Marty Cagan: Strong Opinions, loosely held
In a Nutshell
Marty Cagan regrets underestimating project feasibility and overemphasizing problem identification rather than solution discovery. He criticizes the obsession with predictability through roadmaps and PRDs, which stifles innovation and results. He now emphasizes that product leadership, political awareness, and genuine thinking—rather than processes—are essential, especially as AI makes outcome-focused product work more critical than ever.
These notes were generated by AI and may contain inaccuracies.
Thank you all. Thanks also to Leni for organizing this conference and inviting me to be a speaker. Actually, I don't attend many conferences, but when I do, I really enjoy them. I cannot count the number of people who have thanked me for the content of my books. Some people asked me about their torn copies of the book Inspired and requested my autograph, which is both an honor and a very humble thing. I pointed out to some of them that my talk was about my regret, which is indeed a lot. This is what I wanted to talk about, but I was with you all day listening to all the lectures, and I learned something interesting from each one of them.
I have noticed, and I am sharing this with you for a reason, that there are two types of products in the world. Only two types. I have always called them the project model and the product model. Approximately half of the speakers were from each side. It's clear I'm very biased, but my world includes both. So, it was very interesting for me to listen. But I hope you all remember that day, and that you saw completely different definitions of the job. In fact, while I was listening, I was thinking, oh, I should talk about this, and I should talk about that. I don't want them to leave thinking like that. But then Robbie from Google and Carrie said exactly what I was going to say. I liked that. Again, I'm biased, but I liked them.
What I want to talk about here is slightly different. Honestly, it's something I've known I needed to do for a long time, but I kept putting it off. Because this is a really difficult question, especially for me. I have spent decades discussing certain points. Now, for those who don't know my job, I am actually one of those people. I am not interested in the latest operations. I'm not interested in the latest software frameworks. I don't really care what program you created to automate. I don't care. Because this is not actually the product's job. This may give you some tools to help you perform the task, but that is not the task. The product's mission is to solve our customers' problems and achieve results for our business. This is the crux of the matter. So, that's what I care about.
Because I focus on principles, the truth is that they are very solid. In fact, every time a new technological wave appears, you hear the polite phrase that I must be too old to have seen all these waves. But, yes, with every wave you ask: "Will these principles be invalidated?" You don't really know until you take a look at it. And of course, as you know, all those waves were extremely enjoyable. I am one of those people; I mean, the reason I am a product-focused person is because I love the fact that what is possible now is constantly changing. That's what makes it special. Otherwise, I would not be working in this field, and I would not have chosen it as a profession. But with artificial intelligence, I felt I had to look at everything, even things I had learned were sacred. As if I am not tampering with this principle. I wanted to study everything, and that's what I did.
Now, first I had to decide where to start. Because I actually treat my content as a product, and I discover the product weekly. In fact, with the product differences, many of you - who have been with me - know that this is what I do. I am learning. What I was doing here today was discovering the product's content. It was very helpful for me to hear these different opinions. In any case, the truth is that I learned a lot. Now, most of the time, I write articles and publish things about these simple lessons. But what I wanted to do was define what I thought, and frankly, I started making a long list. It's rather frustrating, but it's a long list. I wanted to choose the top ten things I truly regret. I wish I had known then what I know now. I decided to stop at the publication date of the first edition of the book Inspired in 2008, when some of you were not yet born. At this point I draw the line, but you should realize that I had been working in the products industry for 25 years when I wrote that. I was a product manager, I went through the entire engineering side, and I was a product leader for two and a half decades. So, I learned a lot from some great people at some of the leading companies at that time. Truly excellent companies. So, I decided to stop here, although some of the things I will talk about are regrets that are less than six months old. Let's begin. Let's talk about it. Ten important things.
First, I would like to start with the feasibility of the project, because frankly it is the most embarrassing. I have no escape from this. I have completely underestimated the project's feasibility. So that everyone understands what I mean, the feasibility of the project. One of the tasks of a product manager is to ensure, at least in the product model, that the solutions will be accepted by customers and that they benefit the business. This means that you can market, sell, and provide maintenance services for them, and that they are legal, compliant with laws, and respect privacy, security, and ethics. This is extremely important. Even before artificial intelligence, it was like that. But those of you who work on artificial intelligence products know that it is much more difficult. Project feasibility is a very broad field. But the embarrassing truth is that in the first edition of Inspired, I didn't address this as a risk. Value, ease of use, and feasibility were three risks we discussed. As for feasibility, it was completely neglected. I feel embarrassed to say that. I addressed this issue in the second edition, but ten years later. This is one of the things that made me think a lot, because I kept wondering: "How could I have gone on for so long with such a big blind spot about the product?"
