Transcript

Chris LoSacco: You could have labelmakers that print out those, you know, like, little strips? That you could…

Gina Trapani: Yeah. Yeah, yeah, yeah.

Chris: There was a strip with the name of the server…

Gina: There was a little label! It said, like… 

Chris: …like, taped on the top of the box!

Gina: (Laughs)

Chris: And… but… nobody thought to…

Gina: But no one knew what that meant. 

Chris: Right. 

Gina: No one was like, “When I unplug this…” 

Chris: Right. “-Dev-02” was gonna go offline.

Gina: Like, the entire engineering team is going to be halted in their tracks.

Chris: Right.

Gina: They can no longer do their work.

Chris: Right.

[POSTLIGHT INTRO MUSIC]

Gina: Hello world. Welcome to the Postlight podcast. I’m Gina Trapani, CEO of Postlight. As always, I’m joined by my partner and president of Postlight, Chris LoSacco. Hey, Chris.

Chris: Hey, Gina.

Gina: How are you today?

Chris: I’m great! I’m great, how are you?

Gina: Good. I was thinking about the earliest part of my career and how it sort of… it sort of changed. I think… you and I share something in that, I think deep down… and you tell me if this is correct or not.

Chris: (Laughs) Okay.

Gina: I think deep down, each one of us is, like, an IT person.

Chris: Ooooh.

Gina: Like, a kind of a little bit of an IT person.

Chris: You’re so right. 

Gina: It’s true, right?

Chris: What do you mean?

Gina: (Laughs)

Chris: I have an intrinsic sense of what you mean, but let’s make it explicit. Like…

Gina: Yeah.

Chris: When you say IT person…

Gina: I mean, both of us love empowering people to do their job…

Chris: Yep.

Gina: …and giving people the tools they need to do their job, right? I mean, what we do is, we develop platforms and products so that people can get things done better and more easily, and have a good experience along the way.

Chris: Yep.

Gina: So I think there’s something inherent – and really in anybody who works in technology, whether that’s enabling people to use technology, which is kind of on the IT side, right? Or building new products, that there’s this desire: “I want to get good tools into people’s hands that they can get their stuff done.”

Chris: Yep. Exactly. We should step back. Let’s define IT.

Gina: Let’s define IT.

Chris: Right? IT stands for Information Technology.

Gina: Yep.

Chris: It can mean kind of anything to anyone. But when we say IT, I think we are talking about large-scale ownership of infrastructure and configured tooling that gets rolled out across an organization.

Gina: That’s right.

Chris: Is that a fair definition?

Gina: Yeah. That is a fair definition. You know, I started out kind of in IT. I worked at the help desk through college. I mean, the help desk is one sort of part, I think, of an IT organization.

Chris: Mhm. Sure.

Gina: My first, like, job out of school – and I’m dating myself here, but it was, you know, late ‘90s, and I worked at an elevator company with my brother. Family business. 

Chris: ‘Kay.

Gina: In IT. We were IT support. And the office was two floors. On the bottom floor was the factory floor. So this is a company that designed and manufactured and tested elevator parts.

Chris: That’s super cool.

Gina: The upstairs, the second floor, was the office. And so, it’s late ‘90s, everyone’s got these PCs on their desks. It’s a very dirty and loud kind of… the factory in particular is a particular environment.

Chris: Sure.

Gina: So I spent my days, you know, setting up and configuring the server rack, and running ethernet cords…

Chris: Oh, yeah,

Gina: Like, crimping the ends of ethernet cords.

Chris: Yes!

Gina: Troubleshooting these PCs, these workstations. One of the memories that’s burned into my brain. We had a Novell server.

Chris: Okay.

Gina: And there was a shared folder that the entire company worked off of. So, it’s an elevator company, right? So there’s this one shared folder. And it’s got all these subfolders inside of it. Each folder is the address of the building, okay?

Chris: Oh, wow, okay.

Gina: For that particular project.

Chris: I mean, that makes sense. That’s a logical data model.

Gina: Yeah. Yeah. All the CAD drawings, all the configurations, the billings, the invoices, everything.

Chris: Sure. Yeah.

Gina: So, like, if you… you know, at any given moment, you’re working on a particular building, you go to that folder in this air drive and you access the files. And again, the factory floor was kinda dirty, and the equipment was old. And the mice, like the mouse at each workstation, had a physical trackball inside of it.

Chris: Oh, sure.

