CRAFTED. | The Tech Podcast for Founders, Makers, and Innovators

Making media is hard. Distributing it to the right audience even more so. GenAI can help. 

Matt Monahan is the president of Arc XP at the Washington Post. On this episode, we're digging into the core technology of any media company: the content management system (CMS), why they’re so hard to build right and how new GenAI tools can reduce the toil required for journalists to get their stories to the right audience. 

Arc XP is not only the CMS used by the Washington Post, but it’s also used by many other publishers, broadcasters, and companies with stories to tell. Matt shares more on the challenge of commercializing and white-listing an internal tool for others to use. It’s a tantalizing idea – turn that cost center into a profit center! — but it’s not for the faint of heart. 

Plus, lessons from Jeff Bezos, the owner of the Washington Post, including how the Post uses key Amazon practices, such as the “six page memo” and “one- and two-way doors.” 

Key Moments: 
  • (00:00) - Introduction
  • (02:25) - From print to digital: hard product & organizational changes
  • (04:17) - Why the CMS is where it all comes together
  • (06:27) - Commercializing an internal tool: selling Arc XP to other publishers
  • (10:07) - AWS and the impact of Jeff Bezos on the Washington Post
  • (11:30) - The famous six-page memo and how the Post benefits from it
  • (14:06) - GenAI: New features Matt is building to reduce toil for journalists
  • (20:45) - How to pick the right model and fine tune it
  • (23:24) - Matt’s a pilot! What flying has taught him about running a business
  • (24:33) - Being nimble: Why there are more “one way doors” than you think
  • (26:07) - Outro

CRAFTED. is brought to you in partnership with Docker, which helps developers build, share, run and verify applications anywhere – without environment confirmation or management. More than 20 million developers worldwide use Docker's suite of development tools, services and automations to accelerate the delivery of secure applications. 

CRAFTED. is produced by Modern Product Minds, where CRAFTED. host Dan Blumberg and team can help you take a new product from zero to one... and beyond. We specialize in early stage product discovery, growth, and experimentation. Learn more at modernproductminds.com 
Subscribe to CRAFTED., follow the show, and sign up for the newsletter 👉 crafted.fm

What is CRAFTED. | The Tech Podcast for Founders, Makers, and Innovators?

CRAFTED. is a show about great products and the people who make them. Top technologists reveal how they build game-changing products — and how you can, too. Honored twice by The Webby Awards as a top tech podcast, CRAFTED. is hosted by Dan Blumberg, an entrepreneur, product leader, and former public radio host. Listen to CRAFTED. to find out what it really takes to build great products and companies.

[00:00:00] Matt Monahan: As a pilot, if your original plan takes you straight through, is the thunderstorm, like you don't wanna keep following the plan? I think in the business world, sometimes those challenges are less obvious.
[00:00:09] Dan Blumberg: That's Matt Monahan, the president of ARC XP at the Washington Post, and he's both a product person and a pilot, so he knows a thing or two about turbulence, both in the air and in the media industry.
[00:00:20] And the rise of AI is only making things more unpredictable.
[00:00:24] Matt Monahan: Like as a product person, it's an amazing problem to be able to face right now. Like, what a cool disruptive challenge. You know, the human's still in the loop, but what is it exactly that you're doing? You're not necessarily flipping every switch and every dial.
[00:00:37] Dan Blumberg: On this episode of CRAFTED., we're talking about the core technology of any media company.
[00:00:41] Matt Monahan: And CMS is right in the middle of that. It's easy to get it wrong. It's very difficult to get it right, how AI can automate some content production and promotion. So it's super important, but not the thing that most journalists wanna do. Most users wanna do. It's the perfect place to plug AI into.
[00:00:56] Dan Blumberg: Plus, ARC XP is not only used by the Washington Post, but also sold to other companies. We'll talk about what it takes to turn an internal tool into a business, and we'll also hear what Matt has learned from Jeff Bezos who owns The Washington Post.
[00:01:09] We'll hear how the Post has implemented those famous six page memos and how they discuss one and two-way doors.
[00:01:16] Matt Monahan: So many of these Jeff observations that are superficially obvious. And yet so few companies actually execute in a consistent way in it and practice.
