Feedback Loops and Speed: How to Optimize for Learning and Build Efficient Teams

Aydin Mirzaee 3:01  

Farhan, welcome to the show.

Farhan Thawar  3:02  

Thanks for having me.

Aydin Mirzaee  3:03  

There’s a lot that we’re going to talk about. But one fun thing that I thought would be interesting to start off our conversation was both, you know, obviously, Dan D. is a good friend of yours, mutual contact. And, you know, I messaged him, and I said, Hey, I’m gonna have Farhan on the show. What should I ask him about? And you know, one of the things that he said was, you have to ask him about his long reading list and how he does it. So tell me about your reading list and your reading habits? I don’t know very much about it. 

Farhan Thawar 3:34

Daniel was my co-founder. So like, Yeah, he’s a super intense, extremely sharp person. And the only way for me to work with him was for us to start a company together. So that’s a whole nother story, but happy to go into that. My long reading list. That’s an interesting thing for him to say. I think, you know, I do have opinions about like, what I call like a reading hierarchy. And the reading hierarchy trust, what I try to do, although I fail most times is I try to prioritize the longer standing reading from the shorter standing. So have you heard of Lindley’s law, Lindley’s law is that however long something has existed, it will likely exist for that much time going forward? So there’s a book that’s 20 years old, it likes likely be relevant for another 20 if it’s relevant for the last 100 it might be really relevant for another 100 years. So it’s this idea that long-standing things done with purpose, tend to stand the test of time, versus things that are more fleeting. So for me that typically means like I should pay for like blogs over Twitter, magazines, over blogs, and books over magazines. And so what I try to do is I do try to still read books I’ve now moved to like a Kindle, and the reason for that is one was just portability. When I was traveling, I used to have a fear of running out of things to read. I don’t know if that’s a real phobia, but I do have it. I was traveling a lot and I would always carry multiple books. And so while useful when you get stopped by TSA because they don’t want you to use electronics. It was getting making my bag heavy selection. To a Kindle. And what I do is, I do read a bunch at night, and sometimes on the weekends and then what I do is I highlight phrases in the books and then I use a tool called Read Wise to like playback those phrases like on a daily email, and it reminds me of the book and something that I thought was interesting. I also then import a bunch of other stuff into reading wise as well quotes are things that I want to think about. And so that kind of reinforces the memory of the things that I’ve been reading, because part of it is like you read it, but then you forget it.

Aydin Mirzaee 5:28

That’s awesome. So much wisdom right there. It’s over the course of time. So read wise will only send you things, you know, from a certain amount of time or kind of randomizes. What to randomize, okay.

Farhan Thawar 5:37

Yeah, random. I mean, you can do it’s pretty easy. Like I, it’s, you know, I know, I think even Toby’s got, like a public GitHub repo of a way to have your Kindle highlights sent to you every day. So I might end up just taking that code and modifying it for what I need. But so it’s not super complicated. But there is something to be said about having the ability to like, have your quotes or other things replayed to you over time. And I do I write a lot of things down like one funny thing, like somebody asked me, what’s a funny thing on your phone like I have different I have a different note on my phone for every restaurant that I go to with my kids with all the things that we ate, like, for example, that we liked, so that next time I go to like, you know, to lay down the street, I literally pull up and like, here’s what we had last time, and I write down, never order this. And so, you know, I’ve even got weird notes. Like, where I was for New Year’s Eve for the last like 30 years. I don’t know, I just write it down. It’s cool to kind of go back and be like, Hey, where were we like 2005? And I can look it up on my phone and be like, Oh, we were here?

Aydin Mirzaee 6:33 

Yeah, no, that’s awesome. I think that that kind of sets the stage for a lot of what we’re going to talk about I love that reference to Did you call it? Was it Lindsay’s law? Lindey’s law? Yeah. Yeah. That’s awesome. I’ve heard about that in the context of, you know, things like I guess the was it like the Berlin Wall? Like and some things like that. But that’s really interesting. I didn’t, I’ve never thought about it in the context of books. But I think that makes a lot of sense.

Farhan Thawar 7:00 

Yeah. And so what that would mean, though, in a strange way, is you should favor old book totally things that have been around a long time. And that’s not I mean, I’m not always reading old books. I’m reading the new Reed Hastings Netflix book right now. But I do tend to read even like, definitely a lot of fiction I’ve read is like, many, many years old, and I got started on my like mom’s bookshelf.

Aydin Mirzaee 7:18

Yeah, no, I think I think that makes a lot of sense. There’s a lot that we want to talk about with you today. And so obviously, you’ve had, you know, a lot of diverse experience today, your VP of Engineering at Shopify, but you’ve started a bunch of different companies, including Extreme Labe, you were at Achievers, you’ve been at Microsoft, you got acquired by Pivotal and had leadership roles there. Before we dive into a lot of the specifics and some of the lessons learned in those different companies. I just wanted to start things off by asking you who’s been your most memorable or favorite boss? so far that you’ve reported to?