Gina: But you know what, the dust and just the dirt and things…

Chris: They get pretty grimy.

Gina: They get grimy. And the trackball would get stuck.

Chris: Mhm.

Gina: And sometimes even the physical buttons which you would click on the mice would get kinda stuck. 

Chris: Yep.

Gina: One of my colleagues, like early one morning, was mousing around on her desktop, and everybody’s desktop had a link to the shared drive.

Chris: Sure. Yeah.

Gina: Accidentally…

Chris: I’m already getting nervous.

Gina: Yeah, you’re getting nervous, right?

Chris: Just hearing you talk… (Laughs)

Gina: You’re getting nervous. 

(Both pause)

Chris: Keep going.

Gina: She accidentally… and it’s not her fault. ‘Cause her, literally the mouse got stuck.

Chris: Wouldn’t work, yeah.

Gina: Accidentally dragged and dropped the shared folder, like, into another folder. Which then cut off access, like no one… it was no longer shared.

Chris: To the rest of the company.

Gina: To the rest of the company. 

Chris: Oh my God.

Gina: So the shared drive goes away. It disappears. Literally, work stopped. Everyone’s like, “We can’t do anything.”

Chris: All the stuff is gone. Yeah.

Gina: “We can’t get to anything.” And me and my brother spent the day… now, you gotta imagine this. There are people, like, standing in the hallways, you know, it’s like a cubicle kind of situation, drinking their coffee, going like “Yeah, can’t get to the shared folder, can’t do anything.”

Chris: Yeah. (Laughs) God.

Gina: And we’re going from workstation to workstation, we’re back… we’re like, you know, “When is the last backup?” We’re trying to restore, like, at what point did we do a snapshot? It was painful, and there were several hours of total downtime where no one could get any work done.

Chris: Yeah.

Gina: And it’s burned into my brain because it was so painful, and then when we realized that it was because my colleague’s mouse… like, when we isolated the workstation where it happened and then tried to use that workstation, and saw that the mouse was in such bad shape, it was… it was unbelievable that this happened. But it burned into my brain, as an IT person, as a support person, part of your job… and it becomes a big part of your job when a catastrophe like this happens, is to reduce risk. 

Chris: Yes.

Gina: To put up as many guardrails as possible. To make what happened in that situation not possible to ever happen.

Chris: Right.

Gina: That shouldn’t have been able to… you know. That shouldn’t have happened, it shouldn’t have been able to happen, because you can… one person making an innocent mistake shouldn’t take the whole company offline.

Chris: Exactly. 

Gina: So I think about that a lot, now, 20 million years later.

Chris: (Laughs)

Gina: When I’m frustrated with an IT department…

Chris: Mhm.

Gina: Who doesn’t want to give you access to the thing, or is saying “Here are our rules.” I have to remember, these folks’ job is to reduce risk, to put up guardrails, to give you access…

Chris: To make sure the company can run.

Gina: Make sure the company can run.

Chris: Yep.

Gina: Reduce any risk that someone’s gonna, like, cut off access to that shared folder. (Laughs)

Chris: Yeah. No, no, no! I have… can we just, first of all, appreciate the power of the shared folder? 

Gina: Whoo.

Chris: We’re still working that way today, in a lot of ways.

Gina: Yeah.

Chris: It’s just that the shared folder exists in…

Gina: It’s in the cloud.

Chris: Google Drive.

Gina: That’s right.

Chris: Or Microsoft SharePoint or something like that.

Gina: That’s right, that’s right.

Chris: But, man. So many companies, I guarantee you, are like “Did you put that on the M drive?” Or whatever it’s called.

Gina: Yes! Yeah, that’s right! It was, like, the M drive! 

Chris: Yeah!

Gina: It was some letter very far up the alphabet, right?

Chris: Yeah. And that’s how the world ran.

Gina: That’s how the world ran, that’s right.

Chris: All… I remember early in my career, all of my clients had a similar version of that. I have a story that’s somewhat like yours that again, speaks to reducing risk.

Gina: Mhm.

Chris: I remember an early project that I was working on where one day, you know, the team had a shared development environment.

Gina: Mhm.

Chris: And one day the development environment went offline.

Gina: Mhm.

Chris: And nobody could, you know, do their work. It was a time and place where setting up a local environment was not really possible, right?

Gina: It’s not… yeah.

Chris: This was pre-Docker, pre-virtualized environments. 

Gina: Mhm.

