#126: How to approach PM takehomes for interviews | Ben Erez & Marc Baselga (co-founder, Insider Loops)

All right, what's up, Mark? What's going on, man? Dude, experiment after experiment, and I'm excited. I'm excited. I'm very excited because, like, you know, we've been doing these hourly writing blocks in the morning, and it's often like we get our juices going by talking about what's been top of mind for us. And one of the things that—and then today we're just going to start recording one of these so that we can turn it into hopefully something useful versus have all that context get lost in our off-the-record conversations. So I wanted to tell you about something that has been coming up in some of my coaching sessions that I think will be useful for other people to hear, and we haven't really talked about this yet, so you can kind of, like, ask the quote-unquote dumb questions or the things that someone in the audience might be curious about. But this—what I want to talk about is take-home assignments, and specifically some things I've observed in kind of, like, how people tend to approach them and what I've been kind of coaching them to do differently lately that seems to be very exciting and kind of encouraging and energizing for the people I talk to. So as context, when I say take-home assignment, what comes to mind for you? Like, are you picturing anything in particular? Yeah, so take-home assignment, I mean, I think at, like, a very broad strokes, right, is you get a prompt, and you have to do it on your free time, and then you have to, like, basically present an artifact or just, like, send the artifact back and you'll be graded. I think there's different flavors of, like, homeworks in terms of, like, sometimes companies will be, hey, like, just tell us, like, a two-hour block, you're free and you only have those two hours. Some others be like, hey, like, you have seven days, more or less, and then they give you, like— You know, under basically completely, like, you know, like underscoped times. Like, oh, like people usually spend six hours, which I would translate into most candidates probably spend 30 hours on these. But, you know, we don't want to feel guilty about the fact that we're not paying you, so don't spend more than six hours. So that's basically more or less. And then the other kind of flavors that I've seen as well is, like, some companies will literally just give you the prompt. You can ask some clarifying questions to the recruiter, but there's not really other checkpoints other than, like, you submitting it. Some others actually will have a couple of checkpoints, either with hiring manager or with someone at the company to see, like, to make sure you're not going down the wrong rabbit hole or to kind of test some of the assumptions. So that's more or less what I know. And usually it's like a checkpoint that happens usually after the hiring manager interview, but before the onsite. Yeah, no, that's great. I mean, I think the things I'm hearing there is it's kind of, it's a prompt that you get sent, you work on it on your own time, you submit it, and either it gets graded asynchronously or reviewed asynchronously. Sometimes there's a live presentation as part of the happy path. Sometimes there's no live presentation. Sometimes there's an expectation of a time box. Sometimes there's no expectation of a time box. Sometimes it can come before the hiring manager at some companies. Sometimes it comes after, from what I've seen. Sometimes it's part of the, it can happen in parallel with the full onsite loop, like we've seen can happen at Stripe, for example. So, and then what's interesting about take-homes too is that they kind of seem to be happening across both early-stage companies, and they're happening all the way up to, like, some of the big, bigger tech companies that we cover with our guides too. Yeah, and I think the other thing that I'll mention too is, like— Some of the difference that I noticed as well is, like, how prescriptive some companies are about the artifact medium that you present your work, right? Like, some, like Stripe is very much a writing culture, so they want it to be more of like a doc, versus other ones be like, hey, actually, I want a presentation. Some other ones might be actually, I don't care, like full liberty, you choose the canvas that you feel most comfortable with. So that's, like, another thing that I've noticed. And we know that Anthropic is partial for, like, HTML files for their presentations. Some of the loops require linking to prototypes. I'm also aware of some loops within Anthropic that don't require prototypes at all because it's just, like, not—I know they're not related to the prompt. So takehomes are probably one of the widest, I would argue, let's say least predictable vectors in interviews from what I can tell. And the reason for that, like, for example, if someone tells me, I have a hiring manager call coming up next week, how can I prepare? Knowing nothing about what the hiring manager is going to talk about with you, we can probably come up with a solid strategy for you for, like, how to get ready for that conversation. Yeah. We're not going to cover that in this episode. Or a product sense interview. Or a product sense interview, exactly, or an analytical interview, or engineering collaboration interview, or system design interview. There's a lot of these kind of interviews that you can kind of, you know, every company is going to have its own nuances, of course, as we've found when putting our guides together. But there's kind of like probably 50, 80 percent, you know, that you could get ahead on and then improvise the rest on the spot. With take-homes, what I've been noticing, because some people have been engaging me to help them get ready for interview loops, and sometimes it's like, hey, like I have a take-home. The next step is a take-home. They haven't sent it to me yet, but I want to get your thoughts on kind of like the process and