Farhan Thawar 7:54

Well, good question. So I think there are probably different answers depending on where I was in my career. So one of the things I started doing recently was like trying to articulate to myself how I want to spend my time at work. And what it boils down to, really is that I try to optimize for learning. When I optimize for learning, I tend to gravitate towards bosses who have more of like a sink or swim mentality, meaning like, they just dropped things on you, and you figure it out. And whenever I’ve worked for someone like that, it’s been super fruitful, because we both figure out where the limits are of what I can handle and what I’m able to, you know, spend my time on working through. And usually, that leads to like creativity on getting to some of those crazy goals that get dropped on me. And whenever I’ve been in a situation where even if I’ve not had the task, relevant maturity, and somebody should spend more time with me, and it’s more, you know, you could use the word micromanaging, I tend to not do well, with a structure like I you know, maybe this is something to ask my previous bosses, but I don’t think I’m that easy to manage, which means that like, you want to be more, you know, higher-level like drop large goals and just go away versus like, Hey, can you do like these seven things in like the next seven weeks, and I don’t think I’m as good that way. So who are some people? I would say, you know, two come to mind right away. One is Ammar from extreme labs. And he was a classmate of mine from Waterloo. He called me an O seven to join. I didn’t join I went to achievers, but then in oh nine, I joined when it was pretty early at Experian labs about 10 or so people. The genius thing him and Sonny were the founders, the genius thing that they both did was they trusted me to try all kinds of crazy experiments. I don’t know why they trusted me because I had not really worked with them that much. I worked with, like Co Op with her. But other than that, and we tried so many things. And we not all of them worked, but a lot of a lot, but a lot of them did, like so for example, we were wondering whether you had to adhere to a strict one to 10 ratio between manager to employee and so we tested it by you know, continually adding more people and there was a there was a stage at extreme labs where I had 120 directors Oh my god. And we were not saying that it was working perfectly for like, depending on what person’s like their career goals were. But many people to this day messaged me and say, you know that time when we work together at extreme labs was one of the most formidable in my life. Because, you know, there was probably not that much time I could spend with each person. But there was a lot of responsibility that each team had to think about. They didn’t feel like the management overhead, there was something that was blocking them for their career, right. Like, don’t forget, we had a very religious environment where we had pair programming, structured hours, we shipped every week we demo it every week, we show clients our work and our feedback. So how to all the things that you wanted from a mastery, because you’re pairing autonomy, I’m not really around the to spend time with you because of 100. And money reports. And purpose. We were building mobile apps for that we’re going to be deployed on millions of phones, people felt like they were getting that and didn’t need me to like bog them down with a one on one actually, when I was doing one on ones that extreme labs, people tend to be like, Can I leave now? because there wasn’t much help that I was giving them. So I would say amantani had a bunch of genius insights around like letting us try things. And other one was, can you let interns do real work? A lot of companies we worked with were like, I don’t want insurance on my project. Are you sure like they’re the ones actually who do the creative thinking they grew up with mobile phones more than we did, right? extreme obscenely mobile apps. So I think there was some genius in that. And I would say, at Shopify, I’ve had, you know, two bosses so far. So teaching jml, both with their strengths, and helping me figure out where I should spend my time at Java by actually many ways, both together, because I, you know, have a little slack group where I pick them both. I think, again, there’s the genius and like just dropping large goals and saying, Go figure it out yourself. And then there’s also the like, Hey, you should really look over here, like spend some time in this area, even though you may not feel like it’s the right thing I feel. I’ll give you one example. One of the things I spent a lot of time in my career over the last 10 years or so was like engineering management and agile and really pushing the limits of like, what makes sense what doesn’t, right, I’m probably known as one of the top pair programming folks around who I spent, you know, probably read the biggest team there. That meant I got away from like some of the tactical work in engineering, like writing code. And so one of the things that, you know, JLM and I spoke about before coming to Shopify, and now here is that you should write more code. And what’s amazing about that advice is that a few things one is that 99% of people who would tell me that I would tell them to like, Fuck off, because I’m like, my leverage is here. But he’s got a very specific reason, right? He’s like, I really want you to understand the end the developer experience, I want you to gain trust with the teams by spending time there, I want you to onboard and talk to engineering candidates and still have that closeness. And so it’s, it’s interesting and amazing, because in most cases, I would say no to it, but because of how it’s being delivered, and the reasons behind it, I’m coding now more than I have in the last 10 years or so now, of course, not a lot, and you don’t want me coding lot. But at the same time, it’s reignited a passion in me that I haven’t felt in a while on the coding front.

