a16z Podcast - Grant Lee: Building Gamma’s AI Presentation Company to 100 Million Users

Episode Date: November 11, 2025

Grant Lee was told Gamma was "the worst idea ever heard" by an investor who hung up mid-Zoom—yet he built it to 100 million users and $100M ARR without spending a dollar on advertising.While competitors hired aggressively, Grant's team of seven refused to grow, dedicating 25% of their tiny team to design and personally onboarding every influencer themselves. They reveal how ignoring AI for their first two years, then orchestrating multiple models in ways the frontier labs can't replicate, let them steal the presentation market from Microsoft and Google—going from 60,000 signups in eight months to 50,000 per day. Resources:Follow Grant on X: https://x.com/thisisgrantleeFollow Sarah on X: https://x.com/sarahdingwangFollow Olivia on X: https://x.com/omooretweets Stay Updated: If you enjoyed this episode, be sure to like, subscribe, and share with your friends!Find a16z on X: https://x.com/a16zFind a16z on LinkedIn: https://www.linkedin.com/company/a16zListen to the a16z Podcast on Spotify: https://open.spotify.com/show/5bC65RDvs3oxnLyqqvkUYXListen to the a16z Podcast on Apple Podcasts: https://podcasts.apple.com/us/podcast/a16z-podcast/id842818711Follow our host: https://x.com/eriktorenbergPlease note that the content here is for informational purposes only; should NOT be taken as legal, business, tax, or investment advice or be used to evaluate any investment or security; and is not directed at any investors or potential investors in any a16z fund. a16z and its affiliates may maintain investments in the companies discussed. For more details please see a16z.com/disclosures. Stay Updated:Find a16z on XFind a16z on LinkedInListen to the a16z Podcast on SpotifyListen to the a16z Podcast on Apple PodcastsFollow our host: https://twitter.com/eriktorenberg Please note that the content here is for informational purposes only; should NOT be taken as legal, business, tax, or investment advice or be used to evaluate any investment or security; and is not directed at any investors or potential investors in any a16z fund. a16z and its affiliates may maintain investments in the companies discussed. For more details please see a16z.com/disclosures. Hosted by Simplecast, an AdsWizz company. See pcm.adswizz.com for information about our collection and use of personal data for advertising.

I get into my third pitch, and it goes pretty well, I feel like, and I kind of pause and wait for the, you know, investors' reaction. He basically says, this has got to be the worst idea I've ever heard. Not only are you going up against massive incumbents, they have ultimate distribution. There is no way you're going to compete against them. Before I could say anything, he hangs up. He just drops off the Zoom call.

I kind of gave myself just a chance to internalize what he was saying. Try to, like, pull out some of the truth in that, which is, yeah, going into this market is going to be incredibly hard. We need to think about growth from the very beginning. And if we have any chance of succeeding, product, growth, those are going to be intertwined in our strategy. Building a presentation shouldn't feel like wrestling with formatting. It should feel like thinking out loud. Today, you'll hear from Grant Lee, co-founder and CEO of Gamma,

on building one of AI's breakout success stories, from a brutal pitch rejection to 100 million users and profitability in record time. We discussed why better to be different than better drove them to break the 16 by 9 prison that PowerPoint built 40 years ago, how they went from 60,000 signups in eight months to 50,000 per day after nailing the first 30 seconds of user experience, and why hiring painfully slowly while personally onboarding every influencer created their word-of-mouthed flywheel. Plus, the strategic sequencing that let them build prosumer love before attacking enterprise, and how taste isn't just visual, it's the entire restaurant experience.

Let's get into it. Thank you for coming, Grant. Thanks for having me. You co-founded in our leading one of the most successful AI applications right now. You've crossed 100 million of ARR. You're approaching 100 million users. We are so stoked to talk to you about

all of your recent product launches, expanding in the enterprise, and maybe taking a step back just the process of building Gamma itself as a company. Maybe to just kick that off on that note, we'll rewind time and bring it back five years ago when you and your co-founders started Gamma in 2020. There's obviously been a ton of twists and turns since then,

and it feels like we've been in maybe three macro cycles, two technology cycles. A lot has changed. A lot has changed, exactly. Can you share a little bit more about what that process has been like and what were some of the big unlocks along the way? Totally. It's always fun to kind of tell the initial sort of fundraising story because the world was just such a different place where, you know, peak pandemic.

This is the end of 2020 where most of, you know, fundraising was just done over Zoom. So a lot of the investors I was trying to pitch early on, you know, I never met in person. I was actually living in London at the time. So different part of the world, trying to line up these pitches. I'd have to do so in the middle of the night. I also have two little kids. And so for me, it was, you know, after kids go to sleep, we had a pretty modest flat. So I would kind of find out little corners actually tucked between the kitchenette and the laundry room.

little space, put up a fake, you know, Zoom background so nobody could tell where I was and try to talk somewhat softly so that, you know, don't wake up the kids. And so starting the pitch, I was like, okay, first week, it was just lining back-to-back pitches. And the first day, I get into my third pitch and it goes pretty well, I feel like, you know, going through 20 minutes in, feeling like, hey, this is not too bad. And I kind of pause and wait for the, you know, investor's reaction. And he basically says, this has got to be the worst idea I've ever heard. And he said, not only are you going up against massive incumbents, they have ultimate distribution. There is no way you're going to compete against them. And so in my head, I was like, well,

that wasn't what I was expecting. And I got to think of a really good rebuttal. And before I could say anything, he hangs up. He just dropped off the Zoom call. And so. So I'm like, wow, okay, you know, this feeling of really this moment you're kind of thinking your head, you know, what if he's right? You know, what if he's right about all this? Maybe this is a dumb idea. And so you kind of have those doubts kind of creeping up into your head. I didn't really have much time to kind of get down to myself because I had to prep for

the next pitch. And so in that moment, I kind of gave myself just a chance to internalize what he was saying. Try to like pull out some of the truth in that, which is, yeah, going into this market is going to be incredibly hard. We need to think about growth from the very beginning. And if we have any chance of succeeding, like product, growth, those are going to be intertwined in our strategy. And so kind of took that to heart, even though it's painful to hear. And certainly, maybe there could have been a kinder way to say those words. I still went back and eventually, you know, fortunately, we were able to raise money five years in. We're making some good progress. Well, I'll say, good progress. Yeah.