how I can get ready. And what I've realized is it's actually, it's almost impossible to predict exactly where to— like, it's kind of hard to really prepare for the take-home itself in advance until you get the prompt. Once you get the prompt, then, you know, it starts to— like, I think a really great example of someone chewing on the prompt is like our episode with Yaniv Fatal, my friend from Israel, and how he handled his take-home. But it can be something that looks entirely like, what the hell do I do with this? All the way to like, oh, this is, I know exactly what this is about. So you can kind of have different levels of initial reaction to the take-home, but you still need to take your initial reaction and then plan out kind of like what you're going to do between the moment you get the take-home and when you submit it. And what I've observed, and I'm guilty of this in the past myself as a candidate, is you get a take-home and you just treat it as like a heads-down solo assignment. So you're gonna carve out some— clear your calendar if you need to clear your calendar, and then start working on it and work on it and work on it, iterate on it, sleep on it for a few nights maybe, and then ultimately just get to a point where, you know, you submit it. I maybe once or twice in the past had thought about getting some feedback from people that I know on my take-homes if it was like a really high-stakes kind of like interview process that I was not feeling very confident in myself, I would ask for some advice. And so that turns out to be what, like, basically everyone's doing with take-homes. As I coach people and I ask them, oftentimes I'm reflecting with them on past processes before we kind of look forward together. And, you know, they're like, Oh, I didn't make it through the take-home at this stage and at this company and this company. I'm like, Okay, can you walk me through how you, like, prepared for that? What did you do? Yeah, and then maybe one thing that might be helpful just to, like, make this a little bit more concrete, like maybe, like, share a couple of examples of, like, take-homes, like very broad strokes, so people can relate to, like, what kind of prompts people would get. It doesn't have to be exact, the prompt that they're asking some of these questions. Yeah, I mean, one that's top of mind for us that we've seen across, like, multiple loops, so we're not divulging any confidentiality, but, like, for example, we know that, like, multiple roles at a company like Anthropic will give people a take-home that has basically an area of the product that they're expected to improve or come up with new ideas or something like that. And then they ask the candidate to create, you know, to basically propose one, you know, like small idea or one small bet, one medium bet, one large bet. And obviously they want you to, like, some of them are very prescriptive about what they want to see you touch on as you present those ideas. Some of them are a bit more open-ended. So that's an example of, like, a prompt at Anthropic, like a take-home prompt at Anthropic that we've seen being used across, you know, several roles. So what recently clicked for me on take-home prep is that there are actually things you can do when you're getting ready for a take-home before you have the prompt. And so I wanted to, like, talk through those with you. So at a high level, I think it's important to basically have a plan in place for what you're going to do day by day—not minute by minute, but, like, definitely day by day—between getting the prompt and submitting it. So let's say you know you're going to have to submit it five days after you get the prompt. Let's use that as an example. And, you know, the recruiter is like, Hey, like, we just need you to submit it five days from sending it to you. So which day would you like us to send it to you? So you almost have some—you do have some agency as a candidate, oftentimes, to tell them when you want that clock to start. So this planning, this, like, decision-making starts with, like— By the way, would you say that's true for both people that have a job and people that are in between opportunities, or is it, like— Yes. And the reason for that is it's almost like I'm going to assume a full spectrum of how crazy people's lives are and how diverse their calendars are, okay? So you can go all the way from someone who's, like, single, no kids, unemployed, basically has no obligations to anyone and has a completely wide calendar for themselves, all the way to the spectrum of someone— Bro, why are you talking about me like that? You are not between jobs, my friend. So, but, you know, let's say people that don't have demands on their time, okay? So they kind of have, like, lots and lots of flexibility. Like if you ghosted from Insider Loops for five days, I'd be like, Where you at? Right? Yeah. So—or from the podcast or whatever. So I would say, so that's on one spectrum, and on the other end of the spectrum is people that, you know, like, you know— Dual income household, two or three kids, full-time job, very demanding, work in the office, right? You know, so like busy weekends. So you've got kind of the spectrum of people that you could work, that are going through interviews at any given time. So the first thing I try to do is obviously understand, like, where they're at, like what's their life looking like. But we don't spend too much time on that because, again, coming back to what we were just talking about is like regardless of where you are in that lifestyle, it's like you're still going to get five days to submit the prompt, and you're going to get the same prompt the other person gets, okay? So the first thing you need to do is figure out how am I going to make time for this, and when am I going to have time for this, okay? So if you know you're gearing up for a really big release at work, you know, let's say today we're recording this on a Wednesday. Let's say the recruiter is