Aydin Mirzaee 13:07

Yeah, you know, I think that makes a lot of sense. And I think ultimately, even that conversation could have gone two ways, but it but it sounds like if you can buy into the why of why you know what he’s suggesting? And if that makes sense to you, then then obviously, the rest will will follow.

Farhan Thawar 13:23 

Yeah, I didn’t, you know, to be honest, it didn’t make sense for a while, right. Like, there was a conversation where he said, I didn’t know how much to push you on this. And then and I did it, right, like I started coding. And so we got to a good spot there.

Aydin Mirzaee 13:35

Is it safe to assume that you’re very much I mean, you talk about people dropping big projects, I’m letting you figure it out? Can I assume that if I talk to your direct reports that they’d basically say the same thing about like, your leadership style? Or is that the type of leader you would describe yourself as?

Farhan Thawar 13:52

Yeah, like I try to be as I mean, I don’t know what the right term is. But I think what I’m trying to do is give folks space, like my role really, is to unblock, right? So if I’m in there with like, very tactical feedback, that’s either because we think that things are not going well, or I’m just trying to produce ideas, versus like trying to manage the team very closely. But what I’m trying to do is make sure that people are unblocked. And so whether that’s resources, they’re blocked on a, you know, getting to somebody that they have to it’s information, they don’t understand the priority. And we and a lot of those just mean that we haven’t done our job as leaders to set those things. That’s where I think it can be useful. Actually, one of the lines I use is that I say to folks, I can only help you go faster, not slower. Meaning if you pay me So here’s an example of something that actually happened in Shopify, we were chatting with a merchant. And one of my directors of engineering had written like a draft email to send to the merchant. And he had sent it to me and say, Hey, you want to look it over before I send it? And I was traveling, and so I saw the two emails simultaneously. One said, Hey, you want to read this draft, and then right after it, you know, many hours later It just I just was CC’d on the email to the merchant. Now, some managers might be like, hey, you asked for my feedback, you didn’t let me give you feedback. And you just sent it anyway, like, WTF. But this director, even though we hadn’t worked that, well, well together, like a much together, we just started a job fine. messaged me. And I said, and I said to him, Hey, good job, he sends it out. He goes, he goes, I have a field, I had a feeling that’s how you work. Meaning if I could make it better, or I could, I could, I could interject and give feedback in that email before he sent it. No problem. But if not, he’s sending it anyway. And that’s the way I like to like to work with folks, which is that you don’t want to be a gate. You don’t want to slow people down. And sometimes it can come across that way, right? Maybe, and maybe there is something super important. We were like, do not send until I look at it. But most of the time, it’s probably okay. It’s like Sam, if you give feedback, cool, if you don’t get feedback, send it. Right. Because you know, we’re all busy. And, you know, it also gives you this notion of if they’re going to send it and you’re not looking at it, it’s probably a little more care from the people who are doing that work, because they feel the danger of not having it like checked over.

Aydin Mirzaee 16:05

Yeah, I love that. It’s like you’re defaulting to speed. And it also doesn’t prevent you from, you know, giving the feedback afterward. Or if there was something really that you wanted to add. I mean, you could always reply all and you know, respond that way too. But I love how it just defaults to speed.

Farhan Thawar  16:21  

Yeah, and I think that the default thing to speed is for many reasons. One is that, especially when you’re interacting with others, right, whether it’s like writing code, whether it’s with partners, getting feedback on a product, all of those things, the speed helps you give to get more feedback, right? So the merchant replying to us would happen faster because we emailed faster, the feedback on the product happens faster, because no one has to check it and which means you’re getting high fidelity information back. So I think that speed is underrated like, I know, we talked about it, everybody talks about it. But I think that people don’t understand why it’s so important. It’s the loop. Right? The shorter you can make the loop, the faster you can learn. And again, I said this before, I focus on learning as my number one attribute that I look for that I spend my time on. And in order to be learning, you have to get feedback loops.

Aydin Mirzaee 17:08  

Yeah, I love that, like the cascading effect of like every, you know, waiting to send this email on Monday versus sending it today. All these things kind of just yeah, have this loop effect. As you said, I think that matters a lot. So I think I have another topic that is somewhat related. And that’s talking about the word know, you have a lot of opinions about this word know, from what I can gather. And you know, one of the things that you caution is it you know, knowledge is like a bullet, and you should use the word know, very wisely and not make a habit of it. Like, tell us about that.

