327: Estimate Crafting
The Bike Shed - Een podcast door thoughtbot - Dinsdagen
Categorieën:
Steph joins Chris in trying new things! For her, it's a new email client – the Newton email client – because she really wants to love her inbox. She also talks about implementing a suggestion from Chris on improving CI speed. Chris continues his search for the perfect to-do list app. (It's not going great.) But he has made hiring progress and is excited to move on to the next step: onboarding. Together they answer a listener question who asked for advice on crafting project estimates for clients. This episode is brought to you by ScoutAPM. Give Scout a try for free today and Scout will donate $5 to the open source project of your choice when you deploy. Services down? New Relic offers full stack visibility with 16 different monitoring products in a single platform. Newton Subscribe to Email Newsletters in Feedbin GitHub - Shopify/packwerk Sunsama TickTick: To-do List, Tasks, Calendar, Reminder Become a Sponsor of The Bike Shed! Transcript: CHRIS: I am now recording. STEPH: Me too. CHRIS: [laughs] That's my recording voice. STEPH: [laughs] CHRIS: That's how you can tell. STEPH: I just like how it sounds suspicious where we're like; I'm now recording, so be careful. [laughs] CHRIS: This is now on the record. Hello and welcome to another episode of The Bike Shed, a weekly podcast from your friends at thoughtbot about developing great software. I'm Chris Toomey. STEPH: And I'm Steph Viccari. CHRIS: And together, we're here to share a bit of what we've learned along the way. So, Steph, what's new in your world? STEPH: Hello. Happy, happy Friday. Oh, I have something that I'm excited or intrigued about. I don't know. Okay, I'm hyping it up. [laughs] But I'm realizing I'm also very skeptical of it. CHRIS: This is the best sales pitch I've ever heard. I'm so excited to hear what this is. [laughs] STEPH: I am trying a new email client; it is the Newton email client. And I so want to love my inbox. I want to check on it. I want to help it grow. Okay, that's the opposite. I want to help get through all the emails that come through, but I just want to love it. I want it to be a good space that I want to go to. And I just hate email so much. And it always feels like this chore that it's really hard for me to bring myself to do, but yet it's really important because a lot of good things come through email. So this is my rambly way of saying I'm trying the Newton email client because I saw on Twitter from Andrew Mason, who has very similar feelings that I do about email, where we are just not fans of it. And we rarely check it and have declared email bankruptcy at several points in our life. And he's also one of the co-hosts for Remote Ruby. But I saw on Twitter that Andrew was talking about the Newton email client and how it actually made him feel that he enjoyed writing and looking through his inbox. And I was like, yeah, that's the sales pitch I need. So I'm giving it a go. It's been only a couple of days. But one of the nice things I have noticed about it is it's very focused, and there's not much noise, and it actually feels like very minimal design where if you open up like a new email, so you're opening up a new draft, there's no much noise. You get to just focus, almost like you're writing a little blog post or journal post or something. It takes away a lot of the noise. While in Gmail, it's going to open up a small window in the right, but then you still have the rest of the noise that feels distracting. So I like that very intentional like, hey, you're just doing one thing, just focus on this. And then also you can integrate other email accounts as well. So you can have one-stop shopping versus Gmail, then you have to click around and sign in, sign out, or visit different email accounts. So we'll see if it helps improve my email life, but that's something new I'm trying. CHRIS: Very interesting. So you're fully on inbox zero life now. That's what I'm hearing. [laughs] STEPH: Ah, hmm. I don't want to lie to you. [laughs] We have a good friendship. I won't start lying now. CHRIS: I appreciate that. So you're halfway to inbox zero. You're not even entertaining that idea, right? This is just you want a better tool to do email. STEPH: Exactly. Inbox zero is not incredibly important to me. But I do want to make sure that I know that I've seen everything important, and I know where to find things. And then making sure that I am responding to people in a timely manner. Those are more my goals. Inbox zero, if that supports it, then great, I'll work on it. But not necessarily that has to be the goal that I reach. CHRIS: Gotcha. I'm not seeing Newton, but I'm intrigued. Particularly on mobile, I have the Gmail mobile app, and that has unified inbox, which I appreciate. But Gmail on the web does not, and I find that odd. And I've never found a mail app that I enjoy because I want some of the features of Gmail. I want to do Gmail snoozing because I still want that to be consistent and whatnot. And to be honest, that's the main way that I get to inbox zero. I just say future me will have more time. I actually tweeted recently. It was a screenshot from my Saturday inbox, which I think was 15 emails that I'd snoozed from the previous week into Saturday morning. Because I'm like, Saturday morning me will have so much time, and energy, and coffee, and it'll be great. And then it became Saturday morning and, ooph, what a view. STEPH: [laughs] Yeah, your snoozing tip has been life-changing for me because that's not something that I was using all that much. The two things are, one, schedule send so that way if I do have a sudden burst of energy and I want to write an email, but I want to make sure that person doesn't get notified until a decent time. Being able to schedule an email and snoozing is amazing. I think Newton and Gmail have pretty much similar features. I was trying to do a comparison. I was like, is there something really snazzy that Newton does that Gmail doesn't already give me? But it looks like they all do about the same, having those important features like snoozing and then also being able to schedule emails. So I think it really just comes down to a lot of the UI, and there may be some other stuff I'm missing since I'm new to it. But that's the main appeal for me right now is the focus and the look and feel of it. So then maybe I will find looking through my inbox a more zenful