That's amazing. And it was pre-AI when you started gamma. Is that right? Pre-A.I, I think the spirit of what we want to do all along was to make it easy, effortless for anybody to create content. So the nature of, like, being fast and the speed aspect was always there. But AI happened to be a great gift a few years into the journey that really just expedited the entire process there. Yeah. One of the things I've been curious about, because 100 million users is like a lot of users. And some of them might know and love AI tools and some of them might not. As AI has evolved, how do you think about how much AI you expose? in the product, like even giving users the choice to pick their own image model. Like, how easy or complex do you make that? Totally, yeah.

And it's a moving target. You know, when I was at Sanford, I was lucky enough to have a mentor, David Kelly, who was the founder of IDO. And his story, early on, they invented the first Apple mouse. And so, like, at the time, like, you know, Xerox already had a mouse. It was a three-button mouse, very, very expensive, wasn't built for mass consumption. And so for someone that's super technical, they knew how to navigate. Kind of sounds silly now.

even three buttons kind of intimidating. And so David Kelly's approach was, oh, is you need to be super empathetic for where users are. And so they actually invented, you know, the one button mouse, super easy. Anybody can click one button. And I think it's going to change over time. So I think we're right now maybe in the one button era of like AI where the vast majority of people, like, they're still trying to figure out how do we even use this thing. It's powerful but intimidating. And so we need to recognize that.

I think it's easy to kind of throw shade at, oh, a chat prompt box is not the ultimate form factor. That may or may not be true, but chat is also very accessible and familiar. And so I think we're trying to navigate that ourselves. Like, what is the right entry point? A lot of that is spending time observing users, how they use gamma, and then seeing where they just run into a wall or fall really short of expectations and try to fix things along the way. Yeah.

I think one thing that surprised us in a sense was just how, not only how hard it is to get AI presentations right, but also how different the products are on the market. Olivia and I have literally. tried 20 plus players out there. And the results are night and day. Can you share more about what is so hard about it and what you guys got right? Yeah. I mean, I think, you know, there's been multiple attempts just throughout the past decade, even pre-AI, right? There's a graveyard of presentation tools out there. I think the ones that are pre-AI kind of got stuck in a different

trap, which is they tried to be incrementally better slideware. So they still do the 16 by 9 thing. And you still have this sort of canvas where you as then user have to figure out the layouts, the layers, resizing things. And for the average person, even if you're kind of 10x better on one dimension, PowerPoint's had 40 years to add every single piece of functionality possible. And so how do you ultimately compete? And of course, yeah, they do have distribution. And so it becomes pretty hard. I think for us, like early on, we knew that to compete in this category, you know, the saying is it's better to be different than better. Like, it's better if you're different than better. And so for us, we wanted to be fundamentally different.

What are the different set of primitives or building blocks that aren't 16 by 9 slides that allows the average person that doesn't have design skills, that doesn't have design resources to build to create content super fast? So that was kind of the initial differentiation. I think today there's almost a separate trap, which is being an AI native startup, you can lean a lot on just the AI models and assume they're getting it better and better, which they will. But then that almost becomes a crutch because you're trying to anticipate, well, I don't want to sort of create something that the model is going to disrupt everything I've created. And so you're not really willing to understand, okay, how do I create something in between that gives people a starting point that still has human in the loop? And it's pretty challenging. I think for us, we were sort of lucky in the timing in that we started pre-AI. We had a thesis around

what building blocks would be important. And then we married that with AI along the way so that we kind of get the best of both worlds. Yeah, that's fascinating. And I think maybe just to pull on that thread a little bit, you know, hey, we're coming up against Microsoft, Google, the biggest tech juggernauts in the world. But on the flip side, right, you might hear from others that, hey, is this just going to be solved by the frontier models? As the models get better, you can one shot a great AI presentation. Talk more about why that reality is also maybe one that you're not afraid of. And beyond sort of a single model itself, like what kind of value gamma is adding on top? Yeah. I mean, the way we see it is for certain tasks,

maybe a generalist approach or a super app can probably do a lot of these things. And for one-off presentations, that might be actually a viable solution for that case. I think for any time where the job to be done is really core to what the end user is trying to do. So in this case, like visual communication, business communication, visual storytelling, there's a lot that happens to make that feel right. And you as a creator need to feel like you have a lot of input. You want to be involved in that because it is your story you're telling. It's not the AI story.