[00:01:28] Dan Blumberg: Welcome to CRAFTED., a show about great products and the people who make them. I'm Dan Blumberg. I'm a product and growth leader, and on crafted, I'm here to bring you stories of founders, makers, and innovators that reveal how they build game changing products and how you can too.
[00:01:44] CRAFTED. is brought to you in partnership with Docker, which helps developers build, share, run, and verify applications anywhere without environment confirmation or management. More than 20 million developers worldwide use Docker's suite of development tools, services, and automations to accelerate the delivery of secure applications. Learn more at Docker.com
[00:02:05] And CRAFTED. is produced by Modern Product Minds where my team and I can help you take a new product from zero to one and beyond. We specialize in early stage product discovery, growth, and experimentation. Learn more and sign up for the CRAFTED. newsletter at modernproductminds.com.
[00:02:25] Matt Monahan: It was kind of such an unbelievable time because it was probably 2013 or so, and we recognized that, you know, things would need to change pretty dramatically to get on the right digital footing for the future. And so I think in those early days, I'm really proud of the work that we did because. There wasn't a lot to draw from at that time.
[00:02:46] You know, I mean, years later you could go to these conferences and everybody's kind of talking about the same thing and people are talking about the technology they're building for their consumers, for inside their newsrooms, but then it was all new and so we really had to kind of chart a new path.
[00:02:58] Dan Blumberg: Yeah.
[00:02:59] Matt Monahan: And look a little bit into the future what we thought would be needed. We definitely didn't find anything on the market available and that's why we started building our own CMS.
[00:03:05] Dan Blumberg: Yeah, I mean at at that time I was at the New York Times on the product team from 2012 to 2016. Mobile phones were still really new.
[00:03:10] At that point, the iPad had only just come out and the cms and everyone loves CMSs, right? No one ever complains about the cms. The CMS wasn't built for those things. And so I'm curious how that manifested in, in your experience at the Washington Post.
[00:03:25] Matt Monahan: It's surprising how much it's changed, right? Because exactly. Like you say, the CMSs that were commercially available back then were either too small for large media organizations, so of course, WordPress and Drupal both existed at that time. They were pretty different than they are today, but also they weren't really well suited for large media organizations. And on the other end of the spectrum, you had these big legacy commercial solutions that were mature, but.
[00:03:49] They were really focused on print and to the extent they were focused on digital, you know, it was like a desktop website and everybody's coming to the realization at the same time because they're all beginning as consumers to use phones for everything. You know, at the beginning of this, that was what all the conversation was about, was how do we build something that can kind of address that change in the needs of these big media organizations, but also plan for a future that we don't really know what it's gonna look like. Like even then you sort of recognize if it's changing this fast now. What's gonna come next?
[00:04:17] Dan Blumberg: Yeah. For, for those that have never worked in a newsroom, can you explain the, the importance of the CMS or the interface that people are using to write their stories, manage their stories?
[00:04:26] Matt Monahan: Yeah. So CMS is a content management system.
[00:04:29] It's the thing where the main thing that you produce gets input, gets shared with your audience, gets distributed. So it's really at the center of what media companies do if you're in a big media organization. That's where all the complexity of your organization comes together too. So, you know, in an organization like the Washington Post, you've got hundreds of journalists approaching a thousand these days, more photographers, videographers, um, producers, people who are putting together different experiences for consumers, like parts of the, the website, but also things like email newsletters and other off channel experiences on social.
[00:05:05] And the CMS is right in the middle of that. It's easy to get it wrong. It's very difficult to get it right, especially in the view of the folks that are using it every day.
[00:05:11] Dan Blumberg: Yeah. It can be a very thankless job to work on the CMS because it's, you know, it's hard.
[00:05:16] Matt Monahan: I remember one of the very first products that we built that ultimately became part of ARC was a tool called Web Get and, and the whole purpose of it was to shift the planning workflow and also just the, the behavior inside the newsroom to a digital posture and one that reflected kind of the needs of the audience on our site, on our apps, and on our social channels. But what that meant was you had to think about when are consumers on the site? When are we actually publishing content? And it's. You know, if you haven't worked in one of these journalism companies, it's a bigger change than you think.