experience, I think is how I saw them advertise it. CHRIS: Well, I definitely look forward to hearing more as you explore this space. I will say looping back to what you were just commenting on around deferred send, which is definitely something that I use, but you described one of the reasons that I use it. So the idea of wanting to be respectful of someone else and not send them an email on Sunday night because you happen to be working at that point. But you don't want to put that on their plate. I would say equal amounts; that's the reason I use scheduled send. And then the other reason that I use scheduled send is please, for the love of God, I do not want another email back in my inbox. So I will reply to something such that now I'm done with that, but I will schedule send it for the next morning. Because tomorrow morning me can deal with whatever reply this generates. There's some adage; I don't know if it's an adage, but the idea that every email that you send generates 1.1 emails in reply. So emails just have this weird way of multiplying. And so if you send one out there, you're probably going to get something back. And so often, if I'm trying to clear my inbox, I don't want to get another email in my inbox at that moment. So I will not actually send the reply. I will schedule it for a future time because I do not want to hear. I want no new inputs at this point. I'm trying to process them. So that's part of why I use deferred send. STEPH: I had not thought of that, that yeah, that if you schedule it for tomorrow, you've really gamified this inbox zero because you're like, yeah, if you send something, then you might get an email back. But you're like, if I wait till tomorrow to send it, then I'm less likely to have another email, and then I've hit inbox zero, and I'm set for the day. I like it. It seems helpful. CHRIS: Yeah, inbox zero sounds like an altruistic thing, but it is not. It's a way to force myself to have to make decisions, which is something that I want to get better at broadly. And that's part of the role that I have now. A lot of what I'm interested in exploring is just getting better at making decisions, being more decisive, being more action-oriented. Because I just have a tendency to make many, many spreadsheets and think about stuff for a while and take a long time to make a decision. But I don't get to do that, particularly now. But broadly in life, that's probably not the right mode to be in. So inbox zero is another thing that forces me to deal with things rather than just be like, I don't know, I don't know, I don't know, and keep looking at the same thing over and over. So just more thoughts about inbox zero, but now I'll stop talking about it. STEPH: I do like that, though. And you're totally right; it can be a very helpful constraint. And I think that's sometimes why I fight it because then I haven't curated my inbox enough that then when I go to it, there are so many interesting things that then I feel a little bit overwhelmed where I'm like, oh well, I want to read this, and I will look at that. And this seems interesting, and maybe I should be a part of this. It feels like one of those like; you could be a part of these ten amazing things. Do you want to be a part of all of them? And given a person that it's hard for me to say no to or recognize that no, I'm just going to not do anything with this, that is hard for me and would be a good skill for me to hone in on and practice and make quick decisions and be very realistic. Because I used to be subscribed to more newsletters, and then I finally had to stop subscribing to them because it had that same effect on me of that FOMO of like, I'm missing out on this great article or this great video. And I've become more honest with the fact that my Saturday morning self isn't going to want to read through a bunch of newsletters and videos about coding, that I'm going to want less screen time. So that is a really good constraint and helpful skill to cultivate for sure. CHRIS: All right, I said it was done, but one more thing. I feel like I've mentioned this in the past, but Feedbin is the thing that I use for RSS. I still believe in RSS as a technology. But everyone's moved to newsletters these days that go via email. Feedbin gives you an email address that you can use to subscribe to newsletters, and then they do the job of converting that into an RSS feed. So for me, I take something that was now a push into my inbox, and now I can pull whatever I want from that RSS feed. And on Saturday morning, if I'm feeling like, with a cup of coffee, I can enjoy some newsletter about all the new hot tips in Svelte land or whatever it is or not. But it's not clogging up my inbox. And with that, I think I'm actually done talking about inbox zero. [laughter] STEPH: Yeah, that's a nice separation. We could keep going. I have full faith in us that we could keep going about this. But I'll share a slightly different update. I've been implementing a suggestion that you provided a couple of weeks back where we were talking about Rspec's selective test running and how some applications will speed up their test. If you change one part of the codebase, then perhaps you only need to test this chunk of test. You don't actually need to run the full test suite. And that is complicated and seems hard to get right, and really requires understanding boundaries. But then also knowing Ruby, then how do you really identify? Do you really know where this method is being called and can identify all the tests that need to be run? I think we'd mentioned before there's a really good article from Shopify where they have worked on this and created an open-source project called Packwerk. So we can link to that article in the show notes. But more specifically, you suggested, well, what if you just change a test file? That seems very low stakes and also has the benefit of creating a reward where if someone does see something that they can improve in a test, then that's a very quick feedback. Let me just get this change. It's going to be fast on CI. I can merge it right away and also saves time on CI. So I've been working on implementing that change. And it's one of those the actual change is easy, like checking with Git to say, "Hey, what files have changed?" Does it have an _spec.rb at the end of it? Great. Does it not? Okay, we've changed some application files. So let's run the full test suite. That part's easy. Getting