Artsy Engineering Radio
Artsy Engineering Radio
Shipping Project Retro with Jackie
Jackie Potts joins Steve Hicks to do a project retro. They discuss the chronology, the challenges, and the joys of a recent project integrating with a third-party shipping provider. They maybe make each other cry.
Feel good about this?
Jackie Potts:Definitely Steve. We can edit, right, after?
Steve Hicks:Yeah, we can just delete the whole damn thing. If we want to.
Jackie Potts:We should leave this in, I think it's fun
Steve Hicks:Hey, friends, welcome to another episode of Artsy Engineering Radio. I'm your host this week, Steve, Steve Hicks, you've heard my voice before, engineer at Artsy. I'm here with my good friend Jackie. Jackie, would you like to say hi.
Jackie Potts:Hey, everybody. Um, just before we start, I just want to say, Steve, thanks so much for taking the time to invite me here. And I'm a big fan of yours.
Steve Hicks:And I of yours.
Jackie Potts:You're such a community driven Dev, and really are all about sharing and caring. And that's definitely felt, as we've been on the same team together and just listening to the podcast and everything. So
Steve Hicks:So we're changing the name of this episode to Steve is crying. So that's the sort of, let's talk about things that make me cry. Yes,
Jackie Potts:I think we might have just became best friends. So that
Steve Hicks:I think we actually have been for a while. Jackie, I've wanted to get you on this podcast for a long time, we all have. Because you're also a super inspiring engineer. And honestly, one of the most fun people to work on a team with pairing with you is just, it's always a joy. But I think that...do you wanna cry, also?
Jackie Potts:I got tissues, I got tissues.
Steve Hicks:So we're here to talk about a project, which is kind of fun, because we don't, I don't think we've done very many of these episodes. But we're thinking about this as like a retro about a project that the team that you and I work together on, which is the partner experience team. That's the team that kind of works with partners who are art galleries and art fairs to get their artwork up on Artsy. But we did a project. I mean, we started it a while ago, and it's still ongoing. We did a thing. Yeah, we did. And we're doing it. But our goal was to improve the shipping experience on Artsy. So do you want to maybe just kind of introduce the project, talk about what it is why we went forward with it.
Jackie Potts:Sure, that sounds good. I think there have been some sprinklings, about the project on the podcast already. But yeah, kind of, as you said, the idea here was to build out a better logistics and shipping experience for our E commerce marketplace. So our team typically deals with the supply side of that chain. So like building tools for getting artworks onto the Artsy platform, getting them ready to be sold on Artsy that sort of thing. This project is kind of special in that we did an end to end to marketplace, two sided marketplace projects that had an integration with a third party. So yeah,
Steve Hicks:integrations are always fun. And I definitely want to talk more about that, as we get towards the end of this this episode. But maybe you could talk maybe about the beginnings of the project. Just to be clear. You've been involved with this project for basically the whole time. I think, as long as you've been on PX this project has been going on.
Jackie Potts:It's yeah, it's it started a bit ago, I think we started kind of ramping up maybe March of last year. And it's also your first project on the team, right, you had recently joined px partner experience team. And I'd be curious to get your feedback and experience.
Steve Hicks:That's not what we're here for, you're here to
Jackie Potts:You're still here. So that's, I guess the kind of give, the main problem we were looking to solve on the part well, I guess there's maybe two main problems. On the partner side, we didn't have any like dynamic options for shipping. So if a partner lists $150 artwork on Artsy, for example, they can add a flat fee to that to cover the shipping costs. So in this old world, the partners responsible for packing shipping, any cost associated with that they can add to this order if they would like. But we only offered two options. One was like, if this ships domestically, that's the fee they'll have. Or if you ship internationally, that's the fee. So for example, if I'm living in New York City, and I order a piece from a gallery in Buffalo, New York, that gallery has a $15 shipping fee for that price. But what happens if a collector from California checks out with that piece, the gallery still can only charge the collector that$15. And same thing internationally like the difference between a gallery in Canada to New York versus a gallery, from Canada to Berlin, Germany, like they can only charge one flat fee for the shipping rate. So there's a bit of discrepancy there. And that partners could potentially be fronting up a lot of the shipping costs. Also, as far as like, safe shipping goes a gallery has to make sure they have the right materials. So there's like costs for materials and all of that, for fragile works, glassworks, that sort of thing a lot to kind of think about. And on the collector side of this whole experience, when a collector checks out at Artsy, we didn't have too many notifications after your order was like approved, for example, or confirmed. So a lot of the times the collector was in the dark about the whereabouts of their package, if it was being shipped, we didn't really have access to updates or a tracking information at that side of it.