Chris: And it was just, you know… in order to really see your work you had to do it in a shared environment.

Gina: You had to have access to the dev’s… right.

Chris: Right. Nobody had… for example, nobody had access to a local database with all the data, right?

Gina: Right, right.

Chris: They had a local database with, like, test data, but you couldn’t really see… and the server just went offline. And so we, you know, we were trying to debug, “What’s going on? Why is this unavailable? Is it a problem with our connections or whatever?” Turns out, our operations team was moving offices and they had unplugged…

Gina: They had unplugged it. Something got unplugged.

Chris: It was a computer tower that was just plugged in the wall and tucked into the corner of a room!

Gina: (Laughs) Right.

Chris: And it just got unplugged! And it was this moment where it was like, “Oh! Not only was this totally avoidable, we didn’t even know.”

Gina: “We…” Right.

Chris: We didn’t even… you know? 

Gina: (Laughs)

Chris: And it’s the same kinda thing. Good IT practices is just about reducing risk…

Gina: Yes.

Chris: Knowing where all your access points are.

Gina: Yes.

Chris: Knowing how you have a proper process and redundancy if necessary, so that things…

Gina: Yes.

Chris: And, you know, the world has changed a lot in the past 15 or 20 years…

Gina: Yes.

Chris: …but the ethos is the same.

Gina: Is the same. That’s right. Good reporting, good alerts when something’s offline, right? Like, knowing…

Chris: Yes.

Gina: Yeah, yeah. I mean, fast-forward to… I mean, Postlight was founded 7… we’re 7 years old? Is that right?

Chris: Mhm. Yeah.

Gina: And our entire… (Laughs) This was a fun part of our acquisition, they were like “What is your IT team, and what is it… what are your servers? Like…”

Chris: We just kinda raised our hands, like… (Laughs)

Gina: Right, we were like “The IT team is us…”

Chris: Head of engineering, yep.

Gina: And our IT equipment involves a router in our office. Like, we run the entire company on a Sass stack. Like, IT for us, today, is like… provisioning Google Workspace users, and getting, you know… managing our Airtable and Figma user base.

Chris: Right.

Gina: And our Slack permissions. Right? Like, that’s… it’s all Sass products.

Chris: We have… other than, literally the internet setup in the office, we have no physical hard… and everyone’s laptops.

Gina: That’s right.

Chris: Like, we have no physical hardware.

Gina: We have no physical hardware. And everyone’s laptops. That’s right.

Chris: We don’t even rent a, you know, hardware in a data center. We are all in on software as a service platforms.

Gina: That’s right. That’s right. And for our clients, we exclusively develop in the cloud as well, right?

Chris: Well…

Gina: We’re not configuring data centers. Is it not, has it not been exclusive?

Chris: I wouldn’t say exclusive. I mean, we have some clients who have their own on-premise hardware.

Gina: That’s true, yep. Okay.

Chris: Or run their own, you know, what I would call like mini data centers.

Gina: Mhm, mhm.

Chris: Well, I guess I do agree with you that from the engineer’s perspective, right? From a typical day-to-day development you’re working with “the cloud,” it’s just that…

Gina: “The cloud.”

Chris: …the cloud may be a private cloud.

Gina: Right.

Chris: That’s run by, you know, an internal team.

Gina: Right. Right, right, right.

Chris: But yeah, I mean, we have very much said “We are going to help move you, Client X, into the cloud, and into the future of modern development.”

Gina: Mhm. That’s right. And provisioning a database server is like, clicking some buttons in a console.

Chris: Exactly.

Gina: It’s not actually… (Laughs) Building a server.

Chris: Right. Or running a script!

Gina: Or running a script, even. Or running a script, that’s right.

Chris: Yeah. That’s another story that I remember from early on, is I remember one of our… someone from an IT team asked me, you know, “How did you set up X, Y or Z?” And I was so embarrassed, I was like “I have a text file with a bunch of commands that I ran in the terminal.”

Gina: (Laughs) “And I ran them. I copied and pasted them into the terminal.”

Chris: Yeah. And he was like, “Can you send me the text file?” And I was like “Okay.”

Gina: (Laughs)

Chris: And he was like, “This is perfect! This is amazing! Like, this is exactly what I needed to be able to replicate…” All it was was, like, an early version of a Docker file. Like, it just said “Here’s all the things you have to do to spin up this environment,” you know?

Gina: Good shell scripts…

Chris: Exactly, and we’ve now just automated it.

