The GTM Podcast is available on any major directory, including:
Oji Udezue was most recently Chief Product Officer at Typeform, where he led product strategy across the company’s growing suite of tools. Previously, he was CPO at Calendly, where he helped scale one of the most beloved PLG tools in the market. Before that, he was Head of Product for Creation and Conversation at Twitter, leading core features like Tweets, DMs, and Spaces. At Atlassian, he led all communication products, including the launch of Atlassian’s first post-IPO product.
Oji also spent years at Microsoft, building foundational experiences across Windows, Outlook, Hotmail, and Internet Explorer.
In addition to his product leadership, Oji is the founder of Kernel Fund, an early-stage fund supporting startups across Africa. He writes about product, growth, and leadership at his newsletter Product Mind and is the author of the upcoming book Building Rocketships—a guide for high-growth product teams.
Discussed in this Episode:
- When to hire your first PM (and what to look for)
- Why great product teams include marketing from day one
- A framework for picking winning startup ideas
- How AI is changing how we build (not just what we build)
- What product leaders can learn from Microsoft’s Longhorn failure
- Why 3x better isn’t a nice-to-have — it’s a must
If you missed GTM 143, check it out here: Why Most AI Messaging Fails and How to Actually Stand Out in a Crowded Market | Harmony Anderson
Highlights:
04:52 – Why marketing is essential to product leadership
12:02 – How AI is transforming the process of building products
17:07 – Common product mistakes founders make early on
24:16 – Building Rocket Ships: Why Oji wrote the book
31:08 – Framework: Where to fish for unicorns
33:17 – The “3x better” rule for product value
37:07 – The challenge of true product differentiation
39:22 – What Microsoft’s Longhorn failure taught Oji
44:51 – When (and how) to hire your first PM
48:31 – The importance of fast, early customer feedback
54:46 – How AI is changing go-to-market strategy
57:32 – Why marketers must master signal in a noisy world
Guest Speaker Links (Oji Udezue):
- LinkedIn:https://www.linkedin.com/in/ojiudezue/
- Kernel Fund:https://www.kernelfund.com/
- Product Mind:https://www.productmind.co/
- Building Rocketships (Book):https://a.co/d/0HGjf18
- SubStack:https://ojiudezue.substack.com/
Host Speaker Links (Sophie Buonassisi):
- LinkedIn:https://www.linkedin.com/in/sophiebuonassisi/
- Newsletter:https://substack.com/@sophiebuonassisi
Where to find GTMnow (GTMfund’s media brand):
- Website: gtmnow.com
- LinkedIn:linkedin.com/company/gtmnow
- Twitter/X: x.com/GTMnow_
- YouTube: /@gtm_now
- The GTM Podcast (on all major directories): /gtmnow.com/tag/podcast/
Sponsors:
ZoomInfo’s GTM25 Virtual Summit
Wednesday, May 7th 2 PM – 4:30 PM EST | 11 AM – 1:30 PM PST
This is your moment to rewrite your Go-To-Market playbook with AI-powered strategies. Join ZoomInfo CEO Henry Schuck and top industry leaders to learn how to accelerate pipeline, boost close rates, and transform your revenue teams. The future of GTM is here.
Qualified’s AI SDR Summit
AI SDR agents are rewriting the pipeline playbook.
The AI SDR Summit, put on by Qualified, just took place. If you missed it or want to go back through any parts, the summit is now available on demand.
It explored how leaders from companies like G2, Gartner, 6sense, Salesforce and so many more are scaling their pipe gen with autonomous AI agents.
You’ll learn everything you need to know from AI SDR use cases for inbound, use cases for outbound, the landscape putting a price on digital labor, and exactly how you can use AI SDR agents to transform your pipeline generation.
No fluff—just real strategies, candid conversations, and proven results.
Learn all about AI SDRs at qualified.com/gtmnow which will also be in the show notes.
The GTM Podcast
The GTM Podcast is a weekly podcast featuring interviews with the top 1% GTM executives, VCs, and founders. Conversations reveal the unshared details behind how they have grown companies, and the go-to-market strategies responsible for shaping that growth.
GTM 144 Episode Transcript
Sophie Buonassisi: Hello, and welcome back to the GTM Podcast. This is Sophie esi, VP of Marketing at VC Firm, GTM Fund in our media brand GTM. Now I’m joined by one of the most experienced product leaders in tech, Oji, Esue. Oji, welcome to the podcast.
Oji Udezue: Thank you, Sophie. It is such a pleasure. I’ve been looking forward to this for some time. Thank you.
Sophie Buonassisi: Likewise. So have I, and I’m gonna dive into a quick bio for the listeners. Oji Udezue was most recently the Chief Product Officer at Typeform, where he led product strategy across the company’s growing suite of tools. Previously, he was the CPO of Calendly, where he helped scale one of the most beloved PLG tools in the market.
Before that, he served as head of product for creation and conversation at Twitter, leading features like tweets, dms, and spaces that we all know and was previously at Atlassian, [00:04:00] including the launch of Atlassian’s first post IPO product. He spent years at Microsoft building foundational experiences across Windows, outlook, Hotmail, internet Explorer.
So you can see why I framed it as one of the most experienced product leaders in tech. And in addition to his product leadership, Oji is the founder of Kernel Fund and Early Stage Fund supporting startups across Africa. He writes about product growth and leadership at his newsletter, product Mind, and is newly Congratulations.
The co-author of a book Building Rocket Ships, which is really a guide to high growth product teams.
Oji Udezue: All those things are true, I swear.
