The Future Is Here. Now What? | Joe Futty

speaker-0: Without further ado, let's give a warm welcome to Joe Fuci. Good morning product people. The futures here. We didn't invite it, it just showed up. So I'm gonna spend the next twenty five minutes or so talking about how I approach this. a little bit about me. I've in a handful of countries, born in US and then kinda gone back and forth in Europe. But I live in Portugal now, so it's my seventh country since last October, so I'm super excited to actually speak locally. Like many of you here, I worn many hats tech, small companies, large companies. I started my career as an engineer, and then I moved to ⁓ around 25, 26 years ago. I got invited learn about AI around years ago when I was working on search, search engines. We didn't call it AI back then, we just called it machine learning. So at Microsoft, I transitioned from product, sorry, from engineering to product. And ever since then, I've been doing product until most recently. I worked at eBay for about six years, where I led product for fulfillment, shipping, and loyalty. I worked at Booking, where I led UX All Up, Product for Marketplaces, Loyalty Program, and owned the product vision for the whole company. And then most recently, eight months ago, I moved over to Pipe Drive, where the pendulum swung back for me, going from engineer to product. Now I lead both. So it's super exciting to get a first hand seat on what's happening across the S DLC lifecycle. I always feel like my career has been preparing me all these years for what's about to happen now. Like you heard, I had a lot of patents at Microsoft. We were always talking about the future and how will that work out? speaker-1: Thank speaker-0: So a little bit about pipe drive. very cool story. We're on our sweet 16th birthday celebrating this year. It was five Estonians started this in a garage in Tallinn, which actually happens to be a hotbed startups. And the they started Pipe Drive is because they were looking for a sales system, like a CRM, and there was nothing out there that they liked. So they pivoted their company and built Pipe Drive. Ever since then, We've had this ethos of being easy to use, to learn, and we now have about a hundred thousand companies loving our product. So I'm super excited to be the custodian of this thing, pipe drive, as we move it forward and try to figure out how do we take that same ethos that we built and move into this world of AI and agentic, where sometimes it can be a little bit less human. We have offices all around the world. Our second biggest one here is in Lisbon, which is where I'm based out of. So let's get into the meat of the presentation. I don't know how many of you have seen this quote before. It's actually I keep coming back to this all the time because it'd be so important for product people or anybody that's building. So the futures here, it's not just it's just not evenly distributed. Kinda sounds interesting. But if we think about it, there are people that know how to do magic now. It's almost like the first person that in found fire. Anybody else that has never known fire, they look like magic. So sit with us a little bit. So the capabilities already exist that are, wow, magic, but it's not evenly distributed. What does that mean? So like a good engineer, I like to have a graph about it. And maybe this is correct, maybe it's not correct. I'm not sure where you are on this. But it just shows you that the scheme of the whole world, it's only a few of us that have actually tasted, or the words I use, activated the future. And it's really important because you are talking to your customers. Some your customers are fully activated, some are not. I'm not sure what business you're on. We're in the SMB business, so it actually varies quite a lot. There are some people that are ahead, and there's many, many on the backhand. So this is could be a problem, but for you, this is actually the opportunity. This is where product people shine. When you try to take a complex problem, I really understand what's happening. So it's not only your customers, you can actually think about those products as well. Some of the products are fully activated, some are just trying things for the first time. And probably the one that I'm focusing most of my time these days is around the team. Parts of the team could be activated, some teams are activated, which also makes it Kind of complicated to deal with all this. Also, the one thing that I've been around for a while, I've through a lot of these patterns, nothing as big as this. But for tech shift, I think there's kind of a pattern, especially when it comes to the work. First, we make technology fit how we already work. So I can go back to the time when you first seen ChatGPT, maybe the first week, the first month. Like, wow, this is really cool. I know a lot of people did some consumer type of things like trip a three-day plan in Lisbon. But then we also took email, email threads, we cut and pasted it in there, we summarized it, maybe even created an email response. So that's kind of taken that first pass of here's technology, the things I do, how can I extend that, make it better? I think some of you are probably in that second phase. And that second phase is when we actually reshape how we work. technology. And this is kind of going back to the drawing board. What can we actually do when we take