like, Hey, like can I send you the prompt on Monday, and you'll submit by like end of week or something. You might be like, Oh, like next week's actually looking pretty clear. Like, you know, I know my manager's out of office. We just had a big release a couple weeks ago. Seems like everything's great. We just did like all of our planning for Q3, so we're just kind of like heads down execution mode. I could probably carve out three or four days next week and, you know, maybe just kind of like call in sick or take some time off from work and get this done, right? So that's one way you could think about it. Another way you could think about it is like I actually have a vacation planned next week, right? So I'm not going to work on this on vacation, like while I'm with my family on holiday. So this would have to be following week, right? So, you know, that's the first, you know, there's a bunch of permutations of that scheduling, but you want to make sure you have a solid chunk of time to work on it. Some people have a lot of productive heads-down time that's available to them over the weekend, and some people have no productive heads-down time available on the weekend. So if I was like a weekend work owl, maybe it's better for me to actually get the prompt on Thursday or Friday this week. Right, tomorrow or Friday, like send it over. Like, it's better for me, right? But if weekends are dead, like they are for me, like it's really hard to get anything done. I actually don't want to count on any weekend time to get them done. Okay, so that's the first thing I'll say, is we try to figure out timing. Now, within that, what I've started kind of like pushing people to do, and this is because I'm not going to, like, name anyone, obviously, but like I've had a couple of people who I was coaching and, like, I was helping them get ready for, like, live interviews and interview strategy, and they would come back to me and be like, Oh, I didn't make it past the take-home. And I was like, Would you mind sending it over to me? And so some people would send me their submitted take-home assignment, and I'm looking at it, and there's, like, things—I've realized there's kind of like two buckets of things that kind of feel off about the take-homes to me. And I'm going to diagnose the problem, then I'll tell you kind of how I coach them to solve it. The first bucket is it's hard for me to understand what they're saying. Like, it's hard for me to understand the story. Like, it's not a really good standalone artifact that, like, if I pull up the prompt and I look at what they submitted, it's kind of like, man, like it's hard for me to know what you're trying to say here. Okay, so there's, like, a communication and a legibility issue. The second bucket of problem is even, you know, sometimes it's actually quite legible and it's easy to see what they're trying to say, but the ideas themselves are, like, feel weak or they don't feel like that strong. So the substance is kind of off. Okay, so the way—and obviously I tell them, Why didn't you, like, float this by me before you submitted it? I probably could have given you some, like, feedback, right? Like, you made it this—you made it so far. Why are you not getting help, right? And some people don't know that that's even an option. They're like, I didn't even think about doing that. Dude, it's so funny. Like, I've had the exact same experience of, like, friends going through take-home and either sending it after. Or sending it, like, the day before. And I'm like, and it's just so hard because I'm like, bro, like, you literally have less than 24 hours to submit this. And, like, I think you back yourself into a corner, totally, right, at that point. And at that point it's like, do I really want to rock the boat completely and then take accountability for, you know, like, if— because I think, actually, like, I really like the frame that you used of, like, basically, is the story good and is it clear, or are the ideas just kind of weak, right? And I actually think for the last-minute, like, rush that I was talking about, like, if the ideas are good, I actually think, like, that's a much more easy problem to solve the story, because then just— at that point it's just writing and, like, just basically sharing, hey, here are the disconnects, here are the things that I think that are not really flowing well. The idea one is like, you know, like, usually to me that stems from, like, a really bad foundation. Either you didn't really think about, like, the discovery, you had, like, a wrong framework, maybe you're, like, sleep deprived, so, like, your ideas were shitty because you were working on this and, like, without, like, getting any external help or validation or brainstorming. So, yeah, I totally agree, but it breaks my heart and it actually puts me in a really uncomfortable situation. Like, I just don't like it when people share last minute, because I'm like, you kind of know that when a homework is not going to make it. Like, I look at it as like, there's no chance this happens, but I'm like, what can I do at that point? And the thing is that the bar has increased substantially on what people expect to see with these submissions now. Given how competitive the market is, so a lot of, like, I think that for—I think people are just, like, for some reason taking the—treating the take-homes as a bit more like a checkbox, with less rigor than they would getting ready for, like, behavioral interviews or case-style interviews sometimes. Again, I'm generalizing, but some people are definitely doing that. Like, I've had multiple people tell me, Oh, I was just winging, like, the take-homes. Like, I was, you know, like, thinking—I just put something that generally looks legible on paper, and then I'll get through like I always used to get through in the past. And it's just like the bar has moved, especially for, like, the hyper-competitive companies. Yeah, I think AI makes this even worse in a way. Like, I think