Steve Hicks:So it sounds like it's like a super complex problem, right? And is it true that maybe the way we approach shipping from the start was iteratively, like, we're gonna do this simple thing. And we're gonna try to do this in a way that we account for a lot of things, but it's not going to be the right way. And now an integration is maybe a way to like, get more -- catch more of those edge cases, and let someone else do that work for us. Is that kind of what happened?
Jackie Potts:I think so I think definitely, having this integration partner was a huge win for us, just because you know, they are the experts in this field. And they already have, you know, kind of a network of logistic partners throughout the world. This partner is ARTA, which is a white glove shipping service that specializes in Fine Art shipping. So, you know, they're kind of the leaders in that space, which is really great to have them as a partner here. And yeah, I think we've learned a lot on both sides of that.
Steve Hicks:White Glove is one of my favorite phrases that has come out of this, because I really hadn't heard the phrase white glove before this project.
Jackie Potts:Yeah, it's amazing. It's fun to think about that. It's really cool how, you know, fancy some of the stuff goes because they offer like temperature control, and like climate control shipping for like some really high end and delicate pieces. So it's definitely a world I would like to be involved with more.
Steve Hicks:Yeah, significant investments that people want to take very good care of.
Jackie Potts:If finances allow, yes, of course.
Steve Hicks:Yep. Cool. So that's a little bit about the why. Maybe we can talk about the how. How did we break this problem down? Like it's huge. There's a whole lot of stuff that's going on? How did we approach that?
Jackie Potts:Yeah, good question, Steve. I think kind of like the biggest, maybe the first tier of this is Sarah, you all know, Sarah. But Sarah led a few user story writing sessions with the whole team. So we kind of like just to get a frame of reference. Everybody was in a room together, I think we had like 90 minutes. It was like super fun. But we kind of defined stories just like brainstorm, No Bad Ideas and Brainstorming kind of thing for our three user types. So our partner, the whole supply side of this, our collector, the whole buyer side, and then admin interventions. So we have a lot of admin tools for, you know, trust and safety team marketplace support and like, also, how do we communicate with our logistics partner this way? So we kind of did a big brain dump. And from that, we kind of went through each card, drew like a water line, like what's the most minimal thing we can do? And what things can we potentially do in parallel? That's kind of where a lot of those ideas came from. So from that we were able to define, like, certain epics. And then we kind of knocked out some projects in parallel. So like, I think the first thing we did, the order history screens and experience in the mobile apps. So that gave time for our iOS and Android apps to get updated while you know more of me and Anson, and also you to Steve and Anna, we kind of looked at the docs for this integration. So we had time to really do some research on, you know, the ins and outs of the back end integration. And how that would kind of play out in our Rails apps here.
Steve Hicks:Yeah, I remember there was a lot of exploration with the API just like, what does it do? What can we make it to? And how can we use it? I also remember that when I started on the team, I think that there was already in existence, a tech plan. And I can't remember the chronology, I don't remember if the tech plan came before the user story sessions, or the user stories sessions came first. I'm pretty sure the tech plan came before we did any of those API explorations? what's your what's your memory of this?
Jackie Potts:Oh, good question. It's, it seems like so long.
Steve Hicks:Do you have any memory of this?
Jackie Potts:I would assume the tech plan came first. Sure. And then, you know, user stories and tech, you know, API kind of exploration, that sort of thing. But the tech plan, when we went into the tech plan, we did have a lot of unknowns. Have we talked about tech plans before? Just give like a quick,
Steve Hicks:we've kind of introduced them as like a...it's our way of writing down our thoughts, before we approach a big project. I don't know Is there any other description, you want to give of a tech plan? That might be more clear,
Jackie Potts:I'll say maybe my favorite part of the tech plan is the risks section. So there's a nice area where you kind of have space and time to like, investigate risks. And like, going into an integration with a third party, there's a lot of unknown. So unknowns are risks, right? Like, if we learn something midway, that's something that can augment time or resources or anything like that. So I think the tech plans kind of forced you to get ahead of some of that, which has been really great. And we kind of did that to like, dig into some legal things. And yes, learned a lot about laws and taxation, which is very interesting.