[00:05:44] Like everybody gets used to that idea of, okay, I have my deadline at the end of the day, the deadline's associated with when the last bit of content can go out to the printing press. And then the next day I can kind of start at a leisurely pace because anyways, my deadline that day is again at the end of the day.
[00:05:59] And so, you know, at the time our editor was Marty Baron, who folks would be familiar with from, you know, movies like Spotlight and some of the amazing reporting work he did at Miami Herald and Boston Globe, and then eventually at the Post, but. At the time, he was really closely involved with the work that we were doing and, and directing it at a minute level because he wanted to see that entire workflow change and see the pace of the building change.
[00:06:21] And, um, and it was the collaboration with him actually that led us to being successful and changing the way that the company at that time worked. Got it.
[00:06:27] Dan Blumberg: Tell me about the challenges of, of whitelisting something that is used internally for new customers. So ARC XP is born out of the Washington Post, internal CMS and now, now it is a business offering for other publishers, and it's such a tantalizing idea to take this internal tool and then and open it up to others. I remember the idea came up a lot when it was the New York Times, so I'm curious, you know, what was that process like? What are some of the things that surprised you as you embarked on this plan to take the, the CMS that you built up internally and and offer to others?
[00:06:59] Matt Monahan: There was always a lot of excitement about it, and I think. One of the things that we did, right, even from the early days, we sort of saw that it would be a possibility, and I think within the first year of starting to build even the beginnings of the system. And the reality is it took many years for the whole system to kind of come together.
[00:07:14] And so we architected it well, came up with a multi-tenant approach, but I think as a product person, what you realize pretty quickly is even customers that look very similar to you in that entrepreneurship story, actually, they're kind of different. Their needs are different, their requirements are different, and sometimes even when the business requirements look the same.
[00:07:31] When you're building a product, you start thinking, okay, what's the best way to interpret those business requirements and turn them into an amazing product? And so the contours of how their business works, what kind of journalism do they do? What's their audience like? What's the nature of their business?
[00:07:43] You know, we're a newspaper with one big website. Some of these companies we work with, you know, they don't have a print product, or they have many websites, or they run broadcast channels or audio or video is really important to them. And so the focus really shifts then in the requirements inside these tools.
[00:07:58] And so. I think stuff that superficially looks really easy and we said, Hey, we got great product market fit. That was always true from a marketing perspective, but on the product side it required a lot of work. Then, yeah, to build, you know, a configurable SaaS platform versus. Simply commercialize a tool that we built for one organization.
[00:08:16] Dan Blumberg: Can can you give an example of how that manifested?
[00:08:18] Matt Monahan: Yeah. I mean, one of the first really hard examples of it was the first kind of large multi-site organization we worked with. So if you think about the trend over the last 10 years, there's been a ton of consolidation in the industry. It's no secret that the, the commercial side of these businesses is a real challenge, has been disrupted by the internet.
[00:08:38] And consolidation's been one way the industry's responded to that. When you think about what that means in practice, what it, you know, what they're trying to do is do more with less. They're trying to centralize a lot of operations that otherwise are duplicative across all these different sites. And so, yeah, you know, we had one big customer that we signed, I think a 2015, and they had something like a dozen different sites, you know, big name titles.
[00:08:58] People knew all across the country, otherwise on paper they looked very similar to us. But of course their motive was we want to be able to rationalize this. We wanna be able to have one national content team that sits across all these different titles and they should be able to think about how to distribute content out to them, pull content up from the individual titles and redistribute it out to the other titles.
[00:09:18] And behind that is a lot of hard product work. It changes the nature of the content platform. It changes things like permissioning and roles. You have to build a much more robust structure for things like site management and CDN delivery. And you know, none of those things were necessary. Here inside the Washington Post.
[00:09:33] And so that was probably the first hard lesson we learned about that. You know, you start to get a picture of the difference between building a SaaS platform and a tool built for your internal company.
[00:09:44] Dan Blumberg: Yeah. Yeah. And I, and I should mention that, you know, one, one of the most famous examples of white labeling an internal product is AWS Amazon Web Services, which originally was infrastructure platform built by and for Amazon, and now it's used by.