this technology as a foundation and build a new work around that? pattern's been around for mainframe, PC, internet, probably mobile the biggest one. I remember back when I was early on doing mobile, I was the first person to hire that on a team at Microsoft search for mobile search and all kinds of customer visits. So we were building ⁓ the ability to have maps. This is before the iPhone. So three or four years before iPhone launched. Maps get you from one place to another, which is pretty common now. Everybody's using that feature every day. But when I talked to the customers, how they solved that problem that day, do you guys remember MapQuest, some of you? They used to come in with their printouts of MapQuest. And when we show them what they're gonna do, they're like, wow, that's kind of magic. But nobody ever thought that you're gonna be carrying around these. Mobile phones. That could do everything for you. You're watching videos, you're buying cars, you're booking hotels, all those things. And I think with AI it's gonna be even bigger than that. So this is how I approach things. Everything I always think about is there's customers, product, team. Okay, what changes now? So what's happening with our customers? What does this mean? Product. What's expected of the product from our customers? What are the new capabilities that we can provide? What are the capabilities we have to support so we don't become Kodak? And then last, like I mentioned, the team. How to best approach this? Building a new product? Dealing with the customers needs a different request. So this would be the spine of the talk today. So I mentioned the three buckets. But the one thing I'm finding is you can't treat them as silos. They're one big system. You pull one thread and the other two move. The customer's expectations shift, and the product needs to be rethinked. When that changes the team that you need to build that new shape. So you can't touch one without the other two moving. And that changes the role. The role has to elevate as product people. You can't just run one dimension really well and expect the others to to stay the same. So you have to be brilliant, not only at building the product, but really understanding the customers the teams that you need to provide the new system for. And what's interesting now, they're all moving at the same time. And they're moving very fast, all on a diff different activation curve. So the new norm everything moving all at once. Customers, product, the team changing simultaneously at different places on the activation curve. So we used to feel the ground shift every few years ⁓ mobile took off, the iPhone moment, everybody remembers that, that was a big shift for mobile, cloud, so on. And this one's happening right now. So this change isn't something you prepare for you recover from. We're all in it for the next little while. signals, there could be many, many. I try to keep it to three. you probably fill in your own org right now, and I think there's gonna be a lot of talks around this. The role is being rewritten. So the skills that mattered before, writing the perfect requirement, changes to holding a great conversation, steering it, knowing when you hear an answer that's a good one. The role has to elevate to that. Magic is now mundane. The expiration date on wow has gotten brutally short. A feature that earned you an applause at a demo last year is just table stakes. The customers don't thank you for it, they assume it. Last one, the big one, I think there's gonna be a lot sessions on this one too, is when you can build almost anything fast. The question comes from: can we to should we? Judgment, taste, prioritization becomes a scarce resource. Last week I was just talking to my peer booking, and we used to talk about trying to do annual planning for 2,000 engineers. We had a huge backlog of things. And we could only usually get to the third of it. So we had to prioritize the most important things. But imagine now if we actually had magic and we could build all those things. Who's the custodian of that product just to make sure we don't build a Frankenstein? Thinking about what you need to build is super, super important now. We used to hide behind the fact that we can build everything, so we are forced to prioritize. What do you do now? Alright, let me start with the customer. hope you guys all have secret weapons. I've been lucky enough that all the companies I worked at, we had something called the customer advisory board. So for us, this is full of people already activated the future. We talk to them, we study them, we learn from them. We're actually co-building with them right now. So I spent two days with them in Prague. Just last month. And the one, many stories, but one that's kind of local to hear. person actually from Lisbon lives south of the river, was telling me how they're leveraging AI and vibe coding. And they're taking our platform, which is CRM, so it's a sales cycle, but there are other apps they use as well: marketing, project management, and so on. And he's telling me that they're using Claude. Where they're actually white coding apps that sit on top of both of those to fulfill a need that doesn't exist today. So this idea of building these small little applets to build products or even features is now possible. But the thing that I thought was the most interesting, and I want to share for