it integrated into the build system has been more complicated just because this team has done a lot of work around trying to improve and speed up their test. And there's a fair amount of complexity that's there. So then figuring out a way to stitch my change into all the different build processes that take place has proven to be more difficult. But it's also been insightful just because it has now helped me really understand and forced me to learn, okay, what are all the different steps? What's important for each one? Where can I cut off the rest of the running of the test and instead just focus on running these tests? So in some ways, it's been challenging, but then on the positive side, it's been like, okay, well, this has taught me a lot about the existing system. So at this moment, it is still a work in progress. I'll have more updates in the future. I am excited to see the rewards. I've gotten to the point where I just have a proof of concept where I've gotten pushed up, but it's not production-ready. But it's at least I just wanted the feedback that I'm in the right spot and that we're running just the right test. And so far, it does seem like it's going to be a nice win, even if it's maybe not used by everybody because it's probably rare that someone is altering just a spec file. But for people who are looking specifically to improve the CI build time and working on tests, it will be very helpful to them. So yeah, I'm sure I'll have some more updates in the future. What's going on in your world? CHRIS: Well, I definitely look forward to hearing more about that. However, we can improve CI speed; I'm super interested in that as a topic. Mid-Roll Ad Hey, friends, let's take a quick break to hear from today's sponsor, New Relic. All right, so you've probably experienced this before where you're just starting to fall asleep, and it's a calm, code-free peaceful sleep, and then you're jolted awake by an emergency page. It's your night on call, and something is wrong. But I have some good news because you have New Relic, which means you can quickly run down the incident checklist and find that problem. So let's see, our real user monitoring metrics look good. And that's where New Relic measures the speed and performance of your end-users as they navigate the site. But it looks like there's an error in application performance monitoring. If we click on the error, we can find the deployment marker where it all began, roll back the change, and, ooh, problem is solved. We can go back to bed, back to sleep, and back to happy. That's the power of combining 16 different monitoring products into one platform. You can pinpoint issues down to the line of code so you know exactly why the problem happened and can resolve it quickly. That's why more than 14,000 other companies, including GitHub and Epic Games, use New Relic to improve their software. So you know that next late-night call is just waiting to happen, so get New Relic before it does. And you can get access to the whole New Relic platform and 100 gigabytes of data free forever. No credit card required. Sign up at newrelic.com/bikeshed. That's newrelic N-E-W-R-E-L-I-C .com/bikeshed, newrelic.com/bikeshed. CHRIS: Well, similar to your email adventures, I continue on my search for the perfect to-do list. It's not going great, if we're being honest. [laughs] To be clear, because I've mentioned this on a few different episodes, I'm not spending much time on this at all, some but not much. And so it's not really moving. But there are two interesting things. I took a look at TickTick, which was one that I mentioned in the past, a tool for this. It seems good. It seems like an intersection between things, which is what I'm currently using, Todoist, which I've used in the past, and some other tools. So I think I'll probably explore that a little more. It seems like a good option. Decidedly, the most interesting thing is a tool called Sunsama, which is different in some interesting ways but very interesting. So one thing to note about it is it's $20 a month, which is a lot of money for one of these tools because most of them are like, "We're $20 forever, and then it's free." And it's a surprisingly low-cost space. And so, they're definitely positioning themselves as a more costly entry. I would be fine with paying $20 a month for a tool if it really is like, no, cool, I feel great. I'm more productive. I'm happier when I'm not working, et cetera. But what's interesting is they seem to do a let's reach out to all the places that tasks can live for you. So there's your inbox for email. There's your Trello board that you've got. There are GitHub issues. There's Slack. There are all these different sources of potential tasks. And they do a really good job of integrating with those other tools and then allowing you to pull that list into Sunsama and then make each day you have a list. And those items can be like, this is a reference to a Trello card on that board. This is a reference to a Slack conversation over there. So I'm super intrigued by it. It's also got a very intentional plan your day mode, which I like because that's one of the things that I'm really looking for is at the end of the day, I want to clean everything up, make sense of all of the open items, and then reprioritize and set up for the next morning so that I can just hit the ground running. That said, I tried it, and it just didn't quite click. And I think it's one of those it takes some effort to understand how to use it. So I'm not sure that I'm going to get there. But it is super interesting because that idea of our work lives in all of these different tools these days feels very true. And so, something that is trying to act as a hub between them to integrate them is very interesting to me. Again, I haven't really gotten anywhere on this. I'm kind of just reading blog posts, as it were. So I'll report back if that changes, but -- STEPH: The search continues for the right to-do app. Yeah, that seems interesting. I don't know why I'm feeling hesitant towards it. I'm one of those individuals...you're right; there are so many tools. And the fact that they integrate with a lot of them seems really nice. I'm at the point where I just grab links to stuff, and I'm like, hey, if this is my priority, I grab a link to a Trello ticket, and then I just copy that into my to-do. I guess I like that bit of work over having to integrate with a bunch of different