And so when you start going really, really deep into what it means to be, you know, great visual storytelling tool, like that job to be done, there's a lot of different workflows for us, even orchestrating all the different models, whether it's, you know, on the text side, image side, video, audio, there's a lot going on there. And so how do you package that all up so that it feels good for the end user to be able to still play with this, edit, manipulate all those things? And we just think that, you know, that requires a really specific tool to build to do that. Even as you think about the kind of the evolution of that for us is like, you know,

go from single-player mode to multiplayer mode, the ability to collaborate, have a shared workspace, have like a central repository of things that you're pulling into content, those are also difficult problems to solve. And certainly you can do it, but for a super app to do that, it's like then you're starting to pull away from what maybe their long-term aspiration is, like they do want to be the all-purpose app. And that means they have to sacrifice a lot of these sort of granular point solutions that may not fit with their long-term vision. Yeah. It's interesting because a lot of what's magical or, least some of what's magical about Gamma comes from AI, like the new Agentic 3.0 launch is amazing and

the editing capabilities there. And then some of it just seems to be new insights that your team had around how presentations and communications should work. Like what was the process that led you to doing that differently? Yeah. So part of this was, you know, when we first started Gamma, you know, the inception idea or like how we came about the initial idea was I was working at a different startup before we were acquired. And I went back into my roots, which is consulting. I was advising early stage startups. And what I was using to sort of create presentations was, you know, Google slides. So I'd go in, you know, the really one of the moments was like late at night trying to prepare for a meeting the next day and just feeling

like I was spending all this time trying to format the deck to look decent so that it didn't look like I didn't do any work. I had this moment feeling like, you know, there's got to be a better way. This feels so backwards. Like, why am I spending 90% of the time on the formatting and 10% time on the content. And so when we kind of took a step back just to think about, okay, you know, again, PowerPoint invented almost 40 years ago, what has changed? What are some of the things that we can sort of anticipate, continue to change? One is, you know, the way we share information obviously changed, you know, peak pandemic, of course, so much work shifted to virtual or like, you know, over Zoom. And so, you know, what if instead of having content that was, you know, static 16 by

nine, it was much more malleable. It could be something that could be mobile responsive by default. So if I shared something with you, you open on your phone, you could scroll through it just like you would, you know, a website. And that same content can kind of flow naturally there. We also thought, you know, why instead of just having, you know, most presentations, just text and images, what if it could be more multimedia rich? So you could literally embed an entire website into what you're presenting because, you know, that's a way for you to have much more richness than what you're actually trying to showcase. And then, of course, like, rather than, you know, a standard slide deck where I'm saying,

hey, next slide, please, and clicking on it, what if it could be interactive? You know, what if it could be something where everything you click on it can actually open up? Just like the principles that we've embraced with Web, like, you know, the whole notion of, you know, progressive disclosure, like that should apply to presentations as well. As I can either, if you're reading an async,

you can choose your own adventure, or if I'm presenting it to you, I can be the storyteller and actually guide you through this. And we felt like those are all things that, you know, seem obvious or maybe we've already taken granted, for granted on the website, you know, side of creation and why not be able to offer that same set of things for the average person, again, doesn't have design skills, doesn't have technical resources, maybe doesn't know how to code and just empower them to do way more than they could

otherwise. What do you see? I mean, I love this evolution of what's been unlocked on Gamma. And if you think about what's next, even we're seeing interesting things being done with avatars, often in a sales context. It's still very early innings. But, how do you think about what's most interesting and what's kind of coming up next for Gamma? Totally, yeah. You know, one of the things we try to focus on are universal pain points. So when it comes to just even the foundation for Gamma, the insight was, you know, very, you know, we're all visual learners, but very few of us are visual designers by background, right?

And so how can we empower people to be good visual storytellers? I think within this sort of space, you know, there's another skill set that a lot of people have fear about, which is presenting, and they're not maybe super confident going up into a room and, like, you know, be able to articulate their thoughts. And I feel like that's like a huge, you know, if people can't do that well, then that just means their ideas never make it to, you know, into the world. And I feel like that's, that's a, you know, that's really a disappointing outcome if that's the case. And so, of course, yeah, the whole notion of maybe AI avatars or like, you know, a presentation that can speak or, you know, articulate your thoughts on your behalf using your voice. I think those are all

super exciting explorations because, again, it goes back to some of these universal pain points are like, you know, things that people aren't inherently great at that we can help assist them along the way. The presentation is kind of like a final unit of output is downstream of like a lot of work and also a lot of thinking, usually, if it's a good presentation. How do you think about how far gamma goes in the pre-presentation space? Like how much do you do to help users get from like the initial grain of an idea into like a pretty clear articulation of a thesis. Yeah, it's a great, great question. For us, we want to be end to end, right? The moment someone has some inkling of an idea that they want to share, we want to be there for them to help structure

that, help maybe even create that initial narrative or story arc. And then the moment they want to create that first draft, we want to be there to help them create that outline and be able to actually put something where you're putting some shape around it. And then when you do want a design partner to help you edit and really refine the thinking. We want, you know, that our agent to build a review everything you've already created and offer suggestions, maybe even make the design changes for you. And then, of course, the moment you're actually sharing and presenting to others, we want to, you to have a feedback loop. Like, what are people engaging with? Like, what content actually resonates? What are people getting stuck on so that you can go back and refine the entire

process? And so, yeah, we do hope, you know, today I think we do decently well, sort of in the middle. You have maybe the initial first draft done. You have some of the editing capabilities. I do think we have an opportunity to really close the loop entirely. And to your point, even like the very, very first part of the journey, I think there's a ton more we could do there. Yeah. Yeah, totally. And I know Olivia touched on the Gamma 3.0 launch, which was, you know, a few weeks ago now. Very exciting launch, very successful. And within that, you had agent, teams, API. There's a lot coming out in one day. A big launch day. And, you know, I think when we use agent, it feels like having a design partner alongside you. I'm curious. I know it's early days, but

How has it changed the way your users interact with, with Gamma, for the ones that have tried it, at least? Yeah, totally. I'll maybe quickly just share one perspective, which is, you know, I think, you know, we launched 3.0. I think it's helpful just to think about, like, you know, what is the difference between, like, a 1.0, 2.0 and a 3.0 launch. For me, like, the 1.0 launch is all about really going after the true sort of innovators or the people that are really willing to, like, try whatever's out. They love the new things, right? And you can call them AI terrorists, but they're basically going out there because they love playing around and testing all these tools. Many of them may not stick around, and I think that's okay, but you're trying to get some of this initial energy around, like, what's interesting, what's not, what's actually useful.