Steve Hicks:Yeah. Yeah. Lots of conversations with the legal team for this stuff. And I don't know, do you want to take credit for the tech plan? Because I think that you should take credit for the tech plan. I mean, it looked like it was, it was quite fantastic. It was very thorough, very well thought out. You really went deep on a lot of different stuff. So I guess Thank you, because I think I'm really set the tone, right. Like, having all of those details and identifying all those risks and digging into it was super helpful to give us if nothing else, just a sense of confidence that we had accounted for things.
Jackie Potts:Right. That's high praise, Steve. And thank you. I will say though, there were a lot of people we chatted with when, you know, kind of drying up the tech plan. And I know Anna, shout out to Anna and I was like pairing every step of the way, just very present in all those early day things. So it's fun.
Steve Hicks:That's the thing that I love about our team is the the pairing sessions. So now we've got the tech plan. We've got these user stories, we've done some exploration, we start building the thing, right? Like we have to start digging into the epics that we defined and cutting things off maybe...are there any highs or lows or any significant learnings that you remember from this project along the way that you could share?
Jackie Potts:Sure, I'll say, I'm very proud of us. And of the relationship we've kind of built with this third party company slash team. I think, well, I've like never had the experience of being so friendly and just having so much support from, you know, a third party and having a lot of FaceTime. So that's been really great. And definitely a huge part of this whole integration being successful. I think in the early days, the Artsy side kind of intentionally like maybe showed vulnerability first and like, we kind of were sharing our designs right away with our data asking them for feedback a lot. Like we kind of mentioned before, they're the logistics experts, they have experience in the fine arts space, they have so much, you know, we're making assumptions maybe you know, they have concrete kind of things, they can give us feedback on. And raise these really intricate, nuanced points. So, in the beginning, we kind of built a lot of maybe railroad tracks to have, you know, maybe this space where we can openly communicate. And it's free for feedback from both sides. It's free for just like warmth and empathy on both sides. And just like have anyone feel comfortable from our side to reach out to them or them to reach out to anyone on our side? And I think that's really been great. I mean, I've never really worked with a third party that is so kind. And I don't know, maybe we have something to do with that, hopefully. And they feel it on our side,
Steve Hicks:I think, I think definitely. Yeah, I totally agree. So I think the takeaway is really just be really nice and work with really nice people.
Jackie Potts:There's ways to show and build trust, early, you know, so that now, I have this space where there's no barriers to reaching out? And even within our team, I think like, our team is really good at this. I can approach anyone at any time for help and totally feel not judged or not like I'm bothering someone or, you know, this kind of, like insecurities that you have with communication. So yeah, that's been very, very interesting to play around in that space. And to feel that.
Steve Hicks:Yeah, fair. Very true. It is cool how, I feel like when we would hit milestones on the project, the celebration was collective between us and the people at ARTA, it was just like, everyone was so excited about it. And that was really fun.
Jackie Potts:Yeah, it's different. Nice to see too. Yeah, it's exciting.
Steve Hicks:Are there any other things that you remember being things that we learned along the way with this project?
Jackie Potts:Yeah. Yeah.
Steve Hicks:So I guess one thing, one thing I'll throw out is, I think we, I think there were some process things that we tried, that we did differently on the team that we hadn't done Previous to this. One of them that comes to mind is, for a while, we kind of broke the work down into work streams, and then kind of assigned people to those work streams and called them squads. And had rather than our I think, how many engineers were on our team at the peak? Like eight, nine? Yeah, maybe even 10? BiG crew. Yeah, it was a big crew for a while. But yeah, rather than trying to have like one work stream, or even just one team, just all of those people focus on one thing or everything, breaking it into those squads, or those individual work streams was a nice way, I think to to just help people feel more invested in like the part of the project that they were working on.
Jackie Potts:Yes. So that kind of ties into the how do you make your worlds I guess, like smaller and less overwhelming, like thinking about the whole span. If you think about the whole span of this whole project, you will go to a different space. It's very stressful.
Steve Hicks:Too much.