Sophie Buonassisi: Can confirm. Can confirm. Oji, we have a ton to dig into, but first, you have, you know, incredibly impressive background with a breadth of experience across startups and larger enterprises and served as a PM and product leader. I. You also have product marketing experience, including being a senior lead product marketing manager at Microsoft.
How has this [00:05:00] experience being directly embedded in that go-to-market engine impacted your approach to product leadership?
Oji Udezue: I had spent almost a decade at Microsoft, and I had done it through the Balmer years. The Balmer years were weird, right? They were productive in terms of what Microsoft made, but also just evolutionary was evolving. Windows were getting ass handed to us by the DOJ and because of all anti-competitive stuff we’re doing Internet Explorer at Microsoft.
And we were watching the rise of Google and Amazon and so on and so forth, and just not having a clear map to where to take all our success into. So it was both productive but also lots of change happening. And the reason this is important is we are in a very impressive accelerated change environment now.
I’ve been building products for a decade. I’ve built Windows windows, live Hotmail, all kinds of stuff. Search, right? Some search [00:06:00] innovation. Through those years I had 12 patents in my product work. But I was getting frustrated. being led by a sales leader wasn’t like.
I just, it wasn’t Gates. Gates wasn’t at the helm anymore, and that was a huge draw for me. So I decided a couple of things. When I went to marketing, first I started an MBA, and then I, Dave Melin, who was in the Dev, he a good friend of mine still, I just met with him about a few months ago, gave me a chance to become a product marketing manager.
And what I wanted to learn was, okay, I built it. How do we get it to customers? What do they care about? What are my tactics? Digital, offline, and so on and so forth. And I ended up actually doing the whole range of that stuff, plus all the pre-sales. I remember traveling the world to equip salespeople to sell Visual Studio 2013 at the time.
So my whole intent was what are the new skill? Can I learn? How do I not be bored and do the same thing at Microsoft? And there’s this thing that we do, we hand [00:07:00] off to a bunch of people to go sell our stuff. Like I wanna know how to do it. And so that’s why I did that.
Sophie Buonassisi: it. Well, I’m sure that helped inform the way that you build product now, because of course it’s the go to market podcast, so definitely all ties together.
Oji Udezue: I love marketing. Like sometimes, product people think of marketing as, almost like a great evil on the other side, not pure, but I have learned to understand that product people, we have really two roles. One is we delight the customer and we make a lot of money through a business model.
And that last part cannot be done without marketing. It’s very hard. And if you learn to love marketing, I think you become a better product person. Honestly, and actually through my experience at Microsoft, I’ve grown to mentor CMOs or directors and up who are in the marketing track.
- Every time in the CPO role, I consider marketing like a core partner. I start working with marketing. When we start [00:08:00] thinking of the solution, when we’re doing customer discovery, marketing is involved. And I find that kind of strong partnership, very early partnerships, so beneficial in getting great products out into customer’s hands.
I can’t sing the praises of great marketing hard enough. I cannot.
Sophie Buonassisi: As a marketer, I feel like I’m a little biased, but I love to hear it. Are others involved? Break that down a little bit more tactically for me in a way. Who’s in the room when you’re making these decisions? Who are you pulling into that room? Is it just marketing or are there the other folks?
Oji Udezue: So this is a really big focus of the book that a and I wrote. We think that. Engineering product design as the core of product development is so 1999 in many, cutting edge companies, it hasn’t been like that for some time, although some still do that. They build a thing and they hand it off to marketing, to craft and message houses and to do all these things.
The very best teams that we’ve observed, and we’ve been part of the [00:09:00] creative team is six plus disciplines. It is obviously traditionally product engineering design, but it’s also product marketing Very quickly. It is user research, right? For more extensive kind of market sensing. ’cause you get two things from product and user research.
Good marketers sense the market opportunity. It’s a core function of product marketing. What, what will sell, what is out there, what are the conditions? You can’t even do strategy without that good user research. Tells you about what will delight the customer. Again, the two polls for what product people need to actually deliver on the sixth team is usually data, use a lot of really quant, a lot of product building requires quantitative data.
So as a chief product officer, I get these six teams, whether or not they work for me, sometimes they don’t. To work together from what a physicist will call T equal to zero. Meaning when we [00:10:00] even start to have an inkling of the problem, of the market opportunity, of the thing that will both make customers and will fulfill our mission as a company, but also make money.
All six representatives are there. And the beauty of that is that it prevents you from doing all this relay races. Do you know product managers write a spec? Developers develop, and then after you’re done, someone does some acceptance criteria and says, Hey, marketing, here’s the thing we got.
Actually, as the velocity of product development speeds up, there’s a lot of things coming out and you have to worry about customer impotence and how you make noise in the marketplace. Having marketing at the table super early on is so important. Now, in reality, we also talk about sales account management and customer support.
These are the sensing early nervous systems of customer satisfaction. [00:11:00] We do a lot to integrate those people directly into the product development process. Support leaders are honestly worth their weight in gold because even before I release anything, I run it by them because. They have taken a thousand calls if you’re good sized company.
And so they tell you what’s up and they inform and shape even the marketing insight on what customers really want. Those six teams are the core, I think the new core what we call the shipyard and the other three teams, you should integrate as closely as possible. Those are your sensing system.
Sophie Buonassisi: Got it. And I think you mentioned to me previously you view marketing as the nervous system and those
Oji Udezue: Yes, because the fix is the nervous system, because that’s the thinking, the creation, the all of that stuff. The customer focused teams, the ones that are really out there, sense those things so that the nervous system can process it. So the six teams is the nervous system that you have to assemble. It’s not enough to do EPD anymore.
I think the very best [00:12:00] teams that’s too slow basically.