I believe I know the answer, and I want to share it with you because it is still very relevant today, perhaps even more important today. In my career, for those who may know, I have worked almost entirely on developer tools and platforms, a field that I still love. Who doesn't love cloud computing? I personally love this field. However, you should understand that this is one of the few areas where a product manager can overlook their weak business skills. This is not the only reason, but it is one of them. This is not a criticism, but a feature. For this type of product, this area is much easier to make viable. But this does not apply to most products, especially artificial intelligence products. In fact, I was very excited about my engineering knowledge because I felt it helped me learn new techniques. I have an engineering background, which is what I studied, and that was a great advantage. I felt it was my superpower. But today I say that for all of you, and to achieve success at all levels, commercial viability must be fundamental, especially from the perspective of comprehensive systems thinking.
Second, identifying the problem. In fact, as you know, I have always talked about product discovery, which involves problem discovery, that is, identifying the problem that will be solved, and solution discovery, that is, solving that problem. In fact, my mistake was that I did not realize how much people are drawn to the problem-discovery side of this equation. In fact, many see themselves as gatekeepers. This is their job as product managers, to make sure everyone agrees that this is a problem that needs to be solved. Yes, it goes without saying that you need to make sure you understand the problem, who you are offering the solution to, and what the definition of success is, but honestly, this is not difficult. It's not difficult at all. If you spend too much time on that, you won't be able to do the real part you're being paid for, which is figuring out the solution. You might think that the product's failure is due to insufficient demand for it, or that an insufficient number of people are suffering from this problem. But you are often mistaken, because as soon as someone else offers a better solution, we suddenly discover that the problem never existed in the first place. Therefore, I had to emphasize the importance of identifying the problem. Yes, identifying the problem is important; take some time to discover the solution. This is the essence of the matter, and this is where innovation arises.
The next big problem...and this is another problem. All of these things have positive and negative aspects, but I spent a lot of time talking about the importance of understanding why we are working on anything we are working on. I did not invent this phrase; it was John Doerr who did. But we need teams of missionaries, not teams of mercenaries. If you want missionaries, you need to make sure they understand the reason. Yes, this is important, but it's very easy. In fact, if you have a product strategy, which you should have, and we will talk about that later, then the reasons for your work on these problems will be clear to everyone in the organization. Therefore, there is no ambiguity. What I want to talk about, and I was really pleased when Robbie pointed this out, is that there is a much more important reason than "Why are we working on this problem?" The question is: Why aren't people using our product? He pointed that out. I hope to see that. Frankly, this is somewhat surprising for Google. For a while, they thought the whole thing was data-driven, and that we wouldn't talk to anyone. We really need to talk to our customers to find out why. I am constantly testing products. I bet most of you do that too. This is what product managers do. We are among the first users. Let's try it. Most of them are inferior products. You know this, don't you? We tried it, it wasn't good, so I stopped using it. Almost no one is following up on it anyway. Nothing. No survey, no email, nothing to ask me why I stopped using it. Perhaps the most important question is: Why do people use our product or not? What really surprises me is that this is the key to unlocking so much innovation. However, most teams never ask this question.
The next thing is humility. As you know, the more experience I gain, the more I realize that humility and an open mind are truly critical for product professionals at all levels. What does humility really mean? This means knowing what you cannot know and acknowledging what you do not know. Many of you know this, but the message most people have taken away from my books is that the product manager should be the product chief. This completely contradicts the concept of humility. This may seem like a personality trait to you, but it is not. I believe this contributes directly to the root cause of the failure of many products, which leads me to perhaps my greatest regret.
To be honest, I didn't realize how powerful and deeply rooted the desire to predict was. This applies to both product managers and senior executives, and how this desire conflicts with humility, and even with innovation and results. Imagine for a moment that I am opening the door to discussion. Perhaps I shouldn't, but consider for a moment the two most important elements in the product world: roadmaps and product requirements documentation. Roadmaps, of course, are lists of priority features that we think we need to build and when we should build them. Product Requirements Documentation (PRDs) is the way we describe the requirements for those features. Now, consider for a moment how much these documents contribute to reinforcing the idea that we know more than we actually do. This is a very important point, and many people misunderstand it. As you know, people keep asking the question: "Do we still need product requirements documentation? Do we still need roadmaps?" I hadn't heard this question asked until this morning. Despite Claire's optimism, I assure you that the roadmaps will not disappear. We all know that. The issue is not whether it exists or not, and it never has been. The question is: What is the purpose of using product requirements documents? If you are using it because, excuse me, you are arrogant and think you know the answer, and you put it in the form of requirements for your engineers to build, then that is what is holding the project back. This is the root cause of the failure of many products. But if you have learned, tested, and gathered the necessary evidence to ensure that what is being built will achieve the desired result, then a Product Requirements Document (PRD) is merely a communication tool. It is not only harmless, but actually beneficial. The same applies to roadmaps. If you put together a set of features that represent someone's idea of what might work, and set dates for them, you will waste most of your time. This will end with the waste of most engineering capabilities. Yes, engineering time has decreased significantly, but you are still wasting it, and the code is not free. Therefore, predictability is at the heart of this matter. Predictability is important, but it is not as important or valuable as sacrificing results or trust.