[00:09:57] Everyone. Uh, and the Washington Post is owned by Jeff Bezos and Ark is built on AWS How did Jeff Bezos get involved as he came to own the Washington Post and and what have you learned from him and the growth of AWS?
[00:10:07] Matt Monahan: Yeah, so the example of AWS and Amazon came up a lot in the early days, and it still comes up from time to time.
[00:10:14] The nature of AWS it's fundamental infrastructure, right? It's, it's like infrastructure as a service or platform as a service, depending on the part that you're talking about. And I do think that makes it a little bit easier to abstract away from, you know, in that case an e-commerce business like Amazon.
[00:10:30] I think the, the answer of how Jeff was involved, many of the practices that have been otherwise reported from Amazon, um, were adopted here too. Things like the six page memo, writing, you know, instead of presentations and a real focus on the customer that's always been there. And so in the meetings where we talk about arc, that piece of it was kinda less hard than you'd almost imagine.
[00:10:52] He was excited about the opportunity and, and would ask like, what do these customers care about? What's best for them? And we'd have to present pretty nuanced ideas about how to solve this distinction between the internal customer and the external customers. And I remember some of these product decisions coming to a head.
[00:11:07] He kind of said to us something like, well, you can have two number one customers. And I think, you know, it was always sort of emblematic of his thinking that in those meetings, you know, he'd say something that sort of both seemed true and then at the same time was surprising. It made you go back and, and rethink how you were solving your problem.
[00:11:24] So it was an amazing environment to do something like this, but there's definitely a lot of hard work behind that if you wanna do it successfully.
[00:11:30] Dan Blumberg: Yeah. You mentioned a second ago the, the, the famous six page memo from Amazon. I'd love to, if you could unpack that a little bit for those who are unfamiliar with it and, and how you use it and how it plays into the prioritization that you're doing at the Washington Post at at ARC xp.
[00:11:43] Matt Monahan: Yeah. I mean, the six page memo is just a way to explain that you've got a pretty big decision that you need to make, and we use them not just sort of to present ideas to Jeff, but also internally. It's a great escape from the corporate PowerPoint or some sort of unstructured meeting. What it is, is it's a memo that lays out your ideas.
[00:12:04] You're thinking. In our case, sometimes they're updates, but most commonly there's some big decision in front of us, and in the end we usually make some recommendation or some ask. And so. Um, you kind of reserve it for the stuff where you can't solve it easily. You know, we're not writing a memo for very easy product decisions or business decisions.
[00:12:21] But, you know, examples could be we want to go after a new market that we haven't gone after previously, and we can sort of see what the investment required is, what the challenges will be, what the potential risk for existing business will be. And it's the kind of idea that's not well captured in a deck, but too often businesses would stick something like that in a deck.
[00:12:38] Yep.
[00:12:38] Dan Blumberg: And it's written in prose, right? It's not, not a bunch of bullet points.
[00:12:41] Matt Monahan: No. And in fact, the bullet points are heavily discouraged because they're, you know, they're kind of a way to cheat in the same way that a deck is a way to cheat. Totally. You have to write a compelling narrative. Um, the narrative should be supported by data and it's, it's a fun idea.
[00:12:53] 'cause it didn't come from media company. It, you know, it came from from Amazon and it's obviously came from other places before that. Yeah. But it changes people's way of thinking. And usually what you do is you send it earlier in the week so that people have time to consume. And the idea is that instead of sitting in the meeting and having sort of an off the cuff reaction to what actually is a pretty big business decision, yeah.
[00:13:11] You know, we're gonna invest millions of dollars in this new idea. It deserves a little bit of room to breathe and so you give it room to breathe. In terms of writing it, often the crafting these memos takes a week, sometimes more going back and forth between different people on the team, thinking about different perspectives.
[00:13:27] That part of the experience I. I, I haven't ever had one of those memos that, at the end looks very much like what we started with. You know, the, the process of creating it just forces a ton of good argument that, yeah, you realize some of the initial ideas you have aren't that sharp to begin with, and finally what you wind up with, you know, it gets distributed, people consume it.
[00:13:46] And what it means is going into those meetings now, everybody's informed. You're not talking about the basics. You're not trying to jam some content into a deck that influences people and seeing the set of facts in a very specific way.