Gina: Beautiful thing.

Chris: It’s really interesting. When you think about what we do, right? We do modern custom platform development for clients, where we are often creating new products from scratch. Or at the very least, we are modernizing existing applications by bringing new pieces into existence. There is a very interesting, you know, connection point with the IT team, if you think about the IT team as managing the infrastructure and reducing the risk around the overall platform.

Gina: Right. I mean, you can make the argument that if IT’s… you know, one of IT’s main mandates is to reduce risk, you could say that one of the product team’s biggest mandates is to actually take risks.

Chris: Exactly.

Gina: Meaning, put software out there that’s maybe unfinished. Put out features that aren’t complete but… and you don’t even know if users are gonna take to them, you know?

Chris: I love this. Yes. Yeah.

Gina: Like, it’s a very different ethos. It’s the opposite. 

Chris: Right.

Gina: It’s like, “We’re gonna try some things. We’re gonna experiment. We’re gonna…” Like, you know, an MVP is a “minimum viable product.” Like, it’s not done and it’s not ready, and we don’t even know if this product is gonna get traction.

Chris: Right.

Gina: We don’t actually even know what finished looks like.

Chris: Exactly.

Gina: Right? Which is a huge difference between project work and product work. Right?

Chris: Right. So, okay. I love that you’re drawing this connection, like, this parallel, but how do you bridge the gap, then? That is the complete 180 from a mentality perspective…

Gina: Perspective. Mhm.

Chris: From how an IT group typically looks at things.

Gina: Right.

Chris: They say “I want a really clear set of requirements.”

Gina: Yeah.

Chris: “I want to know exactly what I need to do, I want to know how I can scale this platform and get ahead of things, I want to make sure it never goes down. I want to make sure that it’s very clear how to get help and support.” Like, it’s all about reducing all unknowns.

Gina: Right. I’m meeting every security and compliance regulation I need to because I’ve got legal on my back…

Chris: Right, exactly.

Gina: (Laughs) You know? Like, all those things.

Chris: It is a… you know, you said it beautifully. It is a very different…

Gina: Totally different mindset.

Chris: …mindset. About how you’re approaching the work.

Gina: Yeah. Here’s the thing: reducing risk is a mandate. Also, though, IT’s job is to help empower and enable people to do their jobs, right?

Chris: That’s right. Yep. 

Gina: And if a company’s becoming digital, or if a product team, you know, like, we’re going to build a new product, there has to be a give and take there.

Chris: Yes.

Gina: There has to be a conversation there. You know, we’ve kicked off projects with big clients that have IT teams that are, you know, very risk-averse and very clear about what’s allowed and not allowed, and we need to get started. I mean, velocity and agility are very important to us. And, you know, our approach is always, build a relationship with the IT team, talk through the architecture, if we’re going to use a cloud server, what we’re going to use. And in some cases, you know, we spin up those services on our own, separate from the IT organization…

Chris: Yep.

Gina: But we let them know, like, “This is happening, this is what we’re building on.” And then later… we start with an agreement that we are going to transfer ownership of this cloud service, of this Sass product, into your org at some point in the future, right? So you have to build that relationship and do that negotiation up front. And we’ve had some IT orgs say, like, “Mm, this service, it’s on our block list. We cannot support this. Don’t use it.” 

Chris: “Don’t use it.” Mhm.

Gina: “Find one that you can use.” You know? It’s important to have those conversations up front. I think it’s also important to – and we talked about this in our speedboat versus tanker episode, right? – like, to get started. And sometimes that means spinning stuff up outside of the environment, just to get started and get some velocity.

Chris: Exactly. 

Gina: But with the plan that we’re not, you know, we’re gonna integrate this into your environment. You are gonna take ownership of this, right?

Chris: Yes.

Gina: Like, we… the software that we build is our clients’ property.

Chris: Yes.

Gina: Right?

Chris: One of the ideas that we have really internalized as we’ve worked for our clients is building for the handoff.

Gina: Yes.

Chris: I remember we used to talk a lot about that in the beginning of the company, and it’s still very much a part of our DNA today.

Gina: Mhm.

Chris: An aspect of building for the handoff is knowing that if you’re ultimately gonna end up in this kind of environment, how do you make sure that you’re making the right choices…

Both: …now.

Chris: So that you’re… that handoff is as smooth as possible

Gina: Yep.