this talk, the people that are building this, he was telling me it's a sales manager. I'm like, wow, a sales manager. Usually the opposite in your company of somebody who's the least technical is encouraged, supported to vibe code an app that they will use in production. And I thought, wow, that's that's the future. If I came in here two years ago and told you this would happen, you would probably think I'm some crazy guy. It's happening today for those that are activated. So for those nerds, we have to figure out how do we support them the best? What do we need to do? And Skip the slide. back to customer expectations. They're not just benchmarking you to the competitor, which is usually something we fall into. Well, the competitor doesn't have this, but our your customers don't use competitor projects all the time. What they're benchmarking you against now is the best experience they touched yesterday from anyone in any category. So magic has a half-life and it's collapsing very, very fast. Every delightful thing that you ship decays into the expected faster every cycle. So how do we deal with this? the early on, natural instinct, what probably everyone did, we did, is to ⁓ the solution, spread AI across it. I want to argue the opposite today. Resist polishing the solution, so reopen the problem. So here's a risk. So every existing solution, the one you've been focusing on for years, quietly becomes ⁓ the definition of the problem. So always go back. What is the customer problem we're solving? And it's very hard because as soon as we figure out the problem, we need to talk about solutions because we build solutions. We don't build problems. So sometimes you see you stop seeing the customer needs, especially the new ones, and base reality on the solution that you have today. I go back to that sales manager. When they describe what they're doing, it's like, wow, we never even knew that existed. Because in our previous customer interviews, we didn't think about two systems having a single feature. So think about it. The solution should become the disposable thing. And the problem is the thing that stays around. So what we're doing at PipeDrive, this is a fake journey map. I actually used Miro for this. We went through the customer journey. We kind of extended it as well. And then we revisited all our jobs to be done. And then we're going through one by one to try to figure out. Is in the future is this still gonna be human? Will this be human and agents? Will it be agents? Because for us, maybe many of you, customers won't just be humans anymore. They're gonna be agents. Agents have different requirements. Agents can do things that humans can't really do physically. For example, bulk updates. Do you want agents to have bulk updates? A feature that you allow your users to do right now, things can go very scary. So the customer problems is a stable thing. I really recommend that you go, you go back and do this. Even more important than ever. You always had to do this, now you have to do even more so. Quickly, let's get to product. What do we do? What do we build? We talked about the customers. Here's price something that may be new to some of you. Have you guys ever heard of the futures cone? Verse? we have a few people. Great. This is something used in the past. And I recommend that you go have a look at it. What ⁓ what this allows you to do hold ⁓ outcomes at the same time. So it's a very easy way, very lightweight framework where you can talk about, well, where do we think we'll be based on assumptions that we have today, our understanding of things, which we know will change very quickly. So now you can do three or four. We have words for this. Plausible, possible, preposterous. And the idea is you come back here quarterly and you revisit this. I put preposterous in there, and I think you should do this because it allows you to kind of think really, really crazy, because it's fun. Preposterous for a CRM would be the one our example, and it is preposterous, in the future, salespeople vanish. You don't need salespeople anymore. If you don't need salespeople, then you don't need a CRM system. Because why we have a sales team? Sales team is to get business, sell the products or services. In the future, if you're a hot tub maker, maybe you just come into a system and you say, I could produce five hot tubs a month, you sign up to service and you just get the purchase orders. Preposterous. Maybe not. Here's another uncomfortable truth, especially for successful companies and successful people. You've optimized for a box. And the box just vanished. Mobile first, cloud, architecture, your org chart, your roadmap. You got brilliant at solving inside the constraints that's been set years ago by the business, by the ecosystem, by the company. And that mastery felt like maturity, like safety. So you kind of stick with that. Guess what? When these constraints disappear, the mastery may become a blindfold. So maybe optimizing beautifully for a thing that doesn't exist anymore. Again, back to the Kodak moment. So what do we hear? What do we do here? Again, kind of touched on a little Is mature products like ours, 16 years old, we have two possibilities. First is easy, spread the AI goodness and add the capabilities, improve what's there. We had to do that. Everybody was doing it. Customers were expecting it. But now we're starting to find, well, there's