Farhan Thawar  17:43  

Yeah, so the way I’ve been thinking about this is that, as leaders, you are expected to like have a few bullets career where you go in and you actually say you know what, I’m gonna step in and make the call on something. And the reason you only want to have a few per year is that you do want the team to not get used to you stepping in, but also feel like they have their permission to try things and learn from those things. And so the bullets should be super be used super sparingly. And that’s when you know, when I talk about the word No, I mean that going in and somebody wants to try something, it has to be your one or two shots per year where you’re actually overriding. And I think this is true for leaders in all in all levels. Because when you’re prioritizing something, I think that’s super important to get right. And I think that’s a great bullet to use. And that’s something that as a company can be good at communication perspective we should all be spending time on. But all of the cascading decisions like the No, it’s called like a choice cascade. People make choices at different levels, those choices if they’re not one-way doors most of the time, and they’re not, you know, the team should be able to handle it and learn from that in many cases, they’ll do a better job because you’re not in there because they have better context. And you’re closer. But there are times when you do have a very strong opinion or have context because you’re looking at something else that they may end up seeing, in which case you want to interject. I mean, I think there are there are multiple ways to do this. Actually, you know, there’s a great story in the new Netflix book where Reed Hastings comes across the director that is signing, like a five-year lease for some office in Singapore or something. And Reid was super confused because Netflix really cares about flexibility over certainty, meaning like they’ll pay more for flexibility, more pay more for optionality. And he was surprised that this director was signing a five-year lease for this office because that would reduce flexibility. But instead of correcting the director and his one on one, instead of saying, Hey, you know, you probably shouldn’t lock that in. Because, you know, we care about flexibility. And here’s why Reid didn’t do that. He just made a note. And in the next quarterly off-site, he ran an entire session on flexibility versus certainty. So read, you know, in reading that I was like, wow, we didn’t use a bullet there. He didn’t decide to like, tell the person No. Instead, he felt that the broader thing to do was to step back and teach the entire leadership team, why flexibility matters and why certainty even though it’s cheaper, is not what Netflix wants. And so, you know, that ties into this word know, where if you’re gonna use it, like, I’m sure there are times where Reed Hastings does, you know, say something. But you have to be very careful when you say it, because it is a bullet you want to use sparingly.

Aydin Mirzaee  20:28  

Yeah, I think that makes a lot of sense. And what alike, very interesting observation on that, which is, like, it’s almost from this, this meta-level of looking and saying that, as I could tactically correct one thing, but there might actually be a systematic problem here that like maybe people are not understanding this value in the right way. And like really looking at as at a system is actually pretty interesting.

Farhan Thawar  20:53  

Yeah, no, it’s amazing. It’s amazing. I think there are two things there. One is he stopped himself for intervening, which is super hard. That’s right, I would say, Yeah, super hard, I would have a hard time stopping myself. And then to, he made a note, to then spend, you know, let’s say two hours with every leader at Netflix, talking about flexibility versus certainty to help them understand the value. So they make all sorts of decisions in the favor of flexibility over certainty versus just correcting this one thing. So there’s a lot to unpack there. super impressive. unsure if like totally true, it is a book. But amazing.

Aydin Mirzaee 21:31  

[AD BREAK] Hey, there, just a quick pause on today’s episode to let you know that we really appreciate you helping us spread the word about the Supermanagers podcast, if you’re enjoying what you’re hearing so far, dial into your podcast app of choice, whether that’s on Apple or Android or Spotify. And just leave us a quick review. Now back to the interview. [AD BREAK OVER]. 

Yeah, yeah, I think the lessons stand nonetheless, one of the things that you talk a lot about, and particularly I mean, you’ve obviously hired a lot of people in your career. But one of the things that were really fascinating to me is that you actually hired 1000 people inside of four years, which for any company, even in modern-day high growth companies, I mean, that’s a lot. And so you had this really interesting way of hiring that again, like defaulted to speed, we’d love for you to explain that you know, like how you guys operated, how you hired, and a little about, like how that may have translated and what things look like at Shopify today?