platforms. Because once you get used to integrating...I don't know; I'm just rambling. But I wish you the best on this journey. I'm excited to hear more. [laughter] CHRIS: Thank you. I will certainly report back. But yeah, nothing pointed to share at this point. But I do have something pointed to share on the hiring front, which is that we have hired some folks. STEPH: Hooray! CHRIS: Yay. So this has been a fun saga across a couple of different episodes. And in my mind, it feels like this much longer, more drawn out thing, but it's; actually, I think, come together relatively quickly, all things considering. We've got someone who's starting in a little over a week's time, and then someone else who's starting in, I think, two or three weeks after that. So that'll be great. Hopefully, we can transition into onboarding, which is a different whole approach. But hiring as a distinct activity can scale back significantly. As we discussed last week, I want to be in the always be hiring mindset but in the more passive mode of having conversations with folks, staying connected. And if a great candidate comes along and it's the right time, then bring them on the team but not actually actively reaching out and all that sort of stuff, which will be great. Because it turns out that takes a lot of time and also a lot of energy for me. Having those first conversations, going into it very intentionally trying to communicate about something, and there's a tone of salesmanship to it that is not my natural resting state. So I come away from each conversation being like, that was fun, but also, I'm drained now. Why am I so drained? So not having that be a thing that is filling up my calendar is great. And also super excited with the folks that'll be joining the team and to be able to now grow our little team and define the culture and the shape of the groups that we will be collectively. I'm excited for that work and what we can build together. So yeah, it's an exciting time. STEPH: That's awesome. Congratulations. Because yeah, everything you're saying sounds like it's just been a lot of work. So that's very exciting. There's someone that I was chatting with earlier today where they were talking about the value and the importance of understanding what your natural skills are and the things that bring you energy. And so you're mentioning there are certain activities that you enjoy them, but they're also draining because perhaps they are on the outer boundary of what you might define as your own natural skill or the things that get you really excited. And I found that all very interesting. It had me thinking about that today about where are the natural areas that I find that I get energy that are easier for me? And then making sure that I'm trying to prioritize my day so that I am more focused on the activities that just align with who I am and also that I'm engaged with and then also looking for ways to stretch. But they made the point that if you are always in a space where you are not using your natural talents, and you're always having to stretch, then that can be what leads to burnout. Versus if you're in that sweet spot, that zone of where you are using your natural skills, but then also stretching a bit. And I think there are some assessments and things like that that will help you then determine what are my natural skills, and what do I like to do with my time? I just like that style of thinking and recognizing, like you said, like, hey, I did a thing. It was fun, but I'm drained. So now I know that this is something that requires more effort for me. Like hiring, that's one for me. I really like interviewing. I like talking with people, but I'm so nervous for them because I know interviews suck. [laughs] I just have so much empathy for them where I'm like, this is going to be a hard day. We're going to make it as pleasant and positive as possible, but I know this is a hard day. And so I feel like I'm in it with them. And so afterwards, I feel that same relief of like whoo, okay, interview day is over. CHRIS: I don't know that I quite achieve the same level that you do but in no way am I surprised that that is your experience of hiring. And just to name it, you're a wonderful human being that feels for the people on the other side of the hiring table. Like, oh my God, this must be so stressful for you. It's so kind of you to be in that space with folks. But coming back to what you were saying a moment ago, that idea of, like, understanding where your strengths are and where they're areas that you're not quite as strong. And I think critically, the question of like which are the ones where I want to just kind of say no to? I'm like, that's fine. This is not going to be a competency of mine. And I'm going to just avoid that or find other people to work with that balance that out. So for me, sales is the thing that I don't think that's ever going to be my bag. I don't think I'm ever going to move in that direction, and that's totally fine. Whereas decisiveness, which I was describing, is like, I think that's the thing I could get better at. That is one that I don't want to sleep on that. I don't want to say, "That'll be fine. I'll just have other people make the decision." No, I need to get better at making decisions, making decisions with less information or more rapidly, having a bias towards action. All those things I think will be deeply beneficial. So I'm trying to really lean into that. Whereas yeah, again, the sales stuff I'm like, yeah, and there's plenty of examples of this otherwise. But I've also been coding a bit more this week, which has been lovely because the hiring stuff has ramped down. And that has freed me up amongst some other stuff that's been going on. And you know, I like to code, it turns out. It's fun. I just clack about on my cherry brown keys, and it's great. STEPH: Do you remember when we first got introduced to mechanical keyboards, and we had co-ownership of one of the keyboards? And we literally had days of where it was like your turn to use the keyboard. And then it was my turn to use the keyboard. How long did we keep that up before we were finally like, we should just buy our own keyboards? CHRIS: It was a while because we were working with a colleague who was trying out a Kinesis, I want to say, one of the split little bowl of keys. But yeah, we had a shared custody over a keyboard, and it was fantastic. I remember that very fondly. STEPH: The