we actually were talking about this in our podcast episode with Eric Posen that he landed, like—and one of the things that we're discussing, hey, it's surprising that a lot of these companies haven't incorporated AI, like, you know, AI fluency evaluations into their loops. And actually, I think the take-home is actually one of the best ways in which companies evaluate AI fluency because, you know, in theory you have a limited amount of time, but companies are seeing disproportionate differences between, like, the top candidates and the average ones and the really bad ones, right? And I think the best ones not only are, I think, spending way more time, not only they're, like, just much better about how they use that time, like prioritizing, like kind of like what you were saying about, like, the bucket allocation of, like, hey, I'm going to spending actually a good amount of time in discovery. I'm going to be spending, I may be talking to customers. I'm going to be spending some time, like, really thinking about the story that I want to tell and the why, and then building, plus kind of amplifying every single step along the way with AI, right? So it's not just like that's getting really competitive. It's like you're talking to people that are like maybe haven't, are barely sleeping to get this done, plus like using AI like to another new level. So it's just, yeah, it's, I just want to like emphasize this because I think a lot of people just don't really get it. Like, yeah, I mean, I think AI has increased the floor on what's expected in take-home submissions. And then I think it's increased the ceiling on how high people, how far people go with it. There's like things that you could literally not do without AI within five days that you can do now. So it's like you can do much more. Anyway, so with all that said, the kind of like advice, okay, so like what do I actually share that people should do? So you've got your calendar figured out. You know when you're going to get the prompt. You know how much time you'll have. Then the question is like, well, how should I spend the time? Like, and what, basically what do I do during that time box? Okay, so like from getting the prompt, what do I do next? And this is still happening before they receive the prompt, right? Like this is still prep. No, the prep that you're sharing right now, like what to do with the time. Yeah. Is that still before you know exactly what the prompt will be? Correct. Cool. Correct. I want to take a moment before we get back to the episode to tell you about Insider Loops. This is a company that Mark and I started because we just believe this should exist. Companies should be offering candidates the best possible preparation for their loops, yet what they give them is vague and often not very actionable. So we took the matter into our own hands. We've put together well-researched and continuously updated guides for PMs interviewing at companies like Anthropic, OpenAI, Google, Stripe, DoorDash, and more. If you're getting ready for PM interviews at those companies, go over to insiderloops.com and learn more. Now back to the episode. So, and that's what I meant by coaching. The coaching, because we don't know before they get the prompt what it is, like the coaching I give them is, assuming we don't know what the prompt is going to be, what can you do now? Okay. So here's what I've been telling people. Given what we said about where the submissions can fall short, with, like, the legibility and the storytelling being one bucket, and the kind of quality of the ideas being the other bucket, or the quality of the content being the other issue, you basically want to build a support system for yourself that I think requires one or two people to agree to help you through the take-home. Okay, so here's what I mean. Whether it's a coach like me, or it's a friend who's someone you trust, you want to make that upfront ask before you get the take-home, like, hey, I'm about to get this take-home. I'm gonna have to submit. I'm like, hey, I'm about to get this take-home. I'm gonna have to submit it in five days, and I would like to be able to use about an hour total of your time over the course of that, if you're open to helping me. And assuming you can find people to say yes to that, then here's how I generally use that. So the first thing I would do is I would, the moment you get the prompt, you want to already have ideally some time on the calendar if you know when the prompt will come. Let's say you think it's going to come like on a Monday morning. You want to have 30 minutes on the calendar Monday afternoon with, if it's a coach, with that coach, or if it's your friend, your friend, and basically chew on the prompt together. So you pull up the prompt, you talk through it together, you make sure that your understanding of what is being asked is the same as the other people's understanding of what's being asked. The reason this is so critical is what I recommend next is a 48-hour heads-down period. So this is you, you know, working on creating the first draft of your submission. If you don't have a good— Can I ask a question about, like, the chewing together part? Sure. Let's say what happens if you're, like, understanding it incorrectly? Is it— do you think often one person is just correct and you're just kind of like— I don't know if there's correct. I don't think there's a correct here. I think there's, like, if you and I read the same one-page take-home prompt, and I think they want one thing and you think they want something else, we can debate it and try to read between the lines on, like, given the role, given what we know, like— What do we think they're actually asking for? But in the case of an entropic, like the prompt that we talked about before, with like the small bed, the medium bed, and the large bed, and then like the product that it's about, there's only so much room for us to kind of like misinterpret what the ask is, right? But still, it's important