Jackie Potts:So making these little digestible chunks was, yeah, really great. And I think Kaja and Leandro M when they joined the team, this was kind of like their idea to break down some of these epics into little squad groups, where you know, in sprint planning, we would, people would volunteer, and people naturally went into places that they enjoyed more so like more front end focus versus more back ends and with freedom always to move and rotate around from one squad to another. But it was so nice, it was so fun. I think like, I can speak from I think I mostly hung out on the back end squad but like, working with Leandro and Kaja, we were just like hitting this groove. It was just like, really great. I would wake up to PR reviews and then, you know, because they're in Berlin, I'm in New York. So we were kinda like the sun never set on the backend crew. It's like, really fun.
Steve Hicks:Just kept rolling.
Jackie Potts:Yeah, just like, nice to spend time focused with people who have the same context, there's like no time, you know, you don't have to catch people up. Everyone has the same level of context in everything that's going on for that epic. Yeah, I thought it was a great way to not be overwhelmed by the grandness of everything, for sure.
Steve Hicks:Yeah, make it smaller, and even across squads or back end front end or whatever. Like I do also remember many times you just spinning up a zoom and saying, Hey, I'm going to work on this thing and try to move this forward. And it was always great to just like, Okay, I don't really know what's going on in your squad right now I'm just going to drop in and hang out for an hour and find out. So that was really fun. And I think that I've definitely I know, I've expressed this in retros. But that's one of my favorite things that you do that I want to do more of, is just like, start up that zoom and say, This is what I'm working on. Come hang out with me.
Jackie Potts:Oh, yay. Yes. They started as maybe cries for help. And then over time became just really nice to...I just love pairing, and I just love having people around to kind of keep you emotionally and technically focused and that sort of thing.
Steve Hicks:Yeah, you get a lot of knowledge sharing to, spreading all the knowledge across everyone on the team.
Jackie Potts:Yeah. And just so good for bonding to, just to have I mean, we started this project at the peak of the pandemic, so things were just like, everyone's cups were a little empty, and how do you...How do you regain some of that, you know, trust and just like, friendship? There's something I think, I personally was really starving for. But yeah, those were fun. And it started as pairing and then they kind of turned into coffee hangs, and then beer hangs, the day we launched we were fixing a bug and kind of being really silly and having a beer while we were doing it. So it was just like, Yeah, super fun.
Steve Hicks:But I actually had kind of forgotten about the, I guess I'd call them a launch party. But basically just everybody hanging out together on a zoom as we flip the switch, or as we ran a script to do some some things and enable everything. That was really fun, too. I liked that.
Jackie Potts:Yes, that's true. And even people from other teams came, it was like such a nice, nice space to feel super supported.
Steve Hicks:And now I suddenly am remembering, that's where we saw the celebration too, with with the people from ARTA. They came and joined us too, as we turned stuff on. And so then it was, we were all in one place to celebrate that launch. And it was great.
Jackie Potts:Yeah. It's really nice.
Steve Hicks:I would definitely do that again.
Jackie Potts:Yeah, definitely. All the launches. We were all together, which is really, really special.
Steve Hicks:What about the other side? What, uh, what are some things that didn't go well, or things that you would do differently? Ooh, don't say nothing.
Jackie Potts:I didn't prepare this, Steve.
Steve Hicks:That's why I asked it.
Jackie Potts:Let me think. I do struggle a lot with not getting overwhelmed. So I feel like I have a tendency to overwork, which maybe showed in the beginning parts of this project. I remember talking to you, Steve, about it. And you would always say, was it worth it? And I was always like, that would shake me up into reality a little bit. Steve, I worked this weekend, that sort of thing. Well, was it worth it? Um, so just kind of, I don't know, keeping an eye on mental health and how that trickles into the team?
Steve Hicks:For sure.
Jackie Potts:Especially, I don't know. It was a hard time. Just like, in everyone's lives and stuff. And you're probably asking, when have we learned technically right, and I'm making it a therapy session.
Steve Hicks:This is great, we can do the therapy session. That's cool with me. That's my preference, actually.
Jackie Potts:We can process this, that's for sure.