Sophie Buonassisi: Love it. obviously AI is top of mind. In your view, how is AI transforming their product, lifestyle lifecycle and impact, and even the way that product leaders lead?
Oji Udezue: Yeah, I’ll quickly touch on leadership and I’ll talk about the core beat of it, which is product Mind the consulting group that I am with my wife and thinks about how we can use the insights from the book to help companies grow. We start to think about how AI is gonna fundamentally change, not the products we build, but like the process of building and across small companies, startups, and also big companies.
Because. There’s a fundamental change happening and we want to be at the forefront of it so we can help other people. Acclimatize to this just connected to everything I’ve said so far, there are really three phases of product development. There is opportunity finding or customer discovery, whatever you call it, [00:13:00] customer listening and opportunity finding is twofold.
Like I mentioned. It is what’s happening in the market that is available What do customers have problem with that we can exploit in relation to that so we can make money, right? That’s the thinking that goes into that phase. The middle phase, the one that we spent billions on and the one that people are most focused on in the software industry is the build phase, right?
We are gonna make great user experiences, it’s the code, work is a scale, blah, blah, blah. All that really hard stuff. And then the last one is the thing that people ignore. How do we get into customer’s hands in the best way? A good way to know that we ignore it in the digital space is the physical people, the pg e, and the apples and the hardware people, they are so good at this, right?
Their marketing is generally great, their packaging, like first unboxing experiences is a whole thing. And in the digital space, I don’t think we pay as much [00:14:00] attention to that, and that disparity tells us that we’re ignoring a huge part of customer impetus getting into a customer’s hand generating demand.
That’s the last part. Now, how does AI affect all of this stuff? It’s speeding it all up, up, okay. Ironically, the first part of speeding this up happened, GPT 3, 2, 3 years ago with marketing copy. That’s a tiny slice of marketing, but it goes into demand generation, and so that was revolutionized.
We had companies like Jasper ai, copy, whatever, come out and just speed that up super fast. But it was just a tiny slice of marketing. you are seeing that with like lead generation and so on. There’s a huge amount of innovation going on there. But the place where product world, software world is focusing on right now is actually like development.
You have client, you have all these coding tools, you have vibe coding, you have cps and so on and so forth. That stuff is revolutionary. I’ve written more code. I’m a PM by the way. I’m [00:15:00] an engineer. I have a master’s in engineering, but I’ve spent 20 years helping invent the future, not from a code perspective, but I’ve written more code in the last year than I have in the last 10 years because the tools have made it so much more accessible for me as a non-professional, that part is speeding up.
And of course the earlier part doesn’t speed up that much ’cause people have to sit and think. What do customers want? So I think the biggest change is that the pace is accelerating, but it’s not accelerating in the same look if pace accelerates and that all phases accelerate at the same time, actually it just works because all you did was just increase tempo.
But when they accelerates and they accelerate at different rates, managers and leaders have to think differently. How do they make it all work? Because all the assumptions, all the tools gone, right? What happens when there is no specification from PMs and the specification is actual code that [00:16:00] works, right?
Because it’s easy to do. What happens when developers can do 10 x coding? What happens when marketing and demand generation is still slow related to that? Do customers get inundated? These are all things that leaders have to think through over the next few years. The last thing I’ll say is that one of the big challenges I’m hearing from leaders, and I’ve had building AI products is.
Talent. What kind of talent do you need? If you leadership, that’s a huge pro. Not a problem. A huge challenge. ’cause you have 50% of the time is either communicating or hiring. And then the last one is just uncertainty, right? When you’re in the beginning of a cycle, you can’t see the future. You don’t know which way it will go.
At Microsoft, we thought browsers were the beginning and the end of the internet, where well, it turned out it wasn’t. It was social media, it was search, it was Salesforce and the enterprise. And we missed a bunch of that stuff because we were so focused on the browser. So leaders have to navigate uncertainty.
They have to talk to maybe people like me [00:17:00] who compare into the future. ’cause they’ve seen two versions of this.
Sophie Buonassisi: You’re building a few products. You’ve mentored and invested in many startups. What are common mistakes or maybe even misconceptions founders make in product development?
Oji Udezue: I try to be prolific. So one of the things that I’m doing as I see this supercycle happening is I’ve started a fun pH that’s studio and it builds two companies a year and invests in the rest that we can build. And I’m building two companies right now.
One is a conversational ai. Technology for managing communities, for companies, for marketers, CMOs. ’cause everyone is building with a community, whether you’re a creator or an actual company. The community’s important to, testing stuff, but also making fans, which is, people in marketing understand which really impacts your revenue.
And the fact is, when I was at Twitter, every time I wanted to do that, build a whole new thing. ’cause I had a Twitter labs division where I was [00:18:00] experimenting with new things communities, which was like Reddit and the next version of dms. I needed to assemble a small community to test it. Sometimes there was even non code or some I was, I put 5 million into, not me.
Twitter through me
Sophie Buonassisi: Yeah.
Oji Udezue: million into figuring out how people can monetize audio spaces, Twitter spaces. And I had to assemble a small group of people to test different economic theories on how we could motivate them, right? To earn money on spaces, which was important. And each time I spun up a new effort like that, I had to hire a human being to manage that community.
And so Cavalry is about doing that with a person augmented with ai. And then the second company is about really like I am trying to make agents more intelligent. Agents are big hypey right now, but they’re dumb and you have to tell ’em everything is a type and humans aren’t that [00:19:00] creative. We miss things even that we need, if that makes sense.
So I’m trying to make them a lot more proactive and intelligent. So what are mistakes people are make? Look, I think at the creative level, when you’re super early on for people in your fund, the big mistake to make is to just work on things that you are passionate about personally. Now, personal passion is important because it’s what gives you energy to stick to stuff, to gain insight into stuff.