to have someone to talk to and be like, I want to make sure my nerves are high, right? Like this is my interview process. My, like, adrenaline is kicking in. I just cleared the next five days of my calendar. I'm about to go on a bender to like, you know, work on this thing. I might lose some sleep. I want to make sure I'm going to spend my time in the best possible way as I go into this, right? So you want someone to check your thinking. And the check your thinking is kind of like, let's make sure we're on the same page about what the prompt is. And also, hey, like, here's kind of how I'm thinking about spending my time. Like I'm going to start by doing this, and then I'm going to do this, and then I'm going to do this, and then I'm going to tie it all together into this kind of artifact that will flow in generally like this kind of arc. Is that kind of what you think would be needed? Is there anything that you would change about that? Like I'm thinking about probably having an appendix where I kind of stuff some stuff and maybe link to it. And this prompt also requires some prototypes. And instead of trying to like embed the prototypes into like the HTML and like an iframe or something, I'm probably just going to like link to them in the respective sections. And I'm probably going to spend maybe like, you know, I said 48 hours, so I'm probably—84, sorry—48 hours. I'm probably going to spend 24 hours on like the prototypes and the ideas themselves, and then I'm going to probably spend the second 24 hours kind of like creating the narrative and trying to like packaging it all up into the artifact. And then my goal is to basically share with you by the end of day after two days, like, so 48 hours later. I'm gonna share my first version with you, and then I would love for you to be able to—and this is where kind of like, if it's one or two people, this will vary. If it's a, like for example, I was recently helping someone who's working through one of these take-homes, and the prompt is not in my—I'm not a subject matter expert on, like, the area of the product that they're interviewing for. And so I was like, there's a good chance the prompt that you're gonna get is gonna be something I'm not an expert about. So you might want to get help to find someone in that second bucket, someone who's like someone you trust, who knows a lot about, who can really evaluate the merits of the ideas and the substance. But I can definitely be your go-to person for letting you know if I can understand, does the artifact make sense? Do I understand the ideas at a high level? Maybe I'm not gonna tell you whether, like, that's a good idea or this is wrong for this reason, like I, you know, because I've never worked on that exact product before. But, so, you know, but there's another person I'm helping where the thing that they're gonna interview for is actually something that I don't say I'm a subject matter expert about, but I have some experience with. I could probably have some opinion about the substance of the content too. In that case, they can just work with me as like a coach. I can do both of those jobs for them. But the idea is you want to get a touchpoint with someone after 48 hours, get feedback on what's not landing both on the ideas, the story, iterate. Can we pause there for a sec? Yeah, because I think I have a couple. Yeah, so super helpful. The kind of the meeting with someone right after you get the prompt kind of reminds me of like, like coming together with a plan with Claude or Codex, like you're about to start a big project and you're just kind of, hey, like poke my assumptions, ask clarifying questions, share the plan, you get some feedback on the plan until you feel pretty good. It's like, here's the plan for 48 hours. How do you recommend spending the 48 hours? Like, is it usually, like, I know, like, probably varies depending on the prompt, but like... How much of it, like, for example, like, I remember, I mean, I haven't done a take-home in a long time, but actually, like, the last take-home that I remember doing was for Asana. And actually, one thing that I think they really liked what I did, I did customer research. So I basically interviewed like five or six friends on a specific prompt that they had and just kind of get their thoughts. And that was like a really helpful foundation. And plus, like, also I did some, like, I think, like, quant data of like, hey, here's like some trends that we're seeing in the world and how, like, people, I think it was like a Spotify-related prompt and, like, how, you know, music consumption and podcast consumption is changing. And so I'm curious, like, is there, yeah, like, how much do you make them go through, like, kind of the normal product process, if you will, versus just, like, kind of jumping in and, like, start to, like, you know, putting together prototypes and stuff like that? Yeah. It's really hard to give generic advice on this, or, like, general. It's going to depend on the prompt. It's going to depend on the role. It's going to depend on, like, a bunch. Like, again, like, this is an area of a wide amount of variability. So I tell them, I mean, that's kind of the goal of that first session together after they get the prompt is to be like, okay, now we know what the question is. How do we spend the time? How are we thinking about going about this? Check your thinking. Can someone do that with AI? Sure. But given the stakes, you know, I think one thing that's, like, I just believe is more and more true every day almost is, like, the higher the stakes on something, the more you want to go straight to, like, the most credible source for advice that you can get. And oftentimes that might be... An interview coach, that could be just like a really good friend of yours whose judgment you trust, but an LLM is not going to be, I think, your reliable thought partner on this. So you come up with your plan. You make sure you're on the same page about what's being asked