Farhan Thawar  22:31  

Yeah, sure, actually, let me tell you a difference. So I’ll tell you first, like the speed portion of recruiting, and then I’ll go into what we did, specifically, there are two different stories. When we first started recruiting and extreme lab, we had a pretty traditional process, which is that we would have like two or three, one hour each technical screen interviews, and that would inform us as to whether we thought someone was a fit for Extreme Labs, the thing that I think was different was we still valued speed. So for example, all three interviews were scheduled. So when somebody would apply you know, the interviews were cascading meaning if you only if you did well, on the interview, one would you have two and three, right. And if you had done well, on two, you’d have three, but I think the key was, if you did well, on interview three, interview four, which was with me and was 15 minutes long, was where I gave you the offer letter. So we focused on speed of like, the people during the day would have the interview, and it would get the offer the same day. And what was amazing about this process was a few things. One, I would get like a text during interview three, or someone would be like, this person is doing well, my cool. And I had a standard, you know, offer template, I could look at a resume and come up with you know, you know, obviously, if you have an intern or new grad, you know what it is. And even with people with experience, we have a good feeling of how to do that the reactions of the people in the meeting, when I gave them the offer, was phenomenal. And it was exactly what I wanted them to feel was that we move quickly. And so I went into those meetings, and I would say, Hey, thanks for coming to extreme labs, Rebecca, like everybody like to, I’d like to give you an offer to work in extreme labs. And she would be like, what? And I would be like, here’s the I would have the generated, printed, everything in the details, salary, stock options, everything was there, signed by me. So what it gave, in this case, Rebecca, the notion of was this company moves fast, this company doesn’t have a lot of red tape, there’s not a lot of approval process that has to happen. And then Rebecca would be like, I wonder if the whole company works like this. And I would always say do not sign it right now. This is not like some sort of like pressure tactic. Take it home, you have, I think three to five business days to sign but and let me know if you have any questions, but I could walk through the benefits. I could do all this stuff. And it was amazing to do that the same day and it really felt like we were showing them our culture in the recruiting process. The other thing that it did was, anytime I had a candidate who was also interviewing at like Facebook, Amazon, it didn’t matter because they were like, this is better that process. They met with some folks. It was enjoyable. They got an offer. They walked away. And we had some ridiculous like 95% plus close rate. Because people felt like that was part of the culture. So that was my first part about recruiting as it was all about, again about speed. The second thing we learned was in our interview process, which was three hours of the time, we started trying to figure out if there was any fidelity coming through the interviews, the meaning was there actually, the correlation between the interview process and how they did on the job, as numerous studies have shown is that the connection is tenuous at best. And we tried it by just reducing our interviews. So instead of having three hours, we reduced the two and a half to one and a half, one, half an hour, 15 minutes. So we ended up at 15 minutes. And the reason was, as we kept reducing the interview time, and realizing we were not losing any accuracy in our predictions. And I would say our accuracy was about 80 to 85%. Now, that’s not like some secret interview style. what it was that we had a process, which then said, Okay, we’re gonna give everybody a very quick and dirty programming question just to see that they can program. And you know, you can imagine, you know, if you’ve read, like the Fizz Buzz program, or Fibonacci or factorial, like very easy, I would modify it slightly. So it’s not like, literally just off your memory, but you’d have to think a little bit, we just see how they did on that. And they would say, you know, what, we’d like to offer you a full-time job at extreme labs, the first 90 days is a probation period, like at any company in the world, we’re just explicit. And it’s really a time for you to figure out if you’re a fit for extreme labs, and we’re a fit for you. And at that, during that time period, if you don’t feel like you’re a fit, or we don’t think you’re a fit, you know, just like any probation period, we will let you go. Or you could quit, and there’s no, you know, no hard feelings, either way, what we did then was because we have this system set up with pair programming, and, you know, higher fidelity feedback, instead of the three hour spent to the interview process, that three hours was spent working because I would check in with them 30 days, 60 days, 90 days, I would ask the people they were pairing with for feedback. And it ended up giving us a much, much higher fidelity view into their work. What did it also do, we got more minorities, more women in engineering, we got people without degrees, we got people of all ages, and we had all kinds of folks joining because we weren’t filtering them with who knows what unconscious bias was happening when you did like a whiteboard interview, or a panel interview or anything else. And so we had, you know, I would say like, as I said, 85% of the people would make it past 90 days, which is maybe lower than most companies. But after that attrition was super low, like less than 1%. Because votes were like, like very finely tuned to fit our culture, because they love they did it for the first 90 days. So I think that was something else we did. It is something I’m still very passionate about meaning like, how do we have some sort of interview filter, right? Like that fizz buzz filter, and get people in the door, and then look at their work. Like, here’s example, when people were sometimes let go, they would come to me and say, thank you for letting me try. It wasn’t like they were mad, like, I didn’t make it. And I hate this process. And I can’t believe like we told them right at the beginning exclusively how it works. And they would even sometimes say, I didn’t make it, but I’m going to kind of refer my friend because I think that he or she could make it I’m like, so it was working. And we were so explicit and upfront that like, you didn’t have to like sign up. But I never had one person say, oh, that doesn’t work for me. Because imagine you go work at any other company, you’re still going to right imagine, you know, here’s the thing to try it and when you like are, you know, very rich from fellow doing super well in the world. And you have nothing else to do but go interview at Google or something. And you could say, you know what, I’m going to get it. I’m going to crush the interviews, but I’m gonna do anything. And like, see how long it takes you to like, no longer be at whatever bidco it does take I’ll take it, I bet you’ll take a long time. Because the interviews are a proxy, which probably gets you my guess is maybe 12 months in a big company. 