days that it was my keyboard, I would go to the office and be like, oh, today is my day at the keyboard. This is great. This is going to be such a wonderful day. [laughs] And now I'm just spoiled. CHRIS: It went on for a while, though. And this was something where we both obviously enjoy this keyboard. Why don't we just buy one of these keyboards? We totally could have done that. And yet, for some reason, both of us were like, no, but what if...I got to think about this. Again, decisiveness. [laughs] We come back to this topic of well; I had to really think about it. And then somebody got the 92-Key test or whatever it was in the office. And so I just went over and poked every one of those for a while. STEPH: Exactly. It was option overload where we're like, well, okay, we're going to buy one, and then you open it up and search, and you're like, oh, you want options? We have options. Do you know about the blues, and the browns, and the colors, and these different options? Like, I don't know any of this language that you're talking about. I just want to clackety-clack. So yeah, it took time. We had to do our research. CHRIS: And then I ended up on basic browns. So here we are. Let's see, popping back up the stack a couple of levels, hiring that went on for a while. Now it is less going on. Although to be clear, like I said, always be hiring. So if anyone out there in the world is hearing what I'm talking about with Sagewell or seeing any of the stuff that I'm putting on Twitter, which isn't much, I occasionally just post screenshots of my commit messages, which recently included better snakes as a commit message. [laughs] I have to dig into that or not. But we were just doing some snake case to camel case conversion. But the commit message was better snake, so here we are. Anyway, if any of that sounds interesting, please do reach out. But I'm excited to transition back to focusing more on the work. On that note, actually, I'm going to call it interesting things that is happening right now organizationally is; we are working with an external security firm to help with some...they helped us out with a penetration test when we needed that. And then they have stayed on retainer and are helping with various different configurations, taking our AWS S3 buckets and making sure those are nice and secure, and all that kind of stuff. But we've recently started to focus more on organizational security, specifically a bowl of acronyms. We've got SSO for single sign-on, MDM for something device management. I don't know what that first M is. I probably should learn it, but it's fine. That's why I've got help on this is I think they know what the acronym stands for. But so we're working on each of those. And on the one hand, they're probably going to be kind of annoying, like having to go through the single sign-on. It's a whole thing, and it's harder to sign into stuff sometimes. I mean, ideally, it's actually easier. But in my experience, it adds some friction at some points. And then MDM means that there's now some profile manager on the computer. So I can say like, "Every computer must have full disk encryption or else you can't use it. And we need a passcode, and it must be this long and those sorts of things." So it's organizational controls that I think are good for us having a robust security setup throughout the organization. But yeah, they're the sort of things that I think historically, I probably would have, as someone working in an organization, had been like, do we have to? Do we need these things? Couldn't I just do whatever? But now there's something about it that I really like. I'm trying to name it in my head, but I'm kind of like, I don't know. This feels like growing up as an organization. And there's always weird corollary that I've been thinking about with the Rails app that we've been building, intimately familiar with just everything that's going into it. And I know the vast majority of lines of code. I haven't written them all, but I've had an eye on all of the different features that we're building in. And it's hard to get out of that headspace where it feels like a bunch of pieces. It doesn't feel like a hole to me, even though it definitely is. But when does a bunch of boards that you nail together become a boat? To make a really weird analogy because that's what I do; it's a hobby of mine. But when does that transition happen? At some point, certainly. But that's harder for me to see on the code side. And organizationally, somehow getting these things in place feels like the organization sort of an inflection point for us, a growth point, which is I'm really excited about it. Even though they're probably going to mean a ton of annoying nuisance work for me because I'm the person in charge of making sure it all gets rolled out. And anytime anyone locks themselves out of an account, I have to help with that. And so it's probably just putting a bunch of annoying work on my plate. And yet, I don't know; I'm kind of excited about it. STEPH: I feel like that shows our roots in terms of how we approach projects that we work on where you mentioned do we need this? Do we need this yet? Because I feel that we're constantly as developers and consultants just we're trying to advise on the more simplified do we need this? Is this the right thing to spend the money on? How do we know? What are the metrics? What does success look like? And all those questions. So I feel like the way you just phrased all of that just really shows that sort of mentality that you grew up with in terms of checking in, and yeah, it's cool. Like you said, you're at a growth point where then it's like, yes, we are at this point that I've asked myself all those questions, and we're here. This feels like the right next step. CHRIS: I like the way you described it as that you grew up with, my formative growing years at thoughtbot. Mid-roll Ad Scout APM is an application performance monitoring tool that's designed to help developers find and fix performance issues quickly. With an intuitive interface, Scout will tie bottlenecks to source code, so you can quickly pinpoint and resolve performance abnormalities like N+1 queries, slow database queries, and memory bloat. Scout also recently implemented external service monitoring, adding even more granularity when it comes to HTTP requests and API calls. So give Scout a try today with a free 14-day trial and experience first-hand why developers