[00:13:57] Dan Blumberg: Yeah.
[00:13:57] Matt Monahan: And the conversation really focuses on the meat of the decision and what the real trade offs are.
[00:14:01] And is it a one-way door? Is it a two-way door? And that's how we make big decisions here.
[00:14:06] Dan Blumberg: Yeah. Yeah. And there and there. There's no dog and pony show with the presentation. Yeah. I'd love to understand a little more of how you build. Um, can you tell us about a recent major feature launch, how you identified the opportunity, how you built the first version, iterated from there?
[00:14:21] Matt Monahan: Yeah. You know, like most people in this industry and others right now, there's a ton of focus on generative ai. I think in media, it's especially true that. Generative AI is really close to the challenge and the business that we're doing. It's an obviously disruptive technology, so it's applicable in a lot of different industries, but like, look, we're in the business of creating content and especially written content, so the applicability of generative ai, like, yeah, we're in a hype cycle, but no, it's not that hard to sort of understand that it's especially gonna be disruptive in media.
[00:14:53] We started looking at what are our customers doing? How can we help them with this technology? And I think we were pretty early to the scene with some solutions to that. So in the media industry right now, there's probably a few different categories that people are looking at about how generative AI can help solve problems for them.
[00:15:11] The easiest one is sort of automation of existing processes. Those are things like production of content, and I don't mean, you know, literally writing the entire story. I mean things like writing headlines, writing summaries, writing, SEO, keywords, um, tags, you know, things like that, that maybe more traditionally would've fallen into the job of a producer or in a smaller company, you know, the journalist themselves or an editor would've had to have done it.
[00:15:34] It's a lot of extra work, so that's an easy problem for us to solve. We started building stuff for that right away. Today we have a number of customers using those products out in production and. The stuff that's more interesting is when you start getting into content that becomes visible to your readers and your consumers.
[00:15:50] And maybe that's where we can focus on the product development process. Because for us, you know, we have customers that are users, journalists, editors sitting inside the tool. We also have their consumers and how they think about them. And in the end, our technology kind of touches both layers of that. So we have to be conscious of how our customers perceive that.
[00:16:08] And of course we know from our experience of the post, you know exactly the sensitivities around it. So we've been working on some things like audio articles, summaries, things that still are human in the loop. We tend to do beta programs for a lot of our big product development processes. And what we try to do, because we're in B2B, one of the fun things is you're not just relying on analytics to judge the success of these products.
[00:16:29] You can have product managers that go visit the customers on site, sit with their journalists in their newsrooms, understand what's going on, get the direct feedback, and you can use that qualitative data next to. The quantitative data that you obviously still have to have to judge the, you know, the success of those products.
[00:16:44] We're just nearing the end of that beta program right now for a few of those kind of product features and about to enter production in, in Q3, where it'll go out to a number of them.
[00:16:53] Dan Blumberg: Can you share a little bit about what those are or what you're learning in this beta process?
[00:16:57] Matt Monahan: Yeah, so, um, one of the things that we've built is a, um. a system for sort of abstracting a way of fine tuning and then having that power, things like article summaries and other sort of metadata surrounding a piece of content. I. And, um, you know, for a lot of our customers that fine tuning step is actually kinda hard. Some of them are hiring consultants to help them do it.
[00:17:20] There's obviously solutions out there, but it requires putting a bunch of pieces of technology together, some knowledge of the problem space. And I think we built a really great platform, API that kind of abstracts that away.
[00:17:30] Dan Blumberg: I'm trying to picture the product, is that there's, there's a, there's the, the original piece of content and then there's.
[00:17:35] You know, the 150 character summary, the 100 word summary, the, the version for Facebook, is that, is that a good way to think of it?
[00:17:44] Matt Monahan: Yeah, exactly. You got it. Exactly right. So there's things like social posts, there's things like summaries, which is a total chore,
[00:17:48] Dan Blumberg: and I'm doing it for the podcast and like, it is not my favorite part of the job.
[00:17:51] Like I'd rather spend more time prepping for this very interview than writing show notes for the, you know, the, the five that are in the backlog that need to come out next. So I, I get the pain.
[00:18:00] Matt Monahan: But then if you have a producer, they're gonna tell you that, Hey, that's the only way anyone's gonna find this podcast.