And, of course, you have then to go back and actually continue to improve around the product. When you get to your 2.0 launch, you finally have a chance to kind of win over what you consider as like the early adopters. People that still are, you know, willing to kind of put up with a lot of gaps or, you know, deficiencies or, you know, product but for the first time they're starting to pull product out of you they have a whole long list of like feature requests they see it kind of working for certain use cases and that's great like you take that feedback back and you start you know actually flushing out your roadmap even further and then you finally get to your 3.0 launch which i think is like you know if you were to map it to crossing the chasm like you're for the first time trying to hit the mass market you have a chance

to have a product and can actually be used and be relied on and people can trust it and i think that's where we're at today. And so for us, you know, there's a few different things, you know, we're releasing. One is just this notion that we can actually serve businesses of all sizes now, right? I think we're going from the sort of single player to multiplayer mode, going from just being a pro-sumer tool to being a business application. And I think we've had to earn the right to build and do so. And, you know, I think that just, there's a ton more that, you know, for a business use case where everything from, you know, even your terms of service to security, privacy to, you know, can I actually collaborate in real time with my colleagues and coworkers.

These are all pretty, you know, meaningful constraints for us to actually build, like actually deliver on. And so we've worked a lot to actually be able to, you know, fulfill on that. And as part of that, we're also kind of expanding the aperture of what Gamma even represents. We're also, you know, releasing an API business where I do think for the first time, you know, in addition to just serving end users, we can serve businesses and developers that want to build on top of Gamma. So for us, it's just like an exciting moment in time where I think for a business that has, you know, up until now been pretty straightforward, like pretty, you know, easy to understand. I think we'll be able to add sort of complexity in the sense that, like, we're going to have many different ways we can serve many different types of users.

Yeah, absolutely. And I want to actually touch on the API piece of it because I feel like whenever there's the word agent thrown in, that gets all the attention. But actually, this harkens back to a CRO dinner that we had where a lot of CROs talked about using Gamma for sales decks. And if you think about the ability to connect your data stores, maybe it's a CRM, et cetera, to communicating what's in them and sort of the ideas and getting that to action, that's a very, very powerful concept.

What are some examples of loops like that that you've seen among your users and sort of inspired the API product? Yeah, totally. So a lot of the early requests were, you know, just building on the foundation of our prosumer user base. You know, prosumers, many of them are using, already using, you know, tools like Zapier, make, and now we can connect, you know, the tools are already using. So let's say you're using granola for, you know, all of your notes.

You're meeting, you know, if you're someone that uses gamma, let's say you're using it as a way to communicate to your clients and partners. And, you're sitting with a client, the moment the meeting's over, you can feed everything from Granola into basically creating a beautiful presentation that recaps all of the discussion topics, recaps all the action items, the deliverables, and all the next steps, right? And so as soon as the meetings end, you know, it's over. In their inbox, the client has that beautiful presentation, the beautiful PDF, and they know exactly, you know, basically, you know, what happened during the meeting and what was agreed upon. And they feel good because they feel like you listened and

are able to really fulfill being a good partner for them. On the sort of, you know, go to market side, we see, to your point, you know, being able to integrate with CRM or, you know, in some cases, you know, people are using, you know, sales intelligence tools like Endgame, where, you know, a lot of that information is there that can be pulled into a presentation. So if you want to personalize, you know, a market research or like a customer research deck for a specific client or customer, you can also do that on a super automated basis. and we think that's going to be super powerful. And then I think on the B-to-B side, there's actually two different use cases that we're super excited about. One is being able to partner with someone like Glean

who sits on top of company knowledge, first-party data, where internally, you know, we can be their visual storytelling layer. So you have to prepare for a QBR or a strategy session, you know, pull all the information you already have and create that deck so everybody can be on the same page. We think that's going to be also just like, you know, who wants to spend time formatting the PowerPoint

point when you can just have that automatically created in Gamma. And then I think on the developer side, this is where we're already starting to talk to, you know, folks like someone building a real estate app. They want their end users to build, you know, when they go into their product and search a specific zip code or region, you know, they want to be able to give their users, you know, a directory or listing of all the available homes in that market. And they can have now a beautiful branded PDF sent to that end user. And they don't want to have to build that, right?

we can be their content infrastructure in a way that's very complementary to their actual real estate app and provide them sort of, you know, the under the hood, all this sort of goodness that, again, they can have this, you know, now this content engine running for them, all built on top of Gamma. And as Gamma gets better, even on the prosumer side, that product gets better, everything else gets better too, right? The API becomes more powerful, more effective, and it could then start even more use cases. It's amazing. Hearing you walk through like the 1.0, 2.0, 3.0 launch, your team has always been, been incredibly, I think, clear and intuitive product builders. But I'm sure with tens and now

in the hundreds of millions of users, you get a lot of requests. Like, how do you think about what do you build that people are clamoring for? What do you not build? Like, how do you make those tradeoffs? Yeah, it's definitely becoming increasingly hard. So it's always balancing your sort of own intuition for where you think there's big gaps or opportunities in the market. We also have just many different sources where we try to, you know, obviously tried to just have a good pulse and then try to consolidate all that. So one is like we have a public candy board where anybody can submit future ideas, requests, people can upvote. We have for our power users a separate sort of Slack group. We call it the Gambassiter Program. So we throw them all there.