What the, maybe the biggest mistake I see from founders is they don’t match their personal interest in a problem with the market opportunity, right? And they miss because building a successful startup has nothing to just do with what you’re interested in and how smart you are. It is solving a product that the market wants and there are enough people, target customers for you to make a lot of money.
[00:20:00] So I think this lesson is perennial. I don’t think it has a lot to do with ai. I think that we have to find sharp problems to solve that have commercial appeal. I find many people will just dig in and I make this mistake all the time. I get excited about technology, I get excited about something and I start to build.
I haven’t figured out if there’s a market for it. I’m sure, as in VC world, there’s a lot of filters for this kind of mistake. In general though, I think that you’re gonna be able to do a lot with smaller teams. So people who are thinking about how to do a lot with smaller teams are going to be building things on a cost basis that is too high for their competition.
And so that’s one thing that founders have to look at. How can I do more using all these tools that are out there? You have to be one of the best at doing that at a low cost basis.
Sophie Buonassisi: And have you found any, perhaps, places for how to learn about these tactics and how to stay ahead?
Oji Udezue: Yeah, that’s a question that I [00:21:00] despair about. I find that there is so much noise, there’s so many experts that are not experts that people focus on. I don’t know a great quality answer to this. I personally listen constantly for wisdom. I Try to listen to people like your podcast.
You have really great guests who have a lot of scar tissue and have had time to masticate what the experiences mean. ’cause you can have a lot of experience and not know how to teach it or communicate it or have thought about what it means. And I wish I had a pat answer for you. When I’m operating for regular PMs, there’s all kinds of tools out there.
Reforge can get you level you up as a pm. There are a few good conferences, but if you are [00:22:00] founder, see, a founder has a thousand decisions they need to get right to get to the other side of revenue, that is the problem. And for most founders, 900 decisions are wrong. They haven’t got it right and they can’t proceed.
And so who’s gonna help you with getting those 900 right? Is a really big impediment. So look, I don’t have a bad answer. I will say that we wrote the book to help solve some of these problems because Isa and I have, by the way, my cohor, she’s my wife. Stafford Engineering Master’s degree executives, chief product Officer as well.
We’ve try to distill both startup and enterprise knowledge into this book to talk about the important things that people need to know. Now, I will be the first to tell you that’s a very hard task for us to do. So I would say get the book. It’s not, it’s actually not for [00:23:00] early stage product managers, it’s for founders.
There’s a lot of lessons in there ’cause I’ve been a founder multiple times. And it’s also about leadership. How do you lead innovation in general? But also find wise people, trusted people who have scar tissue. There are many people in my space, either investment space or product space who make a business about being popular and saying things online.
And I find that maybe like only half of them have scar tissue and wisdom that is practical. And I reflective enough for you to listen to them all the time. I would say if you find those and maybe I can share a list of some of the people that I trust with you or places I listen to or sample that’s what you should do in 2025.
Hopefully if I keep spending more time on this kind of thing over the next five years I’ll help collaborate and create a better outlet for builders. ’cause that’s what really what I’m passionate [00:24:00] about that and making sure opportunities international. I want to have great builders come from the Bay Area, but also from LA and from Johannesburg and from Caras.
And it’s a big passion of mine.
Sophie Buonassisi: And I think you’ve built that actually into the book’s availability, which is really cool. So I wanna jump into the book actually on that note. But before I do, first, I just love how you mentioned scar tissue. I’m gonna swipe that or steal that. ’cause we always call it, been there, done that. And it’s synonymous with scar tissue.
It’s have you actually been there building And it’s something that we always vet for in our community, in, in our media. So onto the book now, you co-authored the book. the full name is Building Rocket Ship Product Management for High Growth Companies.
So first, congratulations on its release, I believe it was in February, and great success that it’s had so far now, you wrote it with your wife, which you alluded to is Zine and both co-authored it. She is an incredibly accomplished product leader, most recently serving as CPO at WP Engine. [00:25:00] So talk about a product power couple.
I’m sure your dinner table chats have way more depth than the usual.
Oji Udezue: So talk about a collaboration. Yeah wow. I feel like, you know, my partnership is in a, requires an Oprah couch thing. We’ve known each other for a long time. We met as friends when we were very young in college, and then many years later we started dating. And we both have almost identical traditional backgrounds, but very different companies.
She’s actually worked for, like working for Twitter was pretty big and however you company, but generally she’s worked for actually companies that are more enterprise and her specialty is, she’ll go into any space, whether it’s construction tech or adver tech, and she’ll make a lot of sense out of it.
User experience, data, all that kind of thing. Incredible leader. But we started to vibe, she was part of kernel fund. We are born in Nigeria. We’ve traveled over the world. We’ve seen builders do similar things in every market space. Spending time as CP [00:26:00] of Calendly was very intriguing because I was a West Coast person, Microsoft, Google, all those traditional companies, Amazon.
And to see a company based in Atlanta try to conquer the world was very illuminating. I was like, damn, you can do this from anywhere. And what is that DNA that you need to be able to do it from file if you have to. And we also spent time mentoring startup teams first in traditional places like Techstars, but also places where minority, people of color women were building, right?
And giving them a safe space and the strong mentorship. And so from that was born, this. Passion for, oh my God, product is this mysterious, and even marketing sometimes is this mysterious, arcane art. You can go to school for design, you can go to school for engineering, but you can’t for product for the most part.