[00:18:04] So it's super important, but not the thing that most journalists wanna do, most users wanna do. It's the perfect place to plug AI into. 'cause you get away from this contention about is it really the core part of the content that I'm creating? How deeply should the human be in the loop? And that's what we're trying to solve for.
[00:18:18] Having said that, you've got rooms full of people that have traditionally done this job and they have really strong opinions about what that content should look like. What's the tone of their organization? Yeah. Is it accurate? And so even though it's still human in the loop, it only matters if it actually saves those humans time.
[00:18:33] Yep. That's where fine tuning is so important. Yeah.
[00:18:36] Dan Blumberg: Is there another lesson with regard to fine tuning or the other generative AI explorations that you're doing that you think might apply to other builders out there? Whether they're in, in media or otherwise, is they use these models.
[00:18:47] Matt Monahan: I mean, there's definitely differences between the models.
[00:18:49] Some are more well suited to certain types of prompts and certain types of workloads than others, and I think a lot of times, you know, these products are still being built primarily by engineers. That's a tough thing to communicate between the product requirements, the business users and the engineers.
[00:19:03] You know, you might wind up just trying to choose a public model based on cost, based on model size. Mm-hmm. Based on this is the type of workload, here's how they charge for tokens. This is the most ideal model. Yeah. The reality is the contours of those models, I think, actually matters to producing a really great product on top of them.
[00:19:16] Dan Blumberg: And especially if you're, you're saying this model's cheaper, but the tone is wrong versus this, and that's a hard thing to quantify. I'm sure it is.
[00:19:23] Matt Monahan: You need some deep familiarity with it. And then starts to get into the next question, which is, let's assume we get through this low hanging fruit. There was a survey, I think it was Reuters who, you know, they do a, um, annual survey of media executives, media companies, and lots of attention right now on things like backend automation with the human in the loop stuff, this production stuff that we've been talking about, but.
[00:19:43] You know, it's obvious that at some point everybody's gonna be thinking more deeply about. And what about the content itself? What does it mean for the nature of our product? A lot of trepidation about it very reasonably, especially within, not just media companies at large, but journalism companies where the accuracy of the content, the trust in the brand, is the most important thing that they do.
[00:20:03] And yet, at the same time, it's, it's sort of inescapable that. What's the real value that we're adding? Is it the original reporting, the facts that we can kind of uncover and produce? Or is it the literal production of the content? And I have this, um, example I sometimes give at, at conferences. I'm outside of work.
[00:20:20] I'm a pilot for fun, and so I, I love airplanes. You talk to any pilot, they'll go on and on about it. But if you look at a traditional flight deck, you see sort of all these. Dials and knobs and buttons. You know, like imagine the old flight tech of the Concord or something like that. They've got a flight engineer station.
[00:20:35] And I, I think of traditional CMSs and sort of media company technology kind of like that, where journalists want to have all the dials and all the bells and whistles that they can control themselves, you know, and if you look at a modern flight deck or where it's all going, Honeywell um, is producing a flight tech right now.
[00:20:51] That's. Partly AI powered. It's heavily automated. It looks like a handful of touchscreens, like a bunch of iPads, and it's hard to sort of imagine that that's not where some of this technology for media companies goes. Where. You know the human's still in the loop, but what is it exactly that you're doing?
[00:21:07] You're not necessarily flipping every switch and every dial, but you're sort of thinking about the intent of the content you're creating and definitely the original reporting you're doing and the original sort of idea that's inspiring that content. Once you start kind of opening your mind to that, you can imagine a world where.
[00:21:23] We could be much more heavily personalizing the experience for users. You know, not just personalized recommendations, but really personalized content. You could be anticipating the needs that consumers have to learn better and understand better. Yeah. And surfacing information inside these archives that these companies all have in much more meaningful ways.
[00:21:39] It's an amazing, like as a product person, it's an amazing problem to be able to face right now. Like what a cool disruptive challenge. Um. Obviously a lot of very reasonable trepidation and anxiety and caution, and that's probably right at the heart of the product problem that we're gonna have to solve.