Chris: This is exactly what you’re talking about. Like, having that conversation… don’t wait until you’re 6 months in or a year in before you’re sitting down with the CIO, or the VP of IT…

Gina: Yeah, that’s… very bad idea.

Chris: Right? And saying “We’ve got all this stuff, you now need to support, blah de da de dah.”

Gina: Yeah, “This is now yours! Good luck, here you go, let us know where to transfer it.” 

Chris: Right. When we talk about design, are we talking about the frontend experience and having a great UX, you know, interaction design? Yes, absolutely. But we are also talking about software architecture.

Gina: Right.

Chris: That process happens at the very beginning. Especially when you’re breaking ground on a brand-new platform, a brand-new thing.

Gina: Yes.

Chris: Where you are inherently going to introduce new pieces of technology or new parts of the technology stack to an organization. And that software architecture phase? It has to be in collaboration with the team that’s ultimately going to support it. 

Gina: Yeah,

Chris: And we’ve done that! We’ve sat down with CIOs, and we say “We don’t know exactly what the full feature list is going to be. But we do know… here are a couple of high-level technology choices we want to make. How does that sound to you?” Right? We wanna host it on Amazon Web Services, and we want to use Postgres as our database, and we want to use this as our messaging queue and this as our frontend technology. Does this all feel right to you?” And a lot of the time, the answer is yes, because an early heads-up is like, “Okay great! We can be comfortable with this.” And if it’s no, then we know at the beginning, and we know that we can make a different choice.

Gina: Right. That’s right. Bad news shared early, just news. Right?

Chris: It’s just news.

Gina: We always say that. 

Chris: Yep.

Gina: There are times when we… you know, in the interest of trying something new and moving quickly… you know, I remember one client we had, they had this sort of big, monolithic platform, and we were adding on a piece to it, and we said “You know what? We just want to spin up this couple of Sass products to get there quickly and make sure that we have…” And the engineering team, I remember the engineering team being like “Well, but this isn’t part of our monolith. Why wouldn’t you work in our monolith? But also, in order for you to work in our monolith, we’re gonna have to… you know, we’re gonna have to do a bunch of training…”

Chris: Onboarding…

Gina: Onboarding and on-ramping, and I don’t know that we can get all the permissions for you to see the proprietary data… And so, you know, we went back and forth, and we were like “Look, it would be so much easier for all of us if you… you know, if we could spin up this parallel lane, work in this kinda sandbox, this new sandbox, on this new bit, you know, this new bit of the thing. And then, you know, later on, if this works, we can either integrate it back into the monolith, or are you willing to open your hearts to maintaining this new stack?” (Laughs)

Chris: Sure.

Gina: And it was hard. Like, it was a little… you know, especially, I think, for the engineering team, I think there was a little feeling of like, you know, it’s like “We’ve worked years and years to build this monolith.” It’s a little threatening. It feels like some sort of implicit judgment…

Chris: Of course.

Gina: That their existing platform wasn’t good enough for us to build on.

Chris: Right.

Gina: And it was just like, “No, no, no! In the interest of expediency, and in the interest of keeping us in this one particular lane and doing this one particular thing that’s kinda carved out, like…” And on the frontend it’s completely invisible, you know, users don’t see the difference. And they agreed. And part of this is relationships and trust-building and this understanding that “You are going to own this eventually, and we’re not going to make choices that you’re not okay with owning at some point…” 

Chris: That, just that sentence alone is huge.

Gina: Yeah.

Chris: Because I think it can be very easy for product development teams to come in and say “We know best. We’re gonna make the choices that we wanna make because we’re prioritizing what we need only.” Right? It’s a very… it’s a short-term thinking mindset. And it’s natural, to be clear. Like, it is natural to think “We need to make sure that we have full control because we wanna…” You know, run down the list. Move the fastest, have the greatest control over design, make sure that we know exactly what we’re using so that we can… you know. A lot of it is about moving fast. But especially when you’re standing something up that you know is gonna be a long-term part of this platform…

Gina: Mhm.

Chris: You have to think long-term from the beginning. You have to include those people whose job it’s ultimately gonna be…

Gina: Right.

Chris: …to maintain that platform.

Gina: Mhm.

Chris: And it’s so much better if you do it at the beginning.

Gina: Yeah.

Chris: Versus doing it at the end.

Gina: Yeah. For sure. Some of the skepticism and anxiety we’ve encountered in these situations is like, “Are you going to spin up this thing with this proprietary software and then is this going to make us need you forever?” Basically, like, “Are you fleecing us? Are you going to make us dependent on you forever?”