And so we wanna spread the knowledge to more people. And so is, and I said talking about that, [00:27:00] and it crystallized into a book where even in a few months we’re probably gonna give away, a part of the book, like the part that’s for earlier stage PMs and not just leadership. And make it free for people who want it from all over the world, because that’s how badly we want to get the word out on how to do this.
So yeah, that’s the sum of my collaboration. The bad news is our kids are so tired of product management and product building that they don’t have want nothing to do with, they want nothing to do. One wants to be a doctor, the other person is thinking about liberal arts, which we like a little sad, but also very supportive of their brilliance in different directions.
But yes we have to actually limit for our temporary strategy discussions in my house. ’cause it’s, it gets too much.
Sophie Buonassisi: Yeah, I’m sure you could just keep talking forever on it. And in building rocket ships, you introduced the where to fish to land a unicorn, which is a framework and it looks at how broad a workflow is across departments and how often [00:28:00] it’s done. Would you mind just breaking down this framework a little bit in your words?
And how can a founder use this framework to identify high potential market or workflow to
Oji Udezue: Can I just say that you’re a really great interviewer. I like how you’re steering me to the places where your audience will love. Okay.
The benefit of experience and reflection is that sometimes we can come up with entirely new ways to solve problems. And so a few years ago I was thinking to myself, is there a pattern to success, startup success?
And as I thought about it, the Chalene B2B SaaS, I started to notice some things. And so just a, any clever consultant, I knew I had a world into two. I was like, okay. I noticed from my time at Microsoft that its successful is built on high frequent, high frequency workflows. So Microsoft’s whole thing, aside from the cloud is office.
And the reason Office is part of the reason, not the reason, part of the reason office is so successful is twofold. [00:29:00] One is. It encapsulates primal offline workflows that’s needed for commerce, meaning writing, which is word math and accounting, which is Excel and Outlook, which is communication. All of these things are the kinds of things that people do daily across the enterprise, right?
Not just in a single department. So we class those things as high frequency, everyone workflows. So if you divide the world into frequency and who does it, it’s high frequency. It is no fluke that this is where the biggest and highest revenue companies pitched their attempt Google. Couldn’t allow Microsoft to have that in the, created Google Docs and Google Suite and so on and so forth, because they understood that Atlassian, we try to assault that Citadel, right?
With the product that I built, which was communication, which was Slack, slack esque, right? And Slack tried to assault that because they understood that if you can do high frequency, everyone [00:30:00] workflow, you can make a lot of money. ’cause everyone needs it. From the doorman to the CEO As beside that is things what we call high frequency niche workforce, which is like something that a department needs sales.
Salesforce dominant marketing, Marketo, Adobe dominant. And there’s a whole slew of department tech that have built billion dollar valuations, right? That’s the next most lucrative niche to focus on. And then below those are low frequency. Everyone workflows, low frequency departmental workflows, low frequency, everyone workflows, something like expenses.
Everyone does ’em, but you do it once a month or whatever, right? Whatever your CFO does. And so you see lucrative companies, but you don’t see the biggest. And then low frequency departmental workflows are basically yield the least valuation [00:31:00] in general. It’s not that you can’t build a big business on it, but in, in general, in terms of a free, like a plot of who’s being successful where.
And so this is an interesting filter if you do a graph on how much do the target customers use this every week. And on the other axis. Who does use it? Is it everyone in the company or is it a particular persona? Workplace persona? You can predict super reliably how the end game, the terminal value of your startup will be if it succeeds.
And in general, we just tell people, you gotta pay attention to this. Some people will start building in one of these niches without actually understanding this, and either sooner or later they’ll run into problems or mismatching the expectations of how big the market is for ’em. And so it’s a really easy tool for venture capitalists [00:32:00] to think about opportunity, but also for founders to think about, where am I building in which quadrant and what can I expect from them?
If that makes, and this shows up all the time in our consulting, all the time in our mentorship. And it’s not that complicated, but just people are not aware that this framework exists and they can use it.
Sophie Buonassisi: Do you have a visual of that framework in the book?
Oji Udezue: Yes, I do have a visual in book. I have a visual on my substack. When I started talking about this many years ago, I remember I first rolled it out at a product leader summit in San Francisco. I hadn’t figured out how to talk about it really well, and so the NPS from that talk was like was okay, but it wasn’t my standard.
And I refined it because I thought it was an interesting, important message. And, by the time I finished, I rewrote it like three times so that people will understand it and people start to get it. And yeah, at one point it went viral. But, uh, yes there’s definitely illustration on my substack and there is in the book.
’cause I think it’s an important part of finding [00:33:00] the sharp problem that you need to focus your life force on if you’re a founder.
Sophie Buonassisi: Cool. We will link to that in the show notes. And I love that you’re packaging and distilling your thoughts similar to how you would do for an actual product launch and testing the market and refining. Now, let’s say, so you said frequency, how often and by who. Let’s say that we’ve nailed those two.
’cause you’ve argued that, being slightly better than status quo isn’t enough. So I read a quote by you as to lure them, you have to deliver something like three x plus value. For users to actually make that switch. So how can startups ensure that their product delivers that kind of dramatic, no-brainer improvement?
Can you share an example maybe from Calendly or any kind of company we’re hitting or missing? That three X threshold really made the difference.
Oji Udezue: Yeah. I’ll talk a little bit about the theory because part of the point of the book is to show people how to think, so they don’t [00:34:00] just look at frameworks blindly. Not all frameworks will work for you. and you need to sort of understand the internal logic of them to discriminate when to use ’em or not to use them.
So it’s really simple. It’s 2025 and we are designing for distracted brains for whom even AI is not amazing anymore. It is, but sometimes it’s not. It’s yeah, okay, fine. We know this is supposed to happen, so let get to it already. Technology’s a little bit like electricity. You turn it on, it works.