[00:21:55] Dan Blumberg: Yeah, I mean, trust, that's the big one. I, I wanna go back to, I didn't know you were a pilot and I'm, I'm a sucker for the thing I learned over here that theoretically has nothing to do with the thing I do over there. Like I, so what have you learned from flying planes that applies to how you build products?
[00:22:09] Matt Monahan: Probably like a duality of having a plan and then being ready to deviate from the plan. And I think that's totally applicable. You know, as a product leader and as a business leader. Um, you can't execute without a strategic plan, obviously, and you can't execute without a product roadmap. But it's so common to see product leaders and business leaders get really fixated on their strategy and not be ready to pivot fast on it as a pilot.
[00:22:31] Like, you know, if your original plan takes you straight through, is the thunderstorm, like you don't want to keep following the plan? Yeah. I think in the business world, sometimes those challenges are less obvious. You know, I mean there's analogs, like you see it coming from far away. You think I'll have time to get outta the way of this and, and then, you know, by the time you sort of realize it's too late.
[00:22:46] It's too late. But there's also a lot of decisions, you know, as a product leader that seem less obvious than that. I. You still have customers giving you positive feedback, you still have customers signing on, and yet you sort of sense we're gonna have to change the direction, we have to alter the strategy a little bit.
[00:23:00] And I think about that idea a lot. So it's, it's a good question.
[00:23:04] Dan Blumberg: Yeah. How, how does this relate to the, the one-way door and two-way door
[00:23:06] Matt Monahan: decisions that you referenced earlier? Yeah. So for people who haven't heard that one before, it's exactly what it sounds like. You know, some doors are one way doors, you go through it, it's like the thing at the hospital where someone has to buzz you back in.
[00:23:17] Um, and some are two-way doors. You can walk through door, you can walk through it again and. I think the observation it, it's like so many of these Jeff observations that are superficially obvious and yet like so few companies actually execute in a consistent way in it in practice. And the one way door or two-way door thing is a lot of teams will spend tons of time arguing and arguing and arguing about something that you can reverse the decision later.
[00:23:41] Right. There's tons of product decisions. Software. It's software. Exactly. And, and some of the software is actually, you can change it cheaply. Sometimes it's a two-way door, but it's kind of expensive to go back through the door again. But it's reversible, you know? And then sometimes there's decisions where it's truly not reversible.
[00:23:55] If you build an entirely new version of your platform, you migrate everybody over. It's, you know, at least it's sort of like a one-way door. And sometimes you make business decisions to change your markets. Enter a new market that, um, depending on how intensely you do it, it feels more like a one-way door.
[00:24:09] And, yeah. But there's way fewer one-way doors than people sort of imagine. There's tons of two-way doors, and so the, the idea behind all of it is. Have a little more guts, you know, design leader MVPs. Be more ready to engage in that kind of experimentation. Rather than argue about it, use real data as opposed to kind of hypothetical data.
[00:24:28] And if you don't have the product and engineering team that can deliver at the pace to, to make that practical figure out how you can build that organization. Yeah. I love that, Matt,
[00:24:38] Dan Blumberg: thank you so much. This is really fun.
[00:24:40] Matt Monahan: Yeah, no thank you. This was a ton of fun.
[00:24:44] Dan Blumberg: That's Matt Monahan. I'm Dan Blumberg, and this is CRAFTED..
[00:24:48] CRAFTED. is brought to you in partnership with Docker, which helps developers build, share, run, and verify applications anywhere without environment confirmation or management. More than 20 million developers worldwide use Docker suite of development, tools, services, and automations to accelerate the delivery of secure applications. Learn more at Docker.com.
[00:25:12] Special thanks to Artium where I launched CRAFTED. Artium is a next generation software development consultancy that combines elite human craftsmanship and artificial intelligence. See how Artium can help you build your future at Artium.ai
[00:25:26] And CRAFTED. is produced by Modern Product Minds where my team and I can help you take a new product from zero to one and beyond. We specialize in early stage product discovery, growth, and experimentation. Learn more and sign up for the CRAFTED. newsletter at modernproductminds.com.
[00:25:45] Please share, CRAFTED. with a friend. Rate the show, review it, you know the drill.
[00:25:50] Matt Monahan: And so the, the idea behind all of it is have a little more guts.