It's a great place for me to get real-time feedback because oftentimes we're kind of showcasing early concepts, wireframes, whatever it may be and saying, hey, is this even like, you know, is this completely missing the mark, or is there something interesting here? And so you get some really critical and honest feedback from there. And then we're extreme, like, dog fooders for our own product. And so, you know, every week we have something called gamma-rama where anybody at the company can create a fun presentation on any topic, like educate us on something.

It could be a hobby. It could be something you really believe in. And at the end of every gamarama, you talk about all the sort of pain points you have, all the shortcomings of the product. So, you know, all to say is like multiple different sort of inputs into that. And then we try on a very regular basis to kind of stack rate where we think is going to be most important and try to push that into the road not. I actually want to switch gears a little bit to growth itself. And the crazy thing about approaching 100 million users is not only the sheer number, but the fact that you've done through, you've done so predominantly through organic channels.

And so we'd love for the, I'm sure. a lot of folks out there listening would be interested in how the heck did you guys do that? What channels have been the most effective for you? And I do have to throw in there. I don't know if you're going to say this channel, but I've been very impressed by your Instagram content. I'm a millennial and I'm on Instagram all the time and I engage with your content. And occasionally when I go to like it, I'll notice it has 80,000 likes already. And so very curious how this ballooned, expanded and how you guys bring the community into the effort. Totally. I mean, I'll start by just, you know, my piece of advice for most early stage founders, you know, as they're

approaching or maybe, you know, getting the product market fit, the number one thing I always advise them to focus on is word of mouth. Like, you have to build a product that has strong word of mouth. Everything else becomes so much easier when that's the case. And don't fool yourself into thinking of product market fit until you have strong word of mouth. And so we really did try to get that right in the very beginning. You know, we had a few different sort of launch moments early on where, you know, we launched on product hunt and we had ended up winning, you know, product of the day, product of the week, product of the month, and it felt good.

And then we looked at the signups and initially you see this huge spike in signups. And then there's sort of plateau out. And we're still bringing in new users, but it was very clear we didn't have strong organic virality. Like something wasn't working on the sort of fundamental level. And so we had to go back and say, okay, you know, at that point in time, we could have said, okay, maybe let's just throw much more money at top of the funnel, like advertise a whole bunch, like run massive influencer campaigns. And I think we would have fooled ourselves into thinking, you know, we could just continue to grow and that would be fine. We'll just spend our way out of the problem. But I think

we gave ourselves an honest look. It's like, no, that's the core growth engine is broken. So before we do any of that, let's make sure we do have strong word of mouth. So we ended up spending, you know, three to four months leading up to our first AI launch, the entire team just focused on one thing. We are going to nail the first 30 seconds of the product experience. We're going to make sure that every new user coming in has the sort of a chance of like seeing the aha moment such that they will go on and, you know, tell their friends, colleagues, family about this product. And I think the moment we launched, you know, all of a sudden you see, you know, what took us basically eight months to get to 60,000 signups, within a few days, we surpassed that.

And then we'd have like, you know, 25,000 signups a day, 45,000 signups a day, 50,000 signups a day, it just kept them going up, right? And we weren't doing any advertising, we weren't doing any marketing, and it was just all organic pull of people who are using a product, telling others about it, sharing it with others. And that's where we're on a path where if we can just continue doing that, everything else is amplification on top of that. you as a founding then can actually say, okay, where might we augment and actually accelerate

the growth? And I think doing so before that would have been premature, but like doing that at that moment, then you have like more of this like massive tailwind to support you and everything else.

So that's like fundamental, like do that. And then I think there's many apps out today. There are, you know, very much more prosumer in nature. And so you have, you know,

this, you know, a growing base of users. It's oftentimes hard for a founder because, you know, go from this point for you know every user

to, the user base being the sort of faceless, like, entity. Like, who is even my user? I don't know. And so, you know, I think a lot of your audience is probably familiar with, you know, founder-led sales as like an important concept. You're selling B2B. Like, you probably need to do that.

You need to be the first person to kind of know how to tell the story of your product, be able to then eventually train and onboard your AEs. I think in this era, it's really important to have founder-led marketing. Like, how do you tell your story broadly in a way that is competitively? a way that really kind of is, is, is, uh, understands like the product and why it matters in the world. And, um, and then, and in this case, like be so close to everything that, that is, uh, you know, it's part of like the brand building. Like, you have to be laser focused on that in the beginning. And I think, again, then everything else becomes a little bit easier.

Sometimes our products that are as horizontal as gamma is monetization can get in the way of growth if you're not very careful about how you design pricing. And a lot of your users are maybe students or other people who eventually might pay, but they might use gamma for months or years without paying. Like, how do you think about where you put the paywall and how you design the pricing? Yeah. I mean, this is one where it's definitely, you know, probably more art than science at this point. And so the way we approached it was, you know, when we ended up doing our first big air launch, you know, signups were going, you know, through the roof, we hadn't built any, all hands-on.com building the first, you know, 30 seconds, we had built, we had zero effort on the

monetization piece. So we actually couldn't even charge people for the product. Wow. And Intercom was blowing up with people saying, I ran out of credits and like, how do I purchase more? It was a very fair question. We didn't have an answer to that. I'm like, okay, yeah, I guess we need to go figure out pricing. So we end up spending, you know, all of April, basically running some initial pricing studies. We did, you know, classic sort of Van Westendor. We looked broadly at like what was, you know, how other products were pricing. And I think, you know, amongst like all these different data points, the conclusion end up being, like, part of this is just, you know, in this moment in time, everyone's familiar with chat GPT's pricing, and they're kind of already anchored on