Politics. The truth is, we all know that politics matters, but I thought it was the quality of the product that would decide the matter. That was, of course, naive. The truth is that politics is everywhere. It is ingrained in the fabric of every company. If you look at my content over the past three years, you will find that half of it deals with politics, and the other half deals with the impact of artificial intelligence. It is a broad topic. This is perhaps the subject that has caused the most damage since the publication of my book Inspired.
My initial work revolved around product development and discovery teams, because I love the art of product making. I wanted to write about this art because of its many principles and techniques, while most people were content with building products, launching them, and trying out what worked. This is still true today. This was a deliberate choice to focus on product development teams and the art of discovery. What I didn't realize was that I rarely addressed product leadership, and I didn't understand the consequences of this choice. The result is that for many companies, unfortunately, this is still true today, because despite all these years, the book Inspired still achieves huge sales. Many teams simply read an Inspired book and think that all they have to do is form product teams, empowered product teams, and a strong product manager, and that will be great. They believe that this is all they need, and they will be fine. Of course, what many companies have learned is that in product-enabled teams, you don't just need less management, you need better management. Product leadership is of utmost importance. You have just heard Carrie's point of context. This is the crucial point. At a minimum, it is the product strategy, which is a list of priority problems that need to be solved. Another difficult issue is corporate governance. I am not talking about the project management office, which is concerned with governing operations. I am talking here about corporate governance. Supervising the company. We all acknowledge that this is a problem, but as I said, this is not something the product can do about it. One of the most painful lessons over the years is that many companies have helped improve the quality of their products, only to find themselves ultimately targeted by people with entirely different motivations, seeking to change the course of the product. Therefore, I recommend everyone read Eric Ries's new book, Irredeemable. It may seem a little strange, as the concept of the product is not as simple as I portrayed it in my books. It seems theoretical, doesn't it? Your job is supposed to be to solve problems in ways that satisfy your customers, but at the same time serve your business. That's right, that's correct. But this hides the fact that this is not enough. You actually need to solve problems in ways that are much better than your competitors if you want to convince people to switch to you. The truth is that competition in today's open market for products is more like a bloody, vicious game. Most of my followers won't realize that, and many teams are unprepared.
And the last point, as you know, this is a point that some people have raised, and I am glad to see it, but the essence of good work on a product is thinking. It's thinking. And this is where the mistake lies. I really appreciated the lengths to which people would go to avoid thinking. I wish I was joking. How does that appear? It manifests as a passion for processes, frameworks, and prediction. I structured Inspired based on people, processes, and product. A grave mistake. I am concerned that large language models will be used as a substitute for thinking, but the fact is that this has already happened. With the operations. The truth is that Elon Musk was right about this. In many companies, processes are used as a substitute for thinking. So please don't believe me. I should have pointed that out, pointed to the thinking, and pointed to the essential basis of product sense more clearly and explicitly. Okay, these are the top ten things that I really regret.
However, I will tell you, ironically, that I feel very optimistic about the content in general, and more importantly, about our future as a product. And I want to try to explain why. Frankly, the main criticism I've received about my content over the past 20 years is that people wish they could work as described in the book Inspired. They really wish that. They tell me there are two problems: First, their companies are addicted to output and predictability. Secondly, it is difficult. They believe that working in this way will be difficult. But look, thanks to artificial intelligence, today more than ever there is an understanding of the need for results and not just outputs. It has also become much easier. As you know, we talk about product discovery as being built for learning. I hope you will not leave this session without understanding that there are two types of building: building to learn and building to profit. This term is derived from a term that many of you are familiar with, which is the term Jeff Patton. But we have many terms for this concept. But the difference lies between discovering the product and presenting it. If you want to spend your time writing user interface modification requests, that's fine. Perhaps you should become a developer. But if you want to make sure that what your software engineers are building and delivering to your customers is for the purpose of learning, then you should build with the goal of learning. It's a completely different approach, and a completely different job. Your job is to make sure that when they build that thing, you achieve the result you need. So, here's the thing. Thanks to artificial intelligence, the principles of product modeling, product strategy craft, and product discovery have become more important than ever. Good. I really hope this is helpful to everyone. I would like to thank Shreyas Doshi and Teresa Torres who gave me feedback on the first editions of this. And I would like to thank Leni again for organizing this meeting. I hope this is helpful. Thank you very much.
Keep Lenny's Podcast in your library
Save the videos and channels worth coming back to, and find them again in one place.