only so far you can get there. So the second part, the hard one, is the one you have to go and rebuild to the core, re-architect the parts. around what's possible, what's expected by the customer. It's slower, it's scarier, but this is where the real refit happens when you stop decorating the old box. concrete example of the box vanishing, interface flipping upside down. So for thirty years we've been teaching our customers, our users, how to use the system that somehow we constructed. For now, in the new world with natural language, the software comes to the user. Like I mentioned, we w who would have ever thought that you're using natural language to develop software, like in that case of the sales manager. You have to think about that. What does that mean for your product, for your customers? On top of that, when agents come, how are they going to interact with your system? Last is the team. Probably one of the most important. What do we do? How do we evolve to meet those needs? Do we still have the same ratios? Do we still do the same things? Do we use the same tools? one of the benefits of my job, like I mentioned, I get to see end to end the whole life cycle, living with AI. so the first truth, everyone gets more powerful, especially in the first phase when we start sprinkling AI. The shift really happens you start removing those boundaries, ⁓ the teams are starting to act like a built team instead of separate functions. One of the things that we did early on in January, we formed, let's say, a team of around 25 people. And we told them, remove all the constraints you have. Use whatever tools you want. We're not going to hold you to a timeline. Take as much risk as you want. Forget about our process. Forget about Reports and so on, just go fast. After a couple of weeks, the team was getting so fast they were developing things in a matter of days that would have taken weeks or months before. But the interesting thing we found as we're getting ready to actually put this in the hands of customers, which was only three or four weeks. All the other bottlenecks appeared that we've never seen before. So if you want to do localization, go to market, get the security to team to buy off, get the lawyers to buy off. All of those things all of a sudden used to have two weeks or three weeks SLA, like we can't do that. We're building a product in four days. can't wait two weeks to get this translated. So now we're doing that second phase. We're going back, okay, what do we really need to do for the whole company to be re-architected? So go as fast as now the fastest piece of the S D L C life cycle. so if building is trending to be almost free, what's actually scarce? And I'm sure you'll have lots of sessions on this today. And words using, it's very true judgment and taste. when anyone could build anything almost instantly, the mo the more it's no longer can you make it, it's knowing what to build and whether what you've made is any good. Not sure how much time you guys spent on that, but this is really, really super hard. It's new. So evals, try to figure out when machines are building things for you, when machines are r running non-deterministic things, how do you actually figure out if it's good enough? 80%, 90%, thumbs up, 70, what do you start with? You don't want to give your customers something so bad that they'll never try it again. But you also want to get in there to start building something that day, fit the prop, fit the product. So for twenty years the craft pushed us towards running the machine. Now the machine can run itself. So which means that the job finally gets to be the one that drew us to it. So taste, judgment, and knowing what matters. So four things for you to take back on Monday. So I talked about the the framework. Try out the cone. It's a very fun exercise, even if you ever never use it. ⁓ actually got gutsy and I pitched our three year product strategy, which I don't usually like to do, to our board six weeks ago, and used the cone framework and they loved it. We didn't have any arguments ⁓ on the future looks like, because we said we don't know. These are four different things. if you do the cone framework, what I actually found is the things that you actually have to do, the thing you do in the next three, six, nine months, there's a lot of overlap. So it allows you to get going without having to put a lock in what it is in three years that you're going to be shipping. System thinking, like mentioned, product, customers, and team, they're intertwined. So you also have to level up and think across the whole thing. Third one, problem, how we solve it? Who are your customers? What are your customers looking for? Are there to be done still the same? Has it changed? Are there new ones? Should you retire some? What do you focus on? What can you do that still has that ethos of your current so you can take your customers forward into this new world? And four, I would like to fun. 'Cause you can't really do something really well if you're not having fun, if you're not enjoying it. So it's kind of the moment right now where product leadership matters the most. So don't be don't wait to be invited. Step into it. the future's arrived. Header your plan, and that's the good news. You decide what's next for your company, for your product. So you must move from what you've accumulated to what you choose next. Go build. Thank you.