and kind of where you want to be at the end of that 48-hour block, and then you go heads down. We don't need to spend too much time about what happens while you're heads down. You just imagine you go into the cave, and then you come out of the cave 48 hours later with a first draft. You send it to the people that we're talking about, like your trusted coach or friend or whatever, and ask them specifically what kind of feedback you're looking for. So I think a great question to ask is like, does this make sense? Like, are you following my thinking? Do you understand how I reached the conclusion or the recommendation? Was it hard? Was it anything confusing for you? So just basic legibility and understandability questions is like definitely something you want to get a pulse check on. I do find myself having an easier time giving that kind of feedback in the form of a Loom. So like pulling it up on my screen, recording myself talking through it, highlighting certain things with my cursor perhaps, scrolling through the artifact, clicking through, and then they can take that transcript and they could, you know, do what they want with it if they want to use it to help them kind of iterate on what they put together. So that's the first bucket. Second bucket will be for whoever it is that's like the subject matter expert, if that's relevant for the prompt, which it always—or for the submission, rather, which is generally the case. There's going to be subject matter expertise that's going to be helpful here. Otherwise, why are they interviewing you? Like, they're trying to figure out whether you're, like, a subject matter expert, like, do you have the knowledge that they need for the role, right? So basically, if there's someone that you're like, I know someone who's even better for this role than me, they're just not looking on the market right now, that would be a great person to have occupy that second slot and give you feedback. And then be like, hey, like, real talk, you know, what do you think about these ideas that I came up with? Like, does this, you know, are these, like, bad ideas? Like, if you were evaluating this take-home, like, is this weak? Do these seem well-reasoned and interesting and new and novel and unique? What kind of questions would you have if I was presenting this to you about these ideas? Like, what are the pitfalls or what are, like, the gotchas maybe if I present these ideas and something I need to think through as a follow-up question? So you want to ask, you want to make some space to get that kind of feedback on the submission too. And remember, because you're only 48 hours into the five days at this point, you've given yourself plenty of time to iterate, okay? So you take that feedback that you get. We could spend a bunch more time talking about the kinds of feedback you might get at that stage, but just imagine general two buckets of feedback you'll be collecting. And because you involved the people who give you the feedback on both of these buckets early, even before you got the prompt, remember? Like, you pinged them saying, I'm going to get the prompt on this day. Can I count on you to help me throughout that process? Be like, yeah, sure. Okay, I got the prompt. Can you make, like, 20 minutes to just, like, chew on the prompt with me and make sure, like, you're aligned with how I'm thinking about spending the time? Great. 48 hours later, now here's my first version. So they have the context. They know kind of, like, you've kind of been hooking them in. Now they're invested in the story, right? They want to give you the most helpful feedback because at this point they're like, I kind of want this person to win. You know, like, they've been doing this so, like, I'm impressed with how they pulled me in. They're being so thoughtful about this. They're investing the time. They're putting in the legwork. They're being really pointed with what kind of feedback they're looking for. So you're invested. Now you give them the feedback. They take the feedback. 24 hours goes by. You iterate, okay? I'm not gonna—we don't need to flesh out what happens during the 24-hour heads-down iteration period, but you come up with your version two, okay? Go back to the people. Ask them, Hey, can you review this? Same exact question. Does anything feel confusing about the story, et cetera? Does any feedback on the merit of the ideas? Also, I threw in like an FAQ or like an appendix that includes some Q and A that I want to demonstrate that I've thought through some questions. So I'm gonna, like, literally put the questions I've thought through, or like risks or pitfalls or gotchas that I can anticipate, put them right into the artifact, but not in the main body. So now there's an appendix. Can you review the appendix and let me know what you think about the appendix, right? And then you get another round of feedback, and now you're, what is it, like a day, 24 hours—you have 24 hours left to iterate one more time. If you want to pull them in for one last review, you can, although I'd say at that point you're making like a third ask for a review from people, so you might be coming up against, like, the social capital. Like, you're asking more than is reasonable at this point. With a coach, I think it's less of a concern. You're paying them for their time, but, like, with a friend, you might want to start to be mindful of that. And then you submit, okay? Now, this does not guarantee that you're going to pass the take-home. Let's be really clear about that. What it's doing is it's increasing your chances of success. It's decreasing your chances of failure. And most importantly, in my opinion, the psychological piece here as a coach is if you don't get it, if you fail this stage. Of the interview, you'll know you did your best. You gave this your best. You tried. You didn't hold back punches, but you didn't get it, right? That feels different to people. I'm telling you, it feels so much better to people to know that than