that. And so you can kind of get very sophisticated in how you want to price it, but that adds a lot of friction into like how people want to purchase a product. And so we ultimately landed something similar to kind of how chat GPT was pricing initially, just as we thought as a good starting point. Let's just at least allow people to purchase gamma. And then we wanted to then see the data, right? Is this a model that was actually economical for? us. Could we make money with that price point? And fortunately for us, we were. We went from a point where, you know, our AI launch was really kind of the sort of bet the company moment because we were about 12 months of runway, which, you know, as a startup is not a great place to be, especially if

there's uncertainty ahead. And so the moment we launched pricing and packaging, it took us about three months to get to like one million error and then become profitable. So it was, you know, for us, it felt like, okay, well, we don't need to change much because at least we know the core you know, unit economics work, at least at this stage, and we're going to continue to monitor it. And then we'll focus more on the product. I think if we had launched and, like, you know, things were working, but we're actually burning a lot of cash. And like that 12-month runway is actually going down even faster, then we definitely would have had to revisit that decision. Fortunately for us, that wasn't one we had to do early on. And we actually really haven't touched pricing in,

you know, two plus years. More recently, we introduced a few new plans. But fundamentally, the initial kind of design, more or less has been the same. Yeah. Well. By the way, I love what you were saying about growth. It reminds me, I think you had this quote in an interview recently where you said strategy is building a business on a product people love. Just the simplicity and clarity of that was, you know, I think very eye-opening, actually, because you can really boil it down to that. Maybe just to this point on starting consumer prosumer, we talked about launching the team's plan. And curious how you think about crossing the chasm to the enterprise and B2B users more generally.

It feels like there's probably been a lot of pull already just because some of your users use you for personal reasons and then they bring you into work. But maybe talk more about that. And how did you decide on the timing of when to launch this plan and, you know, when the time was right? Totally. Yeah. I mean, we think a lot about, you know, sequencing. Like set yourself up for success if you can. and that's where a pro-sumer first is the for in our minds kind of you know where you want to start for a product like ours you want to build a lot of bottoms up love before you even attempt like the b-to-be you know emotion or like that sales process i see a lot of start to do both simultaneously i do think that's tricky some categories maybe you can execute on that but if you can start with the prosumer side and build that sort of bottoms-up love like people recognize the brand they learn to trust the brand and then when you're finally actually selling into a company

there's already some, like, internal champions that want you to win. That's always way easier. And, like, the moment we started feeling that was the moment was like, okay, we feel like you can actually start this process and going into this sort of B to me motion where there's already some tailwinds, right? We have some momentum. We're not like a cold start. We're not just going into this completely cold.

And so that was the sort of early signals we're looking for. And then we didn't want to rush into it. And so, like, then you spend a lot of time talking to those users, like, what are their main points of friction? this is where we started recognizing that this might be just unique into this moment in time but the sort of for many businesses they're looking for ways to unlock AI productivity

for the companies, right? What are tools that go beyond just chat GPT that everybody at the company can use that can familiarize themselves with like the power of AI where other employees can train other employees and it just so happens like slides is one of the formats where like everybody, almost everybody at the company is,

is using on a very regular basis. And so now we're having the conversation where not only is there the bottoms up love, there's a top-down, like, demand for we need something tomorrow to help, you know, really, you know, educate our broad employee base. And I think those two together, you know, hopefully will make it just easier for us to kind of, you know, enter that market in a way where it doesn't feel like just pushing this massive

boulder up a hill. Yeah. I'm really interested in what you said about the kind of bottoms up user love because I think that's something every company inspired. to, but you can't, you know, manufacture it. It just happens. To what extent has that been like a compounding advantage for you? Like is, are there any flywheels that kind of get going as you get users in the tens of millions? Totally. Yeah. You know, I think, you know, oftentimes startups could ask like, what does defensibility mean or durability? And I think there's, you know,

many ways for a startup to build, you know, defensibility. There's some form of network effects that happens. And then there's certainly like brand as a moat. On the network effect, side, you know, we don't have any, I would say, you know, when it comes to like true direct network effect, so not like a social platform. But we do have this sort of local network effect that happens where if you do have that bottoms up love, what happens is you have users that are teaching other users how to use gamma, right? So every incremental user, there's almost a social proof aspect of it. And what ends up happening is you become a standard. You become a language that everybody internally wants to learn. And that becomes a very powerful thing. I think similar things

happen with, you know, tools like cursor, where, like, you become the standard and it's like, okay, yeah, no-brainer, we're just going to use cursor, right? And so I think you only earn that by having some of that bottoms up love. And then, of course, going back to the brand piece of it, you know, if you are top of mind for a category, when anybody, anytime anybody says air presentations and all they can think of is gamma, that's a great, you know, obviously a great place to be for us. And so for us, we want to be that. You can only earn that by, of course, like building that brand awareness over and over again and being invested in doing so over a long time horizon.

Yeah. Yeah, absolutely. And I guess to the point on collaboration, you wouldn't necessarily call this a network effect, but the fact that you can share gamma so easily is a huge advantage. I know we share gaminternally all the time, but it reminds me a bit of Figma and how, you know, they sort of replaced email over time in terms of sending that PDF format. Yeah. I mean, when we, you know, when we're just getting started, we basically just had two mantras.