worldwide call Scout their best friend. And as an added bonus for Bike Shed listeners, Scout will donate $5 to the open-source project of your choice when you deploy. To learn more, visit scoutapm.com/bikeshed. That's scoutapm.com/bikeshed. STEPH: Well, switching gears just a bit, we have a listener question for today, and this one comes from Stephanie. So not me, another Stephanie in the world. Hello, other Stephanie out there in the world. And they wrote in, "Hi, Steph and Chris, fellow software consultant here. And I'm wondering if you'd consider talking about how to craft a project estimate for a client on the pod. It's such an important aspect of consulting." Amen. I added the amen. "And I feel like I'm very much impacted as a project team member when the estimate isn't accurate." Double amen. So true. [laughs] "Would appreciate any and all thoughts, especially since it might be part of my job in the future. Thanks." I just realized I put us in consultant church by adding all those amens, but here we are. [laughs] CHRIS: I'm glad you clarified that they were additions by you and not part of the original question coming in. STEPH: Sure. I don't want to speak on behalf of Stephanie. So I have some thoughts on the matter. I think there are a couple of different ways that we can talk about this particular question because I think there are different formats as to when you're estimating and who you're providing the estimate for. But I'm going to pause because I'd love to see what you think. How do you go about approaching crafting an estimate? CHRIS: Sure. I'm happy to share some thoughts. And for a bit of context, this question came in to us, frankly, many months ago, but I did send an initial reply to Stephanie because I know that sometimes we take a little bit of time to get back to folks. So if ever you do send in a question, know that one of us will probably respond via email earlier, and then eventually, will make it on the show. And again, just to say, we do so appreciate when folks send in these questions. It's an interesting way to shape the conversation and a way to get topics that you're more interested in into the fold here. But so the two main ideas that I shared in my initial reply were, first, is an estimate really necessary? I think that's a critical question because an estimate implies that this thing is knowable. And as many of us, probably all of us, have found out at some point in our lives as software developers, it's really hard to do software estimation, like wildly difficult. And not just the thing that we'll eventually get better at it, which you do, but there's just some chaos. There's some noise in this work that we do that makes it so, so difficult to get it right. So pretty much always, I will ask, like, do we need to estimate here? What if, instead, we were to flip the whole question on its head and say, let's set a deadline. Let's say two months from now that's our deadline. And let's ruthlessly reprioritize every single week to make sure that we're building something that's meaningful, and we're getting there. And obviously, we have to have some general idea of what we're doing. Is two months a meaningful amount of time to build a rocket to go to Mars? Probably not. But is it enough time to build an app that can allow users to sign in and manage a simple list of items? Yeah, we can definitely do that, and we can probably add a bunch of more features. The other thing that I think is worth highlighting is there's a bunch of stuff that is table stakes and very easy to do. But I would, whenever doing estimation, emphasize unknowns. So, where are the external integrations with other systems? Where are the dependencies that rely on other folks to provide some inputs into this process that we can't be certain where there'll be? In my experience, the places where estimates go awry are often these little intersection points that you're like, well, this will probably take a day, maybe two. And it turns out; actually, this can somehow balloon into a month. That's not a thing that feels comfortable saying in an estimation process, but it is definitely real. I've seen it happen so many times. And so it's those unknowns. It's those little bits that I would emphasize as part of the process if you do need to do an estimate and say, all right, here's the boring stuff. I think we can do that pretty easily. But this part, I don't know, it could be a week, could be three months. And frame it in that way that there is this ambiguity there. Because if someone's asking you for an estimate and they're looking for like it is seven days and two hours exactly, it's like, well, that's not realistic. That's not how this thing works. Unfortunately, I wish it did. But pushing back and changing the conversation is the thing that I have found valuable. I think there's some other really interesting stuff in here around the team dynamics that Stephanie is talking about. But I want to send this over to other Stephanie to see your thoughts because I'm super interested to hear what you have to say as well. STEPH: Oh, I like how you hinted at the team dynamics. Yeah, that could be a fun one to circle back to. So I love how you called out highlighting the unknowns. There are a couple of ways that this comes to mind for me. So there's the idea of the weekly or the bi-weekly estimates that we make as developers and designers. So let's say we as a team are getting together to focus on a chunk of work and decide what we can and can't get through. And that feels one of those the more you get to practice it more frequently; you get to ask a bunch of questions. And that feels like a good rehearsal and exercise of how to go through estimates. And I know you and I have pretty similar strong feelings around how those estimates are then treated by the company. They should really just be used for the team to talk through the complexities in the work to be done versus used to communicate outwardly as to this is when it's definitely going to ship. So there's that more immediate practice of providing estimates. And then there's the idea for more of a consultancy or a company, and someone is coming to you, so thoughtbot being a great example of then how do we work with teams that are looking to come to us and gain an estimate for getting a certain feature implemented? So actually, I went to the source on this one. I went to Josh Clayton, who does a lot of the conversations for the Boost