Steve Hicks:No. So I mean, especially at the beginning of the project, when we were thinking about how big it was, and how...I don't really know what we had in terms of deadlines at the beginning, but you're thinking about how you want to get this thing done. And there's so much of it to do. And it's really hard to figure out, what is the sustainable pace right now that's going to help us meet that goal. And so you make decisions like, Well, I think I really want to get ahead right now, so that we aren't behind in two months. And that is tough. And I don't know. I mean, maybe I know we're always trying to improve how we estimate things at Artsy and I think it's, it's even hard to estimate a single thing that you're going to do let alone a project of all this massive scope. So I don't know maybe that's something we can improve on.
Jackie Potts:Yeah, you know, everything's a trade off, right. And it's all about how you balance that. And think about risk mitigation. And I don't know.
Steve Hicks:It's just lots of things. It's lots of things,
Jackie Potts:It's a whole thing. We'll need a part two for the what we learned.
Steve Hicks:So another, this isn't a learning, I guess. But one of the things that I always think about in this kind of project is, it's an integration and integrations are always hard. And it's, it's kind of like, you want to do this one part of your app better. But you don't want to devote an entire section of your organization to this one thing. So you think, okay, let's find somebody that does this. And we'll integrate with them. And that's their entire business model, right? Or their entire business, and they'll just take care of it. And we just have to integrate with it. And it'll be fine. And it ends up being it typically ends up being like 80% of the work is great, and super easy. And then that last 20% is just like we didn't know that this would be a problem, or this actually completely impacts how we're doing this. And we need to kind of shift and...they don't have the features available that we need, or they're coming later. And we just have to wait for it or whatever. And so there's this difficult...it's always difficult, I think, to look back at an integration and evaluate, was it a good idea? Or was it a bad idea? In this case, when you look back at the ARTA integration, How do you feel about it? Do you think it was like a slam dunk no-brainer? Do you think it was somewhere in the middle? Do you think we should have just done all this shipping stuff ourselves?
Jackie Potts:Oh, I definitely think it was a successful integration. Just thinking about this world is, it's so nuanced. And so so intricate. Part of what ARTA does is on some of these flows, they'll actually go to the gallery or to the warehouse or the artists stored and pick it up, do like a condition report, you know, package it, and then they handle the shipping themselves. So they have so much knowledge on fine art packing, fine art fragility, and that sort of thing. I I think that was really, definitely needed for this space. Also, we're kind of exploring, like expanding this to international markets. So for our current launch, that's live, we're doing just continental US shipping. And when you have to cross international borders, it's like this whole whole new world, like mind blown, like certain object materials can't cross international borders, things like that. Things that we're not like the experts in. So I think it's really great. In a space that's so huge to have a partner in that world for sure.
Steve Hicks:Yeah, well, it definitely would have been a lot for us to recreate all of the things that they provide for us. But also, I think one of the reasons that it worked out so well and we look back on it so favorably is like we talked about earlier that communication between us and them and the collaboration and the teamwork, and, you know, their willingness to answer questions and change things and adapt based on our needs. It comes down to find a good partner. That's it.
Jackie Potts:Yeah, find a good partner, have empathy, be kind, be patient, things will work out.
Steve Hicks:Jackie, is there anything that you would like to discuss about this shipping project that we haven't talked about? Anything we missed?
Jackie Potts:Uh, just kind of props to the team. PX I'm looking at you. I've never really worked on a project and at the end of it been so excited and so bonded with my teammates, you know, like I feel like we kind of all were super united and getting this done and I was like, just like kind of the things that kept us going right? Each other and like, you know, we have this life event, this core memory we all have now and we're all a global team too. Yes, super cool. So props everyone, thanks for putting up with me.
Steve Hicks:Hardly what it is, hardly what it is.
Jackie Potts:Yeah, we need to come hang in Buffalo.
Steve Hicks:I'm sorry to hear about your Bills in the playoffs Too soon. Anyways, okay, Jackie, I guess this where we rather than shift to being an NFL playoffs podcast, we wrap it up. And so I will say thank you for hanging out and thanks for being amazing. And I appreciate working with you.
Jackie Potts:Oh my god, it's the best Steve, working with you, and thanks for everything. Cheers, see you online.
Steve Hicks:Cheers. Thanks for listening. You can follow us on Twitter@artsyopensource. Keep up with our blog at artsy.github.io This episode was mixed and edited by Alex Higgins and thank you Eve Essex for our theme music. You can find her on all major streaming platforms. Until next time, this is Artsy Engineering Radio