Make it dead simple to create, dead simple to share. If you do that, you complete the flywheel. And so, yeah, we wanted to make sure that was the case from the very beginning. You build for that organic virality by design and make sure, like, obviously you have to improve everything, but that needs to be fundamental to how you approach it. Yeah, love that. And actually, that's probably a good segue into company building overall. And there's a lot of lore around gamma in terms of just how you built this company a bit differently, especially given the fact that you start in 2020, a lot of companies in 2020 and 2021 were built very different. differently. Oftentimes they got bloated pretty early on, which is pretty antithetical to how you

and John and James built Gamma. Can you say more about these kind of principles you had in the early days? You know, you shared a little bit on the product side, but the culture of Gamma is also very important, distinct. I've heard you talk about having this live culture deck. Say more about like how you guys came up with that. What's in that culture deck today to the extent you can share. Yeah, yeah, totally. Yeah. So, you know, a couple of principles I think, you know, we've tried to stay true to. We had this internal mantra around higher, painfully, slowly from the very beginning. And even as we hit, you know, a lot of growth, we still have to stay true that. So it's allowed us to maintain relatively lean, given our traction. But I think most importantly, you know,

when we first started, we thought about the initial core team. There was seven or eight of us, actually, that all came from optimizely. And we had worked together. for, you know, five plus years. And for us, you know, there was already the sort of foundation of trust. But what I, what we really thought about was we wanted that initial, you know, seven of us, we thought of that as like our MVP crew. We needed that group to be, um, everything we needed to ship a product end to end. So we can create the product. We can also, you know, market the product. And then for me, be able to even sell the product. Like that, that initial seven, that's, that we wanted that to be there so that we would be able to move incredibly fast. So what did

that looked like in practice. Me as one of the co-founders, John CPO co-founder, James, CTO co-founder. Then our first hire was our head of design, product design. Our next hire was our first front-end engineer, and then we hired two backend engineers. And so again, going back to like, this is the MVP crew that could build end to end, and we could test a lot of different ideas along the way, and a lot of us to move incredibly fast in that first year. We didn't have to, you know, wait for the next hire to come aboard. Like, we had everything we needed. And I think that, sort of foundation, getting that first seven right, allowed us to have this core DNA where I could easily imagine hiring the next seven or next 10 and just trying to replicate that DNA.

I think Brian Chesky talks about this where you're like, you really want to get that first 10 right or, you know, first 100 right? Because what you're trying to do is like set the blueprint for the next set of folks that join so that they all share the same values, same principles, hopefully the same level of ambition for where you want to build. And then everything becomes easier, right? like if you know that team is all in it together, team full of hopefully missionaries that anything is possible.

And so for us, getting that first seven right was really, really critical. And then even after that, like having this notion of hiring painfully slowly, just allowed us to always focus on the quality of the people we're bringing in and not the quantity. Sometimes when you hit product market fit,

there's this temptation to just set these really ambitious hiring targets. And we set the hiring targets, the hiring target becomes the goal. And the goal is to hit that target. the target is no longer to hire the best people. And so we've always wanted to know, like, yeah, we need to grow. We need to continue to build a team, but we're never going to lower the bar for who we let in, you know, into the door at Gamma.

And so that's always been, you know, top of mind. We've met many candidates that are maybe strong super technically or maybe good, culturally, a good fit, but didn't, but didn't weren't both. If they're not both, then they're not someone that we're going to, you know, work with a gamma. And it becomes hard because, again, the temptation always is, hey, we have to keep growing. You're on this growth flywheel. You can't hit those without hitting the hiring target.

And I think that's the part where if you can try your best to resist that temptation and be honest to your own core principles, you know, that gives you a chance of actually replicating the DNA over and over. Yeah. So you said the head of product design is higher number one. And we've heard, I don't know if this number is right, but 25% of your team is designers. Yeah. In the beginning, I think it's a little bit less now. But for a while, yeah, it was like a quarter of our team, those product designers. But I also don't think, yeah, is common, maybe among startups.

And, yeah, for us, we cared a lot about, you know, in many ways, AI companies are trying to invent new surface areas, new user experiences. I don't think that's possible without a really strong product design team. And, you know, Zach, who's our head of design, you know, he's been with us since the beginning. He's still with us today. And I honestly think, yeah, without him and being able to really deeply understand what the user's trying to accomplish, that matters so much, right? And we're glad we invested so much in that team.

And I think for us, it's translated into hopefully a better just overall product experience. Yeah. I mean, I guess on the design topic, there's this, I feel like it's overused now, but the concept of taste, people are like, hey, developer taste, researcher taste. But you could kind of think of design as the original arbiter of taste. Do you think there's a consistent sense of what good taste is at gamma? And I ask that because there's an incredible diversity of what you can build on gamma. And yet, when I use it at least, there's sort of clear design choices that have been made that are automated that make my presentation look really beautiful. And so I'm curious, like, what does taste look like at gamma and how do you balance those two sometimes conflicting principles?

Oh, yeah, it's a great question. I'd start by just saying, you know, when people say taste, they oftentimes just think about only the visuals. It's like, what is like the visual output? I oftentimes, you know, the analogy I have more is, you know, if you think about like a restaurant, you go into a restaurant, You know, the moment you go in, you know, you're greeted, you're seated at, you know, your table, you're handed the menu. You finally get, you know, place in your order. The food comes. It's good.