Chris: Mm. Right. “Are you trying to… are you trying to lock yourselves in?”

Gina: “Are you trying to lock yourselves in? Are we not going to be able to operate without you?” And so this is why we have emphasized this…

Both: Build for the handoff.

Gina: And, like, you own this. Our job is, you know, we really see our job as to work ourselves out of a job at some point.

Chris: Exactly.

Gina: If you need us we’ll be here, but we’re not going to set you up with something, you’re not licensing software from us, we’re not setting up something that is… we always use pragmatic, mainstream tools that are, you know, nothing obscure and crazy that only we know. You know? ‘Cause that just doesn’t make sense.

Chris: Right.

Gina: These have to be stacks that you can hire for, and that you can… that are well-documented and well-established. 

Chris: Exactly. If you’re making an obscure choice, there has to be a very compelling reason why you’re doing it.

Gina: Very compelling reason, yes. Yes.

Chris: And sometimes there are.

Gina: Sometimes there are.

Chris: Sometimes you have to make a niche decision because you’re like, the requirement, there’s very specific, you know, latency requirements or something, and you need to make sure that you are…

Gina: Right.

Chris: But most of the time, make the boring choice, because that’s gonna be…

Gina: …the one that’s most maintainable.

Chris: Exactly. We had someone say to us one time, when they were talking about the sort of divide between digital product development and IT teams, we had someone say, you know, “IT teams were built to configure exchange servers, not build product.”

Gina: Mhm.

Chris: And it’s a heck of a quote. I don’t know if it’s totally true, and I don’t know if it captures where we are today.

Gina: Mhm.

Chris: I think there’s some of that, for sure.

Gina: Yeah.

Chris: And I think you can certainly talk to IT teams who don’t want to be bothered. And that can be the source of a lot of friction.

Gina: Mhm.

Chris: But we’ve also talked to some very progressive…

Gina: IT orgs. 

Chris: CEOs and IT orgs, who are very much about enabling really great product development on their platforms.

Gina: Yes. Yes.

Chris: And if you think about it, going back to what you said at the beginning, if you think about their orientation, about reducing risk for their users, right? In the short term and in the long term. And you can build in a collaborative environment from the beginning, then you’re gonna get a much better outcome, right? Versus the “I don’t know what this is, and so I don’t wanna engage with it,” which can happen the further the paths diverge.

Gina: That’s right. That’s right. Our advice, our best advice, is to pull your IT department team close…

Chris: Into your project.

Gina: Into your world. 

Chris: Yeah.

Gina: What are we trying to do? Why are we trying to do it? How is this gonna move our business forward? Like, how… I mean, the most progressive CIOs and IT departments that we work with, they want to meet users where they are. They want to make sure the business is becoming digital.

Chris: Exactly.

Gina: They see the future and they see, they take their job of empowering their teams and employees to get their job done more seriously than just a “no” department. “Nope, can’t have that, nope, can’t have that.” And I think on our side, you know, in the product community, and the product teams have to meet their IT folks where they are as well.

Chris: Yep.

Gina: And not just say “Ugh, all these guys do is set up exchange servers and say no.” (Laughs)

Chris: Right.

Gina: Some empathy. You’ve gotta think about where they’re coming from, right? You gotta bring them close.

Chris: That’s right.

Gina: And say “Here’s why I need access to Figma,” or “Here’s why I need this,” you know, whatever it is. “Here’s why it’s so important that we bring up this new cloud platform.”

Chris: Exactly.

Gina: Because it’s gonna, this is gonna grow our business. Or at least, our theory is, this is gonna grow our business in leaps and bounds, and this is why it’s so important for all of us. 

Chris: Yes. The other thing that I’ve seen good IT groups really care about, that product development teams sometimes overlook, is cost. Cost of all these services, right?

Gina: Mm, yes.

Chris: And it’s very easy, especially in the world where everything is a service, right?

Gina: Yes.

Chris: It’s very easy to think, “Well, this is gonna enable us to move so much faster, and so we’re gonna save on developer energy, which is cost.” And that’s true! 

Gina: Mhm.

Chris: But also, there is an ongoing and often growing service cost and maintenance cost, that gets bolted on at the end, that product development teams are not thinking about.

Gina: They’re not thinking about. That’s right.

Chris: And there can be a really healthy give and take there, where you’ve got IT teams who are really thoughtful and thinking about how they can optimize for cost…

Gina: That’s right.