to not do their best on something where they really want the job. So, like, they really want the job, but they don't do their best on this assignment. They hold back, and then they don't get the job, and they're like, shit, right? Like, this was like I shot myself in the foot. This is like I could have done more. I could have done better, right? But if you're in that bucket of the person who's gonna do what I'm describing, like who follows the advice I'm giving here, you're likely gonna have like a sense of like peace with like what you did. You might be a little pissed off that you spent all this time and didn't get it. That's normal. But at least you're not gonna be thinking like, Hmm, I messed up. Like I didn't give it enough. Yeah. No, I think that makes a lot of sense. And another thought that I was having is like when you're sharing like the B2 or the B3, the version three, like something that I like to do sometimes is like when I'm working on something that's like a little bit like a longer project that needs to stand on its own, is like sharing with like a third person that has like the fresh eyes perspective. Because I think sometimes like when you have someone who's been helping you make edits along the way, they already have all the context, so like they'll kind of like lose perspective. So I think just like having someone who hasn't seen any of the versions and making sure that that stands on its own, I think it's, so I think that's one thing that I would just throw into the mix, assuming you're very popular and have a bunch of friends. Although I will say the risk with that is that they put that person in the situation that you described not like, that you didn't like being in in the beginning of our conversation today. Where, like, they come to you and there's like 24 hours left to submit it, and you're seeing it for the first time, right? So they can do it, but the risk there is that the recipient feels the way you felt earlier, the way you were describing feeling earlier, and that they potentially have to spend a lot of time kind of bringing you up to speed on the context, which can take away a lot of potential. Yeah. And if I was going to boil down, like, that feeling that I had, I think the reason why I felt, when that has happened to me, has felt so bad is because, like, one, I feel like they've been on the— like, I know the process they followed hasn't been the one that we just described. They've just been on the cave for maybe 72 hours, or they waited last minute, they procrastinated. The other one that I think is equally important is, like, they didn't really give me the guardrails of what kind of feedback they want. So I think, like, so for that last B3, it's like, hey, like, does this story— like, I would probably— the most important thing is, like, does this story feel cohesive? Like, I don't have time to, like, go back and come up with better ideas, but, like, can you follow the narrative, right? Like, is this presented— it's like more almost like kind of like the presentation, like the substance. Like, we don't have time to go back to the substance. So I think, like, it would feel okay if you give me very clear guardrails. I think I can work with that, right? Like, you're giving me constraints in which I can operate and give feedback. But no, I really like that process, and I think it just— I like the fact that it creates accountability too, because, like, if you put those meetings in the calendar for people, like, hey, like, we're going to be checking in in 48 hours, like, it forces you to get a lot done. So I think, like, that just, like— Not only it's good for the review, but just, like, forcing you to behave like, I committed with this person. And so I think that that is really—I like that idea a lot. A question I had for you, when you were, especially around the 48 hours, I'm someone who really struggles kind of, like, actioning on feedback when it's, like, two different levels of, like, altitude, like the substance versus the story. And, like, I prefer—I usually prefer to do those in serial almost, so, like, first do one and then do the other. And what are your thoughts there? Because, like, I usually prefer, like, you know, well, yeah, getting the substance right and then getting the story. So I had a feeling that's what you're gonna say. The issue with that is that it's actually kind of hard to give feedback on the substance if the story makes the substance illegible. The story is the packaging, right? So if you're like, Hey, I just—I don't really want your feedback on the packaging. I want the feedback on the contents, and I'm like, It's hard for me to give you feedback on the contents if I can't navigate the packaging, right? I don't know how to open this box. It's like you want me to give you feedback on what's in the box, but I can't open the box. Like, it's really, really hard. And I don't think the people that you present this to in an interview are gonna spend a lot of reps figuring out how to open the box so they could find out what they think about the contents of the box. So I'm trying to—I'm like a super high—like, I have the most positive interpretation of the artifact than anyone in the world at that point, right? Like, I'm your—I'm literally helping you. I'm a coach trying to, like, help you ace these interviews. There's literally no one who's— Rooting for you more than I am at that point. And if I can't figure out how to get, how to see the idea, like what, how to find the idea, your problem is the packaging. Your problem is—and then look, if you get the packaging right, and then you get me another iteration, and now the ideas are legible, I'll probably have some thoughts about the ideas. But, and, you know, ideally you want to kind of be thinking about both in parallel, like kind of like any great product manager, right? Like you want to be thinking about the positioning and the packaging of the thing you're working on, as well as, like, the underlying value it's