Aydin Mirzaee  28:52  

That’s crazy. You know, as you were kind of describing that, that did seem kind of fun to me. I’m like, maybe I will do that. But the other thing is, I feel like it would make a great movie, maybe a Harold and Kumar type movie, where they go and get jobs.

Farhan Thawar  29:06  

Well, you’ve seen I mean, we’ve probably seen it right? Like I definitely was I was at a company many years ago, where we interviewed somebody who was phenomenal. I crushed everything. And then when the person started, we were like, is this this I literally thought we had interviewed like twin, like a twin brother, because I’m like, this is not the same person based on the company and the processes and procedures, it takes a long time, maybe to get that person into a spot where you put them on like a pip and then you let them go. And that’s a lot of wasted time from a lot of people. And so versus a process which is very explicit and says, you know, let’s figure let’s work together for a little while. Internships are a great example of this, right? You work together with somebody, the intern figures out if they like what you’re doing, you figured out the intern is a good fit, you give them a job offer. No like that’s you don’t have to do anything else that that is the or you know, contract to full time, all of those things. But I’ve seen many places where it’s really hard to set that explicitly and hard to do. I mean in Europe for example, with the labor law I don’t know if you can do that kind of thing.

Aydin Mirzaee  30:01  

Yeah, it’s very interesting because like we notionally have these sorts of concepts, like you said, For internship programs and internship program is three to four months, typically. And so basically people are doing that in other contexts or I know investment banks have this sort of, you know, not everybody kind of makes it and they’re very serious about

Farhan Thawar  30:18  

I want to correct you on one point, yes, everybody can make it. I wasn’t trying to hire 50 people to only hire 40. My goal was, I want everybody good. You know, if I go to hire, you know, 1000 people and all 1000 are good. I will keep them all. It was not like some sort of horse race. Because some people think of it as like, Oh, yeah, you need 10 people. So you hire 15. And you hit the top 10. I’m like, No, I want everybody good. So if I can have a class of folks, a cohort come in, everybody crushes and does well, I would hire them all. So it was not about that it was just more about it was more of a fit. And we were constantly hiring. Like I said, we hired 1000 over four years, because we were constantly growing. And not everybody made it all the way.

Aydin Mirzaee  30:57  

Yeah, no, I think Yeah, and Thanks for clarifying that. Yeah, no, that makes a lot of sense. So how is this translated to the way that that you hire at Shopify today?

Farhan Thawar  31:06  

Yeah, so good point, we have a different process and a different level of scale. So a few things. One is, Shopify has figured out that we want to ensure that we look at many other aspects of a candidate than their just their technical background in order to like fight this exact problem. So in many ways, they’re using a different approach to fight a similar problem. So there’s an I think it’s well known now we have a life story interview. And what the life story interview does is we try to talk to people about their background in a way that helps us figure out what are the interesting things that this person has done in their past, the ways they’ve interacted in the past, the situations they’ve been in, and how they’ve reacted to those situations, in order to tease away any unconscious bias we have may have about something that may only come if you’ve gone to like a top school, or if you’ve grown up in a certain environment, or you come from a certain background. And so using the life story has helped us potentially filter for something else. And then we do have more about like pair programming, problem-solving, management, problem-solving style interview, and were you looking at that data to figure out constantly, are there ways to improve, reduce the number of touchpoints, improve the fidelity and signal through that, along with, that’s why we do pair programming as part of the interview, along with then when people start at Shopify, ensuring that they’re a fit for us, and we’re a fit for them in that process. So one thing, one thing that I experimented with a few weeks ago is that when we typically do a pair programming interview, the candidate drives and the interviewer is navigating, like saying, hey, let’s try this. And what if you did this, but the candidate is typing, right? So one of the things I learned at Pivotal, which does pair programming all the time, is that their interview process lets the candidate only navigate. So the candidate never touched the keyboard, there pair programming, but the interviewer is typing, and the candidate is navigating. And so what it’s doing is it’s taking away all the burden of the syntax of, you know, typing skill, like all the things that maybe are not relevant, and bringing in instead, the thought process, empathy, pushback, good ideas, like all the things you want from like a good person to pair with. And so we’ve been experimenting with that type of process in order to see if we can increase the fidelity of the interview using those kinds of techniques. And so it’s just a, it’s a different level of scale. But you know, truth be told, I spend a lot of time on this type of problem. I’m open to ideas, but one of the main is like having a giant internship program, right? We have like 400 interns this year, hundreds of interns that we then, you know, give out many, many offers to those who are graduating to then come back to Shopify. So that’s another way we do to like increase the fidelity.