team when it comes to talking with clients and about the potential work that they would like to be done. And mostly our work is often teams will hire us. They have specific goals in mind, but they're really looking to hire ongoing development and services. So they really want to add to their existing staff. And then it's going to be an ongoing relationship versus a hey, we need you to quote us for how long it's going to take to implement this particular feature. And on that note, we don't do fixed-bid work. So we don't say it's X dollars for specific features. But on the realistic side, customers are often capped by a budget. And so that estimate is very important to them because it could be a difference between it's a go versus a no go. So if you have larger companies that are like, "Yep, we want to engage with thoughtbot. We really just want additional development power and design services," that's great. For those that are smaller, it could be an individual product owner, and they need to say, "I really want this feature, but I only have this much money. And frankly, if I can't ship it by this time, I'm not going to do it because it's not worth the investment to my company." And then, in those cases, those are the ones that we're going to spend more time with them to talk about what does the fallback plan look like? And what's our opportunity for simplifying the features? And Josh, in particular, referenced this as systems thinking. So he will go through the idea of drawing out the set of steps, understanding the complexity of the different screens. So what are the validations? What are the external dependencies? What is owned by us and what isn't? What is the likelihood that we're going to get permission to simplify or remove complexity? And even then, when we start to provide some estimates, it's going to be in weeks. It's not in hours; it's not in days. It's going to be in a slightly larger time frame. And then we're also going to spend more time in the discovery phase to say, okay, well, we know you need to fix this particular issue, or you need to integrate with this particular service. So we're going to need to ask a lot more questions about your codebase. What problems have you already run into? Have you tried to do this before? Do we have experience doing this? Is this something that we can lean on and ask someone in the team? And, say, how long do you think it would take for us to work on this? And that's knowledge that isn't privy to everybody. It depends on where you're at in your career as to like, oh yeah, I've done this like five times before, and I know exactly how this stuff can fall apart. I know where the complexity lies. So I think that's why estimation is so difficult is just because it does often pull from that existing experience. And so, if you don't have that experience for a particular set of work, of course, it's going to be hard to estimate because you just don't know. So that was a very broad scope of as day-to-day developer and designers; I feel like we're constantly getting practice and estimating and communicating the progress of our work. And then on the larger scope of if you are a consultant who's then looking to give estimates to clients, then understanding what other need can you sell them? Just ongoing development services. Or, if they are a smaller team and very focused, then what legwork can you do ahead of time to de-risk the project? And then understand how much control you're going to have to be able to simplify as you learn more as you go. Because you're going to, you're going to uncover some things, and you're going to learn some things. And what's that collaboration going to look like? I do have one more concrete example I can provide around some of the smaller projects that we take on. So when we are helping someone that's, say, getting a new product out to market, then we do have a more deliberate three, four-phase approach where we first focus on discovery, and ideation, and validation. And then, we move on to iteration and then launching. And I really like how you said about providing a deadline because then that helps us scope aggressively as to what is the minimum thing that we can get out into the world that will be valuable? And then there's usually some post-launch support as well. But that's often how we will structure those smaller, more specific engagements. CHRIS: I think one of the critical things that you highlighted in there is that thoughtbot doesn't do fixed-bid work. So we're going to do these 20 features, and it's going to take four months. thoughtbot does not do that, and frankly, that's a privilege to be able to take that position and say, "No, no, no," we're not going to work that way. But it is, I think, a trade-off. It's not just something that thoughtbot does to be like, listen, that doesn't sound fun. So I'm not going to do that. It's a trade-off. Not doing that comes in concert with saying, "But weekly, we're going to talk about the work that we have done and the work that remains and constantly, ruthlessly, reprioritize and re-decide what we're doing." And it's that engagement, the idea that you can have a body of work, look at it and say, "Yeah, that'll take about six months," and then go away for six months, and then come back with the finished software. Our strong belief is that that's not the way good software gets built. But instead, it's a very engaged team where the product owner and the development team are in constant communication about each of the features that are being developed. And then again, ideally, on a weekly cadence, coming up for air and saying, "How are we doing? Are we moving in the right direction? Are we getting towards the goals? If not, do we need to simplify? Do we need to change things?" And similarly, as I mentioned deadlines, I feel like deadlines is probably a word that many people think of as very bad because deadlines often come with also a fixed scope, but that can't happen. That's two constraints, and you can't have them fighting that way. But a deadline can be super useful as a way to say we're going to put something out there in the future and say we're heading towards that moment. And let's, again, cut scope. Let's change what we're building, et cetera. But critically, not say, "We got a deadline and a fixed scope. We're going to do that." And so it's, again, just ways to gently shift the conversation around and say, what if we were to look at