Chris: …that gets built into the product development effort.

Gina: For sure.

Chris: A prime example of this, I feel like, recent to Postlight, is the advent of serverless technology. Right?

Gina: Yes.

Chris: Building backend functions instead of entire, you know, servers that have to be running 24/7, that just get invoked on the fly, and reduce costs dramatically.

Gina: Yes.

Chris: And when you have an IT org that really embraces that and sees the power of that, there can be huge cost savings, because there’s a symbiosis, right? There’s a give and take that’s happening with the people maintaining the infrastructure, and the people that are building the software that’s running on it.

Gina: That’s right. I mean, we lived this. It’s very common for a product team to be like “Let’s try something out, let’s just stand up what we need to enable it and see if it works.” And then it works, and there’s incredible growth, and now suddenly you’ve got all these cloud instance… you know, databases, growing to gargantuan sizes, logging… you know, this happened at Postlight, right? With Postlight Reader. We started to see our AWS bill just get bigger and bigger and bigger. And then we’re like, “We’re spending, like, you know…” What was it? It was like, seven figures…

Chris: It was a lot of money.

Gina: …a month on, to run a free service. And then, you know, one of our more devops-minded folks said, “If we switch to serverless we’ll only pay for as much as we use,” and it required a rewrite. But at that point we had millions of users, and we knew we wanted the product to run, and so now we could optimize. And we reduced our bill. That’s the kind of thinking, you know, you need to have, right? ‘Cause if you’re not collaborating with your IT group, right? And these costs are just mounting over time, it will build resentment. And then eventually, the CIO or the CFO or somebody is gonna be like “Why is this costing so much?” And it’s like, “Well, these… this product team wasn’t thinking about our IT costs.” You know? It’s like, well, they would be if there was a collaboration between those two.

Chris: Bingo. When I think about, like, the rise of devops, and the idea that product development and operations need to actually be closer together, and in many cases be one thing, as opposed to “Development team does their work and then they hand it over to operations to run the thing,” it often gets interpreted as, like, development infrastructure, in terms of, like, continuous integration pipelines and automated testing. And those are great and really important, and we often use those approaches in our development. But it also is about what we’re talking about right now, which is like, infrastructure optimization. That has to be done hand in hand.

Gina: Mhm. That’s right.

Chris: You have to have operational, really good operational thinkers, and engineering working together in concert to really make smart choices for these platforms. And this is the thing: in our experience, there are some really good IT teams out there that care about this stuff. 

Gina: Yeah.

Chris: That want this to be really good. That do prioritize the ultimate end-user experience, right? Along with the overall cost and maintainability and things like that. 

Gina: Mhm.

Chris: And it can be really, really helpful, and really fruitful, to have these energies combined. Versus, you know, the sort of friction and throwing it over the wall…

Gina: Right, “IT always says no, the product team costs us so much money…” Yeah. Right, exactly, exactly. 

Chris: Yeah.

Gina: It also just encourages good modular platform architecture, right?

Chris: Totally.

Gina: Like, “Do this on RDS. Start with a small instance at this growth rate. Like, what size is the database? Okay, now we’re in a position where we can increase the size of the database with, like, the click of a button, and now… and it makes just the whole experience for everybody so much better.” 

Chris: Yes.

Gina: You know, I think that the IT versus digital divide is real. But it’s kind of, it’s all of our jobs to bridge that gap.

Chris: Yes.

Gina: I think this is the main takeaway that we wanna make, right? That it’s so important for both product and IT – and I think members of both of those communities listen to this show – to meet in the middle and work together, and be empathetic to each one…

Chris: To each side.

Gina: To each other’s plights and mandates, right? That reducing risk versus taking risk, but also making… making progress and moving forward. If you are struggling with your product team, who keeps asking for stuff that you can’t give them, or your IT team who’s constantly telling you no and isn’t thinking like a product team, please get in touch. We’d love to talk to you about this. Send us a note at hello@postlight.com. This is a core struggle that so many businesses, particularly technology businesses that have been around for more than 10-15 years, are going through. We see this at all of our clients, so we love talking about this stuff. And we think it’s just so important for our industry. So get in touch, hello@postlight.com. Thanks, as always, for listening.

Chris: We hope to hear from you soon.

Gina: Yeah. Bye, everybody.

Chris: Bye.

[POSTLIGHT OUTRO MUSIC]