Aydin Mirzaee 33:53  

Yeah. You know, I think I mean, there’s a lot to unpack and a lot to learn from that. But I think the part that really sticks out to me is just this constant experimentation, like even when you were talking about we reduce the interview from an hour to 45 minutes to half an hour down to 15 minutes, and actually experimenting and looking at the results afterward, like really treating hiring, like in the scientific method, I think I think that’s awesome. And I think there’s there’s a lot to learn just from that.

Farhan Thawar  34:21  

Yeah. And I think that there are so many experiments and things to try. I’m where I get worried when I hear you know, companies where the startups are big companies who like to follow some, like pre-determined plan or formula. They’re not really thinking of like the first like, they just copy somebody, right? They’re like, Oh, yeah, we should always have like, this department and that department. Here’s another example. Right? I mean, for the last before I got to Shopify, right? I had finance, HR recruiting, reporting into engineering, right? Not because it was like, I didn’t have any idea. I was just like, well, I want I think those things really contribute to the employee experience, right? How long you get how long It takes for you to get your expenses paid is part of your employee experience. how you interact with the vacation tracking system is part of your employee experience. So I’m like, why should those be like completely separate departments that have like this, you know, monopoly problem, right? Your HR department is the only HR department in your company, right? You can’t go to another HR department. So they don’t, necessarily, right, unless they are actively focused on making the experience better. Right, then, in many companies, they’re not designed to, give a great experience to their employees.

Aydin Mirzaee  35:31  

Yeah, I think and, you know, obviously, there’s no playbook for this stuff. And if you do just take, I mean, any process from a company and just like straight up, apply it somewhere else. And you know, what’s tricky about this is that you know, a lot of times, you’ll see a successful company or successful manager, and they may be successful in spite of something that they’re doing. And so just copying straight out without understanding context are within your own environment could be really dangerous.

Farhan Thawar  35:58  

Yeah. And I think I mean, even the same person, right? Like, there was a I think I tweeted it out last week, or something said, to copy someone is necessary, but to copy oneself is pathetic. Right? Like, I had this big debate, as I mentioned about extreme labs, I have 100 120, direct reports. And one ones were a burden for me and the report, so I never did them. Yet, at Shopify, I do one on ones because I find them a great way to share context on what someone is doing and possibly leverage if I can share context, that helps that person for the next year. If I do a bi-weekly one on one for the next 80 hours, that’s high leverage. So I even try to like an experiment on things that you know, or people should experiment on things. They thought if it worked over there doesn’t mean it’s going to work here, you should reevaluate that. Right? So I have a strong opinion about one on one. I no longer have that opinion, because I think it’s everything is situational. You got to try it.

Aydin Mirzaee  36:44  

Yeah, I think that makes a lot of sense. Farhan, I did want to ask you just one more question that I think is, is relevant, which is just about prediction and planning. You have this nice quote like I think in one of the blogs that you wrote, which is from Niels Bohr, and it says ‘prediction is just difficult, especially about the future’. You know, I know from, you know, a past lifetime at hearing Toby speak about prediction was, you know, there’s this mentality of like, anything past three months is really guessing. And what are your thoughts about planning today at the scale of Shopify?

Farhan Thawar  37:23  

Yeah, it’s a good question. I mean, I think what’s amazing about how we’ve set up the company, of course, I’ve only been there two years, and mostly like everybody else, right, I’ve got very little to do with their success is that the values really enforced where we want to spend our time, right. So one of our values, for example, is merchant obsession. So even if you go through, like some long planning process, and we don’t have like, super long planning processes the job anyway. But if we had a long planning process, which ended us up having a team working on something for some merchants, and you just ask yourself the simple question like, is this going to be good for merchants? And the answer was, no, I don’t care. Nobody cares how many hours were put into the planning process, we should stop doing that thing, and reevaluate how we can better spend our time to then give value to merchants. So I think that that has been ingrained in how we think about things. You know, another good example was when COVID hit, I think Toby went out and tweeted, or went on the news and said, we just deleted all of our plans. Like, we don’t care what we said we’re going to do and how much time was spent in there, and how much sunk cost we had in there, we’re instead going to focus on helping our merchants through a very difficult time, and still a very difficult time for many folks. And so that shows you the level of like, how we orient ourselves. And so of course, there’s planning and parts and parts of that to help us think through what are the best ways to spend our time for the future. But then as soon as something comes up, that potentially shows us a better direction, we’re happy to turn it off and say, let’s do this instead. Right. Another good example is, as we just talked about before, the call was, you know, we have these amazing offices, beautiful offices. They’re like featured in I know, the one in Toronto was featured in Toronto life, it’s so beautiful. But when we’re moving to remote, we didn’t say, you know, what, we spent all this time and money to kind of we just said, what’s the right thing for us to move forward on, and the right thing for us, for us to lean into the certainty of us being remote versus the uncertainty of when we’re gonna go back to the office. And so we did those offices. That was yesterday, like, what do we spend time on today? How do we make lives better for our employees to help serve our merchants tomorrow? So it’s a very interesting way to think about it, but we don’t try to plan too much from that perspective. Because it is, it is tough to know what the best thing is going to be to do for our merchants in like a year from now, for example.