It doesn’t, the people aren’t paying attention to that stuff if they don’t have to. And frankly, proper people shouldn’t want them to. You should just work. So that’s your subject of every class in leisure and business, whatever that is. Now, people get used to stuff the way they do things, even to right now you have comfort levels with riverside or fam because you know how it works or whatever your tool set is.
If something new comes into the market, [00:35:00] whether it’s endorsed by a friend of yours or not, you look at it and say, damn that I gotta switch. And that’s difficult. This is just, we’re still in the realm of psychology. I got into tech. And
Sophie Buonassisi: That’s what a lot of
it comes down to.
Oji Udezue: it’s, I think work managers have to be anthropologists and psychologists.
That’s the best, that’s the best people. And even marketers have to be the same. I think if you’re gonna be good, you have to understand, mad men style, like what motivates people. And yeah, all you CMOs, yeah, go get your psych degree. It’s very important. But the point is to do something else than what you are doing, which is comfortable, you incur what in economics we call switching costs.
So the real question is what can make you take on switching costs? Whether you go from paper to word or from word to confluence and notion what gives you the impetus to expend more energy than you did [00:36:00] yesterday? Right now that’s what we call the zone of benefit in the book. The zone of benefit is constructing enough value that people want to spend the energy to switch.
And there’s several dimensions to this, but it all boils down to you gotta do more for cheaper. Two main things that the book goes into a lot more detail on how to test this, but two main things are if something is five x cheaper, I might try because I’m saving money if something shortens my workflow time.
By three x, 10 x, then I might try it. ’cause that makes a lot of sense to me. If it’s in the mushy middle, it doesn’t. You can do clever marketing, clever messaging, clever positioning, but the slope, the water’s not falling far enough, if that makes sense. In building, you have to be very careful that you do some quantifying of the zone of [00:37:00] benefit in addition to fight being in the right quadrant. If you are in the right quadrant, lemme give you an example. We built HipChat, right? Was created, and Atlassian being like an edgy company, understood that this is the new outlook.
And they bought HipChat and sold it to, I don’t know, maybe a million people or whatever, decent business. And then Slack came along with great marketing. Maybe I give a better marketing and took all the shine. And actually I was hired to build HipChat V two, right? We called Stride and we started to imagine like visionary, like what’s the next generation of this?
The core problem we were dealing with was how much better is this than Slack, right? And so all our effort was in creating a zone of benefit distinction between the two. The thing that upended that was Microsoft jumped in with teams, [00:38:00] and I knew that Microsoft was going to spend billions of dollars to win that fight.
You know why? Microsoft understood the benefits of being in that quadrant. They’ve been in that quadrant for 30 years. They know they have to dominate that quadrant. In order to stay for the next 30 years. So they were gonna spend whatever Microsoft was ready for teams to be the next outlook, and they were gonna spend whatever they needed.
And the calculus changed dramatically. I still think Atlassian Stride was better than the other two offerings, but what we came against were business issues, which is how do we differentiate, how do we compete against not just Microsoft teams of product, but Microsoft sales team that’s selling it for marginal cost of me, zero in all the office accounts.
By the way, I think this is interesting for your marketing people. ’cause like sometimes the product won’t save, but there’s not enough zone of benefit. Differentiation shall be one of the [00:39:00] next generation plus all those builtin advantages for our competitors for us to endure basically. And so we ended up just like selling the IP and the software to Slack.
But that’s one practical place where zone of benefit mattered a ton. To thinking about what to build, how to build it, and whether to endure in the marketplace, if that makes sense.
Sophie Buonassisi: That’s a fantastic story and I know you’ve just got tons of great stories I’d love to hear one that maybe you haven’t widely shared as much.
Oji Udezue: this is a little bit of an old story, but I think it’s illustrative and sometimes I like to tell old stories so that people understand that there’s nothing new under the sun.
Even in the age of ai, there are stable undercurrents in this industry, that you have to pay attention to. ’cause everyone pays attention to all the flashy stuff. So here’s the story. It is 2000 and, I don’t know, two, 2000, I don’t know. Some, at some point Microsoft is working on Longhorn, which [00:40:00] is what became Windows Vista.
for the last 10 years, bill Gates has been talking about building an operating system that’s object oriented. Object orientation is just a way to build less, at least it was circuit 20 years ago. he was tired of file systems, which were just like in the, if computer science that just point us into memory and very primitive.
And he wanted to build an object oriented operating system. He wanted to elevate what windows would become. By the way, that still hasn’t happened because it’s incredibly hard to do. Even Apple isn’t an object operating system at this time. And so the next version of Windows were multiple new bets.
It was an object oriented operating system kernel built on top of an object-oriented file system with a little bit of an object-oriented shell. The thing you see, right? ’cause all the arrest is just like behind the scenes. And so three gigantic divisions of Microsoft geared up to build this new thing that Bill had asked us to do.
Long story [00:41:00] short, it was a complete failure. by the time we canceled that version of Longhorn, which the world never saw, it was 10 times slower than Windows xp, right? It just was a dog, basically. And in the end, we took Windows XP as a base and freshened it up, right? Built in the most stable technologies we’d invented over the last three to four years and released it as Vista.
Now, because Microsoft is in a niche, which is everyone, workflows and everyone still needed operating systems, Microsoft did not die, right? If Microsoft, this would’ve been akin to Nokia versus iPhone, if. Software was more brittle. You know, Hardware is way more brittle as a business. So what did I learn from that?
I thought about this for years, but one thing you can take away from it, at least for the [00:42:00] people who are building, is if you’re doing something completely new, try to do one new thing at a time. Don’t concatenate your risks so you’re building something the world has never seen. Sure. And then you build on a new API that’s not tested.