this from a different angle? Because just having a pile of work and saying, "That'll take six months," I've never seen that play out. STEPH: Yeah, to me, deadline is a bad word when the deadline is set by a team that's not doing the work. So if you have leadership or if you have someone else that is setting this deadline and then just passing that down to someone else to then fulfill, regardless of the feedback or how things are going, then yeah, then it can be a nasty thing, which I think is a little bit of in that question that you picked up on that you highlighted where there could be some interesting team dynamics that Stephanie called out, highlighting that I'm very much impacted as a project team member when the estimate isn't accurate. And I'm making some assumptions here because I don't actually know the exact situation that Stephanie is experiencing. But it sounds like someone else externally is setting these team estimates. And so then you're handed this deadline, and then stuff goes wrong, but you're still pressured to meet this deadline. And I've certainly been part of projects that are like that. And then that is one of the number one things that then often comes up in a retro or like, we don't have control over these deadlines, or we don't know why these deadlines are being set. And then people are working extra hours and working nights and weekends to then meet this arbitrary deadline that none of us signed up for, and that's just not fair to treat deadlines in that way. So full-heartedly agree that deadlines can be a very positive thing, but they need to be set by the people doing the work. And then there has to be discussions and updates about how is this going? Do we have control to simplify this? We thought we could do this with this particular external provider. It turns out that that's a nightmare. Is there another provider we can go with? Can we ship this incrementally? Like some features, you can't. They may have to go out wholesale. But is there a small chunk of this that we can deliver that is then a success that leadership and others can brag about? And then we can keep working on the rest of it. So it's always identifying what are the smallest wins, and how do we get there and getting buy-in from the team? Going back to something that you said earlier around, it is a privilege, where so as thoughtbot, we don't do fixed-bid work. And that is a nice thing for us to be able to focus on. But for people who do need to do fixed bid work and are relying on that, I think that often requires more legwork. And maybe that becomes part of your estimate. I'm just making up how I might approach this if I were trying to do fixed bid work. But there's a discovery phase that's very important. So maybe the first part of your estimate is I need to really understand the feature and see the different screens and know what materials we do or don't have. What does the codebase look like? Do I feel like this is a codebase that I can work quickly in? And is it going to be hindersome for me? But answering a lot of those questions to then help me paint a picture of, like, okay, this is a feature that I've implemented before, so I feel pretty confident that I could do this in a month. And then also communicating that this is my estimate but just know it's an estimate. And I will continue to update you each day as to how things are going or each week as to how things are going, and things may adjust. And we can always talk about ways about simplifying this. But I think that's how I would go about it is; frankly, it's going to require more legwork for me to feel more confident as to then telling someone as to how long I think the work will take. I think that's a nice, broad scope of the different types of estimate work to be done with the general idea of if you can avoid estimates and go for more frequent updates, then that's wonderful. But then, if you are forced into a corner where you need to provide an update, then just do as much research and honesty as possible and then still include the frequent updates. CHRIS: Yeah, that I think summarizes it quite well. STEPH: As a side note, it's been a lot of fun to feel like I'm referring to myself as a third person as Stephanie is working through this problem. So that's been novel. But yeah, thank you, Stephanie, for the great question. I hope that was helpful. On that note, Shall we wrap up? CHRIS: Let's wrap up. The show notes for this episode can be found at bikeshed.fm. STEPH: This show is produced and edited by Mandy Moore. CHRIS: If you enjoyed listening, one really easy way to support the show is to leave us a quick rating or even a review on iTunes, as it really helps other folks find the show. STEPH: If you have any feedback for this or any of our other episodes, you can reach us at @_bikeshed or reach me on Twitter @SViccari. CHRIS: And I'm @christoomey. STEPH: Or you can reach us at [email protected] via email. CHRIS: Thanks so much for listening to The Bike Shed, and we'll see you next week. ALL: Byeeeeeeeeee!!! ANNOUNCER: This podcast was brought to you by thoughtbot. thoughtbot is your expert design and development partner. Let's make your product and team a success.Sponsored By:Scout: Scout APM is leading-edge application performance monitoring designed to help Rails developers quickly find and fix performance issues without having to deal with the headache or overhead of enterprise-platform feature bloat. With a developer-centric UI and tracing logic that ties bottlenecks to source code, you can quickly pinpoint and resolve performance abnormalities -- like N+1 queries, slow database queries, memory bloat, and more. Scout's real-time alerting and weekly digest emails let you rest easy knowing Scout's on watch and resolving performance issues before your customers ever see them. Scout has also launched its new error monitoring feature add-on for Python applications. Now you can connect your error reporting and application monitoring data on one platform. See for yourself why developers worldwide call Scout their best friend and try our error monitoring and APM free for 14-days, no credit card needed! And as an added bonus for Bikeshed listeners: Scout will donate $5 to the open-source project of your choice when you deploy. Learn more at scoutapm.com/bikeshed.New Relic: Services down? New Relic offers full stack visibility with 16 different monitoring products in a single platform.Support The Bike Shed