delivering and, like, the meat of it. If a tree falls in a forest and no one hears it, did it really happen? kind of thing, right? So, you know, you need to make sure you're thinking about the story from the very beginning. The way you— even your directions of where to explore in those 48 hours should be based on kind of, like, potentially working backwards from the kind of story you want to be able to tell. Makes sense. But, you know, this is an area where, like, I'll be honest, like, as a coach, I enjoy most working with people who actually don't want you to interfere with what they do in those 48 hours. Like, there's a certain caliber of— and it's mostly PMs that I work with, but let's just say person— who, once they kind of get aligned with you on, like, how they're going to spend time and what they're going to do, and you've created the kind of very productive, like, set of constraints and stuff, and you have some accountability check-in points, that's where you get to find out, like, I don't say how good they are, because it sounds like fixed mindset-y, but, like, you get to find out kind of, like, how much they want it. You get to find out how creative they are. You get to find out what their skills look like, right? Like, what they come up with after 48 hours is usually going to tell me a lot about— Them. And I often work with people as a coach who I don't have a working history with in the past. So it's like, that's often like the first time I'm actually seeing a work product from that person. And so I'm putting myself in the shoes of a coach, but I'm also putting myself in the shoes of like a potential peer or a hiring manager. And oftentimes I will, I mean, it's kind of hard to not feel something in that moment. It's like, wow, like this is good, you know? Like, I'm impressed with this. This is interesting, right? Or it's more like, hmm, how do I untangle this? Right? Like, you know, those are kind of like two spectrums how you could feel looking at a first draft of one of those things. I will say like how much someone can turn around a first version to a second version also tells you a lot about them. Like the ability to take feedback and like a little nudge and run with it and like overhaul something to go from like, ugh, like this, how do I untangle this? to like, wow, this is really good. That is possible in a single iteration over 24 hours. Like, I've seen it. But the point I'm making with all this is like, you want to give yourself the best chance of success in these interviews. Your friends are rooting for you. If you're working with a coach, they're rooting... If you're working with a coach, they're rooting for you. You don't get to do that many of these. Like, if you're going to talk to, you know, anywhere from, call it three or four companies to, in some cases, dozens, like, you know, 20 companies or something over the course of a job search, they're not all going to do take-homes. But, you know, I'd say it's probably safe to assume 30 to 50% of them will do take-homes, if not more, if you're going to go with startups. Like, I know earlier-stage startups are doing more take-homes than, like, a lot of big tech companies. Like, Meta doesn't do take-homes. We know DoorDash doesn't do take-homes. So, you know, there's those kind of companies. But many companies do, and I'd say, yeah, in this market, you got to show you want it. And the way you show you want it is to do a really good job on the take-home assignment. So, like, it's going to count. So it's going to count. And I think the last thing I'll say there too, that I heard many stories of—not many, but like a couple—of, like, creating a really good artifact actually is probably the component of an interview process that has the longest shelf life. Like, you may not, like, get the specific role, it might get filled up, or they might choose someone better, but, like, that artifact is going to live in the hiring manager's memory. It's going to live in the ATS. And if they can be, hey, like, it's a great way to be, hey, I couldn't hire this person, but, like, look at this awesome thing they put together. We saw that at Uber. We saw that at Uber and saw, I forgot where, but someone was sharing that, like, they did something for a hiring manager, put it together, that got shared internally and then ended up... Kind of leading to another, you know, conversation with someone else. So if you do it correctly, it might pay dividends too down the line. And it's like another artifact that you can share with people, and that it's like a great way to show your product thinking. And so I think that's like another way to think about it too. And I think the last thing that I'll say too is like, choose those people that you're asking for help correctly. Like, I think there's some people that actually sometimes don't have the courage to be honest with you. And I think a lot of times, like, people need to hear like, Hey, this is not good. Like, you can do better. So I think someone who, like, you know well, that kind of knows your potential, that can be like, Hey, like, this is not up to your mark, and can tell you with a straight face because they care about you, it's way more important than someone random you just met on Reddit or, you know, another, like, matching company that doesn't really know any history about you and who has no incentive to be, like, brutally honest with you. A hundred percent. Cool. So yeah, my goal, my goal this whole episode was to have something that I can share with people who are either not available, like time-wise, to coach through, like, take homes, or people are just trying to, like, think through how to come up with stuff on their own and don't want to work, don't necessarily want to engage with a coach, but are maybe in a bind and looking for, like, very specific advice. So I think this captures it pretty well. Yeah, this is awesome, man. I'm excited to share this with people. Cool. Thanks for diving in. Bye-bye. Bye-bye, everyone.