A new operating system does not tested a new cloud that’s not tested. Or even new people who don’t know what they’re doing, they’ve never done something hard before. So try to take on only one revolutionary unpro thing at once. Don’t join them up. Right now in building a company, we are calling Fabric, which I’m building right now.
One of the things that happened is the release of model control, MCP service. And we sat down and started to think, what is this? Is it important? Should we change your strategy? Should we take a hard dependency on cps? And by the [00:43:00] way, I’m visionary. I always wanna work on new thing. This lesson though, prevents me from being completely driven by that kind of excitement because I’ve seen how it ended.
And so one of the things I said is, if CPS are the right thing for us, we will use it. But if we have something critical to do and MCPS are too risky, we will not do it. Because in the end, what we want is data access. And this is a data access of MCP. What we want is data access to data, to customer data.
You can say do, there’s a MCP server in front of Gmail and that’s a new thing. But there’s also an old rest, API that’s been around for 15 years that people currently was built on that API, right? It works. It’s been debugged for 10 years, right? The MCP stuff is new. Do not put all your chips in that basket because if you do, let’s [00:44:00] imagine in the future the MCP protocol is found to have a gigantic security hole.
There goes your business or six months of development. And so this is a lesson baffled worn scar tissue lesson is on the infrastructure layer. A business is technology product and the business at the technology layer. try to be fundamental skate to the future, but only take one risk at a time if you can.
and that’s not a hard rule. Maybe you can take more, but be conservative. add that layer because if you really wanna build things that are solid, that people love, that don’t go down, that are tried and tested doing too many of those new things will hurt you eventually.
Sophie Buonassisi: That’s great advice, especially right now is there’s just no shortage of kinda shiny objects, if you will, or exciting things to build.
Oji Udezue: Correct?
Sophie Buonassisi: And Oji. My next question is actually from our founder community when should a startup hire its first product manager, beyond the founders?
What [00:45:00] qualities would you advise looking for in that first hire and that person setting that foundation for a product driven culture?
Oji Udezue: When should you hire your first product manager? I wanna make some assumptions. Product is something that someone is always doing, whether or not they have a title in a standard cookie cutter startup, the founder or the founders are the product managers. They have a vision for the product. And so they are product managers even though they’re the founder.
And what typically happens because, you know, we invest in some of these, is they’re the founder. They work with developers and all kinds of consultants and investors to help achieve that vision. And as they succeed, if they don’t succeed it doesn’t matter. But as they succeed, their time, the ability to devote to listening to what customers want goes way down, right?
you have to literally carve out time to inhabit the mind of a customer. And in the meantime. There’s fundraising to do, [00:46:00] there is go to market to do, you’re the chief salesperson and you just don’t have the time. And if you take on that role forever, it will suffer. So lesson one, someone is always doing product whether or not they have the time.
But the trick is to give that function, not a person. Oxygen, as much oxygen as possible. Austin, I’m gonna make an asymmetrical example. Austin is a great tech city, maybe number six, behind la, the Bay, Seattle maybe New York. Austin’s a pretty decent top six minimum. The thing that I find lacking here are great product people.
I went to a product meetup next week, so they’re gonna hate me, but, but I just think that there needs to be better product people here because what happens is developer focused founders start [00:47:00] gets on to sell it. They find some value, maybe then the right quadrant, they have a sharp problem, and then they can’t evolve it because no one is, has the product sense to figure out what else should we build, what’s going to add to the value, right?
In a way that is fundamental, if that makes sense. So the function needs oxygen for a startup to continue. The minute it doesn’t have enough oxygen is the minute you hire dedicated perk manager, hopefully with enough scar tissue that they don’t just, plod along and they’re not taking risks and so on and so forth, which can happen if.
Because, the founder has the most grit usually. But yes, if you start noticing low oxygen for product management, whatever the title, hire your first product manager. Now in practice, that means as soon as possible, really, right? Product is a [00:48:00] weird multiplier of return on capital. It is the thing that makes the product.
Without a product, you have no business. Your salespeople can sell for as long as they can and delay the inevitable, but if you don’t evolve a product, you won’t have a company for very long. So it’s very important. Do it as early as you can.
Sophie Buonassisi: Great advice. And Oji, this brings us to our last two questions always the same. What is one tactic or strategy that’s working for you or the companies that you are serving, advising, or in your case building?
Oji Udezue: In 20 25, 1 tactic, which is old, that is working is customer discovery and listening. Spend time with customers. Don’t just go off in a hole and build, spend time building community people who will tell you if what you’re building is valuable. I think this was valuable in, and it’s always valuable in building as a core tactic.
People [00:49:00] express it in different ways. Some people build in the open da, but regardless, you gotta have feedback loops as early as possible when it’s uncomfortable. The second one in the age of AI is, oh my god, so much curiosity. imagine if the river just sped up 10 XA hundred x and in the river are shiny objects, you need to build your next company.
If you’re not paying attention, trying to sort. What is hype, what is real? You might miss something. And so in our creative Slack channel, we have like channels dedicated to discussing tools that we have found that we need to build in this new generation. So it’s what are the tools to speed us up and the practices to speed us up, which I’ve talked about.
But on the other side is what is happening with customer taste on the market that is changing that we can’t see? And so we spent probably a more inordinate amount of time [00:50:00] on that than I did two, three years ago just because of the uncertainty. So it’s my way of managing the pace change and the uncertainty that people are seeing is those two things.
So one is perennial, it’s a thing that we always forget, do it. The other one is an artifact of the world changing. I might say when I change it, I mean, transformer based AI is going to transform how we all do this and how we all invest in this and et cetera.
