) What didn't we cover today?

(Joel Beasley at 00:56:04) Anything you wanna get out to the world? Careers page? You have a careers page? Some of the people that like your style, maybe they wanna go check out more and go through that interview process?

(Dave Pickles at 00:56:13) Yeah. Yeah. We do. Trade Desk Careers for sure. So, yeah, I mean, anybody out there that's listening, I think, hopefully, the main thing you heard from me was about the way to do modern software engineering and to embrace that world, to embrace remote, to embrace whatever form of Agile works best for you and to embrace an employee-forward management style. Like, that's something that I think people don't really understand is how much you get for what you give when it comes to working with your teams and working with your engineers. And if you put people first, you get way more. Right? And so even if you are only focused on what you get, the right way to achieve that is to put people first, to invest in their careers and invest in them, and people will always remember that. And they'll work way harder. You know, some people were down on millennials for a while and hopefully people have figured out that that's totally wrong because all millennials want is a mission to work for.

(Dave Pickles at 00:57:16) And if you give them something they believe in to work themselves to the bone, you know, then, and that's what they thrive on and that's what their success thrives on. So just understanding that, I think, actually helps you understand all people. That's not just a millennial thing. That works better on everybody. It's just, you know, help them understand something that is exciting. Don't try to make something exciting. Give them something that is exciting and give them the tools to go build it and all the success will flow from that.

(Joel Beasley at 00:57:44) I love that. There's this concept or this book about like meeting 100 people, and it's this idea that, and I didn't read the book for disclosure. Somebody was telling, I was talking, I was giving a talk, and I think like in Georgia, and somebody told me about it. And basically, when you go out and you meet like a hundred-plus people and get to know them, all of a sudden, color, sex, age, all of those things fade to the background and it all becomes this segmentation by mindset, by like how people think, these ways people think.

(Joel Beasley at 00:58:16) And so when you get with them, you can kind of figure out, okay, these people think this way and you can identify with that, you can know how those operating procedures go. And to your point exactly, with the millennial stuff, once I was giving this talk, it was actually in Georgia, I think. And it was to an older group, and they asked me about, like, one of their questions in the Q&A was, like, why are millennials, why do they suck or, like, why are they distracted? And I said, well, I think in any group of people, there are some people who suck. And maybe you're just running into those. But we have this obsession, I think, as a species of really amplifying the exceptions.

(Joel Beasley at 00:58:59) I don't know if you've picked up on this, but I have. It's like, oh, things are mostly this way, but then it's one time that way, and then we try to amplify the one time it's different. And I think that that's a little bit noisy, and it kind of is a disservice to us when we're trying to wrap our minds and create models of the world. And it's just something that I picked up on in the past couple months that I noticed happening a lot. Have you noticed that?

(Dave Pickles at 00:59:27) You know, the way I think about the millennial thing is different than that, which is that there actually has been a big change in the way that people respond to authority in particular, and it's just generational. Like, if you, a successful manager at IBM in the 1980s had a certain style. That style doesn't work anymore on anyone. And so it kind of started with younger people, and this is how the world works, right?

(Dave Pickles at 00:59:55) Younger people have a little different thing than the generation before them based on their experience, based on their reaction to what came before them. And they're just not gonna respond well to, you know, the sort of browbeating, tell them what to do management style. They have to be led. And so it creates an interesting dynamic where now you have a bunch of older people in positions of power that now have to adapt. And I think that's where a lot of the weird rhetoric comes from, is that there's people that don't wanna adapt, don't think they should have to, think other people should adapt to them.

(Dave Pickles at 01:00:29) And that's just not how the world works.

(Joel Beasley at 01:00:33) I like that. That's pretty interesting. You give me a lot of stuff to think about, by the way. I like your perspective because I really value it, and it's like, whenever you respond, it's something I didn't think about, but it's slightly different. And that doesn't happen a lot.

(Joel Beasley at 01:00:49) So I like you, my friend.

(Dave Pickles at 01:00:52) Awesome, man. Well, thanks. This has been great.

(Joel Beasley at 01:00:54) This is awesome. Yeah. And if you ever need anything like introductions to past guests or anything at all, never hesitate. Just reach out, and we'll hook you up with whatever we can do.

(Dave Pickles at 01:01:04) Very good.

(Joel Beasley at 01:01:05) Thank you so much, Dave. You have a fantastic day.

(Dave Pickles at 01:01:07) Thanks. You too.