Aydin Mirzaee  39:41  

Yeah, I think that makes a lot of sense. I mean, you know, the sunk cost fallacy I think when you have situations or stories, they become part of, you know, the myth and the lore within the company that remembers that one time this happened and so that becomes self-reinforcing, I suppose over the course of time, so you can’t almost get upset at it, which is we changed our plans. No. Like we’ve like this is what our company does. We look at the context, the recent situation and we do what’s best. But you know, founded and based on the values. 

Farhan Thawar  40:13  

I mean one other thing I’ll say is that when there is something that’s being done for a reason, it’s super important to have either a decision log or writing down why we got to that decision so that people don’t end up in this place where they don’t understand why it is because then they can question it. Right? There are many examples. I mean, there’s that famous monkeys trying to get the example of the banana. Have you heard that one? Right? Yeah. And then eventually, nobody has been hit with the firehose, but nobody is allowed to get a banana. And so

Aydin Mirzaee   40:39  

That one is actually a really interesting story. I don’t know that everybody’s actually heard of it would love for you to tell the story.

Farhan Thawar  40:46  

Yeah, sure. So there’s this experiment, where you have four like monkeys in a cage. And there’s a ladder leading up to a bunch of bananas at the top of the ladder. And so what happens is, the first monkey climbs the ladder, but there’s somebody outside and an experimenter who sprays the monkey with a fire hose, and so the monkey falls off the ladder. And then, you know, that tries it a few more times, and every time gets sprayed down. So all kinds of the first monkey realize he can’t go up. But what happens is, the second monkey tries to go up, and the other monkeys pull the monkey down, because they know that the fire hose is going to happen. So what happens then is they take out one monkey and put in a new monkey who has not seen this, these events, and the new monkey sees the bananas and starts going up. And before they even the fire hose comes out, all the other monkeys pull the monkey down, because they know that the monkey is going to hit with the fire hose. And over time, what you do is you remove every monkey one by one until there are all-new monkeys, none of the monkeys have felt or seen the fire hose. But anytime a monkey goes up the ladder, they all pull that monkey down, because they all you know been will have remembered or learn some sort of meme from, from the other monkeys that you never go up the ladder. But they don’t know why. And so this happens in companies all the time where somebody says, Oh, yeah, you never do this. So you never touch that. And without understanding why you can never question it, right? So verbs and this example, the experimenter could leave without the fire hose, right? And with the fire hose, and now a monkey could go up and probably get the banana, but no one will ever try. And I think that that’s a good example of people just like clinging on to the past or not knowing the history. And that’s why you do want to question everything from first principles and say that, well, in the past, there was an experimenter with a fire hose, but that’s no longer there. So if you’ve documented and said, our assumption is that, you know, this, this trend is totally true. And then somebody reads that and says, Oh, that’s no longer true anymore. Like maybe we’re everybody buys on desktops. And now the assumption is actually most people buy on mobile. If that was stated somewhere, someone could say, Oh, this has changed. That might mean all the things that follow are no longer true. We can experiment with this thing now and I think that really helps with setting the context for folks. 

Aydin Mirzaee  42:54  

Yeah, I think that that’s awesome. And a great way to kind of come to an end,  Farhan, this has been awesome. The final question that we asked all the guests is basically for all the managers and leaders looking to get better at their craft, any words of wisdom, parting advice, books, resources, anything that you would leave them with? 

Farhan Thawar  43:13  

Cool, I think my number one piece of feedback for folks is like, I spent a lot of time writing things down. And I think that that’s a really good exercise, mostly because the writing is more important than the reading, right? Like the fact that you wrote it down, or somebody said, something that was interesting to you, or you figured something out, or, you know, for me, I write down my framework of how I want to choose opportunities. I think that really helps clarify people’s thinking. And so what I would tell folks is like, you know, come up with a shortlist of, of goals for the year and I did this year was like, do fewer things better, write more and code more, the writing really helped me clarify a lot of the ways I think about these things. And if you can’t write in a succinct way about what you’re working on, it shows you that you don’t really understand it. And that really forces me to reevaluate how I think about these things. So my advice for folks is like, write more. You can share more, it’s scalable, and it does help you clarify your thinking

Aydin Mirzaee  44:04  

And a great place to end it. Thank you so much, Farhan

Farhan Thawar  44:07  

No problem. Thanks.