Sophie Buonassisi: And they’re almost linked in a way the customer obsession feeds the curiosity and it makes sense. ’cause I think the CEO of Calendly described your book as a practical playbook for building customer obsessed products to achieve exponential growth. So I can see that actually baked in. Now last question, Oji.
What is one widely held belief that revenue leaders have that you think is bullshit or no longer serving us?
Oji Udezue: This is a little bit of a. An inversion of the sharp problem hypothesis, meaning that you should [00:51:00] always select the sharpest problem. And, revenue leaders in some cases overestimate
their impact on the trajectory of revenue growth. Now, I have to say, cleverly done sales and cleverly done Marketing is an accelerator, right? It is because it’s not just about, you know, the right marketing campaign, the right sales launch in a different company, et cetera can make a huge difference.
But at the same time, one huge discriminating factor is, do you have a good thing to solve? Is it fundamental to your target customer? And I think that’s what we’re trying to say. This is by far one of the biggest dimensions of success. So one way revenue leaders can help isn’t just in being clever at their job or clever at their campaigns or clever [00:52:00] at their channels.
The thing that one of the big things they can do is contribute, honestly, not in a fake way to understanding, helping the team understand and build the right thing, right? People who turf fight so much about here’s my budget for this campaign. Here is the thing that I do that I’m great at, and I did at the last company, and I don’t focus hard enough on.
What is the insights? How do I get to the table to really satisfy customer needs in the core of this? Because if I can do that, my job becomes easier. And in some cases, I don’t have to do very much, if that makes sense. so that’s one thing. Invert yourself. There’s a lot you do outside.
That’s you. You’re a marketing person. You’re shiny and bright, you’re a go-getter. You’re an extrovert. Think about how you can send signals that help build a better product. Your life [00:53:00] becomes easier. And so I think that’s one big thing. It’s not just about your campaigns and how clever all those things are.
It’s about how you are feeding back into the engine of returning capital.
Sophie Buonassisi: Yeah, establish and kinda run the feedback through those feedback loops themselves,
Oji Udezue: Yeah, just get into it like it is, you can throw it over the fence. So this is what customer said, this is the, what the tool said, but get into it, work with people like me, to really figure it out. Lemme give you an example because maybe that makes it more, when I got to Typeform, we were selling to everybody basically.
Everyone needs a form in some way, but I wasn’t interested in that because I knew that within that everybody, there were small, smaller groups of people who, for was mission critical. I was looking for the people for whom it was high frequency, mission critical. I knew that those are the people who were converting [00:54:00] faster and all those things.
And so we needed a definition of customers narrower than what we had. And so that definition was important to build a better product for them and not build it for everybody. I wasn’t interested in building for everybody anymore, and so to the extent that the marketing team inverted and help me define that right customer so I could understand them more specifically, not everybody, and build the right thing that led to more revenue than just marketing to everybody.
Sophie Buonassisi: I love it. I love it. I’m gonna sneak a last question in here if that’s okay with you. I’m curious from the product lens, with the rise of ai. What’s your perspective on what’s that, what that is doing to go to market overall?
Oji Udezue: We are in an era where [00:55:00] just a lot more magic is happening. There’s a lot of noise in the market. Two years ago, just slapping AI on your product did something. I don’t know that it does anything right now, if that makes sense. People are going back to the fundamentals, like, how much acceleration does this do for me?
So some of those things that are basic and, AI is part of the story, but it’s not the whole story. and we are about to see, like I said, accelerated development. So if we are already at a point where. For our lower tier customers, we ship a lot of stuff, right? We’re like, take it for our enterprise customers.
We’re like, okay, here, take it once a quarter in one nice chunk, right? Because your users can’t absorb the whole thing. Surely if there are errors, you do that and we don’t want take on the support costs and we are selling you an enterprise license. Now imagine that on steroids, right? New things all the time.
In the old world, product [00:56:00] marketing was like, holy shit. Okay, we’re gonna do a newsletter every month, every quarter to describe all the things that have changed. And those weren’t very effective anyway, but they were part of the product marketing cadence. What is the next cadence when this is 10 x?
What do you say? How do you attract attention? How do you even describe the new value you made? In a way that people pay attention to. Remember, we’re this, we’re designing and marketing to distracted brains more so than ever. And so AI is going to weirdly reward marketers who try to figure this out. How this pace change affects how to attract attention.
What, what is marketing? Fundamentally it is draw attention to something amazing. Generate demand for it, right? We can talk to the cows, [00:57:00] come home about all the different departments, content marketing, da. But that’s it. And so I think it is weirdly gonna be harder in the next couple years because people are gonna be talking a lot about things that don’t matter very much to draw attention.
And so people who crack the code of. Value marketing in a noisy environment, the professionals that worth their weight in gold and people who do it will succeed much, much better.
Sophie Buonassisi: There we have it. What a way to end it. That’s fantastic. Thank you, Oji. This has been fantastic. Really enjoyed the conversation. For anyone looking for Oji’s book, we’ll pop it into the show notes. Oji, where else can people find you?
Oji Udezue: They can find me on Substack, where I try to write about product leadership, product management, and the future. That’s rolling up on all of us around AI, product, AI startups, AI pro mar marketing as [00:58:00] well. So Substack is a good place. You can go to product mind Co to see all the stuff that a I do to help companies succeed.
And who knows, maybe we can help you succeed as well. So those are the two places.
Sophie Buonassisi: Wonderful. Thank you for joining us today. Really appreciate the time and we’ll catch you next week. Thank you to all the listeners for hanging with us.