And then you get your check. Like, the end-to-end experience needs to be magical, right? If someone just came in through a plate down on your table and, like, even if it looked good, like, you probably wouldn't go back and tell your friends. And so taste is really thinking about that entire, you know, experience. The moment you walked into the door for us, the moment you, you know, log into our product, to the moment you, you know, you know, to the moment you're actually sharing something with somebody else, your colleague, like we want that to feel great. And the set of user experiences, like the accumulation of all of that,

we want that to feel like something where, again, it's a no-brainer for you to tell your friends about. Like, it's such a delightful experience. And I think that's really where, like, product designers come into place. Like, they deeply understand that entire, you know, journey that the user is going on. And visuals are a big part of it. Like, certainly if the food doesn't look good or, like, taste horrible, like, no one's going to come back regardless.

but it's like, I'd say like only one part of like this bigger journey. And so we think about a lot of that end to end. We don't want Gamma to be, I think, you know, the incumbent tools are much more like a, maybe I'm going to kill this sort of food analogy here, but we're much more like a buffet. You go in and like figure out what you want to grab, throw it on your plate, go back and eat it.

And, you know, that's the end of the day. We want to design the experience for you. And so we want that sort of experience to feel well crafted. You still have a user. You choose what you want. on, but when you get what you order, it's going to be great. You're guaranteed it's going to be great. And so we're trying to move towards that vision where we can help curate a lot of it for you. We can make it feel effortless. And at the end of the day, hopefully, just feels

delightful and magical. And that's really the ultimate goal we have in mind. I guess going back to this point that you were making on hiring, Ben has this great quote from, man, I think 15 years ago. That's aged really well. Yeah. You know, if you haven't done the job yourself, how do you know what you're hiring for. And the reason I think it's aged well, but there's kind of been twists and turns along the way is that there was probably a school of thought where it was like, hey, you have product market fit, immediately go hire the expert and help you just take that over versus doing it yourself for a good bit of time. And I think, you know, you've sort of gone on the record for saying there are jobs that you did in the early days for Gamma

that maybe you didn't have experience doing. Some of them I'm still doing today. Exactly. Founder mode. But maybe share a little bit more about your experience with this concept of do the job yourself before you hire for it. Yeah, I'm fully aligned with what Ben said. I think going back to like hire painfully slowly,

the pain part of it is pretty important. Like where are you actually feeling the pain as an organization? And that's where you start looking for, okay, maybe we do need a hire. somebody or like build out that function. You know, for me, a lot of it early days was around marketing. You know, I come from a finance background operations. I had never done marketing.

And so what does good marketing look like? I wanted to kind of appreciate the job before we went out and hired and built out a team. And so, you know, in the beginning, when we started doing things like influencer marketing, I was onboarding every single influencer myself, right? I wanted to make sure, you know, influencers are going to be an extension of your team. They are essentially the equivalent of, you know, a sales team if you're B2B. And so I needed to make sure that they could tell our story well. I wasn't going to be prescriptive in how they tell the story, but they needed to understand

Gamma. I wanted to make sure they got like white gloves onboarding so that they could fully appreciate the product so that they could tell our story and their voice in the most impactful way. And so those are things like I could have easily just hired somebody to build out the influencer marketing team because it was working already. But then I don't think I would have appreciated all the things. nuances, right? And even doing so, it allowed me to think about even being, you know, become a creator myself, like actually posting more regularly on LinkedIn and Twitter. And then that you go into

the sort of mindset of, okay, what does it mean to be a creator? Well, I need to learn how to be a better copywriter. And I need to learn, like, what's the difference between a good hook and a bad hook? What's a way to engage an audience? And like, those are all things that ends up becoming super important as you actually are scaling up a marketing function. And then when you're finally interviewing people to join the marketing team, are they up-leveling you in some way? If they're not even as good as you and you basically just learned it yourself, then okay, well, you know, that's probably not going to be a great fit. But now that I've at least dabbled in it, I can ask better questions and I have a better understanding of what great looks like. I also know how hard it is

for like overcome like the, you know, the bar of mediocrity. Like you can get sucked there and just be okay. Great greatness takes a lot of work many times, many years to get that. And so I think it just raised my bar of, like, who do we even look for? And when it goes back to, like, when you're actually feeling the pain, you want to be able to put out that pain with, like, the very, very best option out there. And I think that's only possible when you actually start understanding it yourself. Yeah, totally. Maybe to tie that back to your point on founder-led marketing, which I really loved, you are the star of one of the recent launch videos. I was like, wait, is this a paid actor? Oh, no, it's great. You're amazing. And, you know, to your point on

putting yourself out there, not just Gamma the company, but Grant as founder, co-founder, CEO, how did you think about the go-direct process? And, you know, we think it's very powerful at A16Z when a founder can go out, bring your company and product to the forefront and connected to the zeitgeist of what people actually care about and need and what their pain points are. But how did you decide on this journey? And, you know, maybe for founders out there who are also thinking about do I go direct? It's not natural to me. Totally. There's layers of comms in between. Like, what was that decision-making process for you?

Yeah. I mean, I think there's a few different things driving this. One was just, you know, we talk about internally one of our values around just, you know, don't be boring. And so a lot of the content we put out there is just to have a little bit of fun and be playful. And I also think it's a responsibility for the founders to, you know, really be the sort of steward of, like, what is the brand? Right? Like if you don't care about the brand, nobody else will. Nobody will ever care enough, you know, as much about the brand as probably you do because this is, this is like you've created this from the ground up. And so for us, even as we get bigger, you know, we like to have a little bit fun. I think me being part of some of these videos just ensures that, you know, what does the brand represent? It's really, you know, we always think about branding culture being two sides of the same coin. And so for us, you know, we care a lot about our culture as well and our culture is our team. And so being able to confuse. some of that into the brand as well, the external representation. I think it's just been energizing. And personally, for me, you know, working with our creative team, it really does energize me.

Like, they're just so creative. And they dream up things that I could never do on my own. And to be able to play even just a small role of that, I mean, it would be a cameo or an actor, whatever they want me to do, I'll do it. And I think, you know, hopefully we can preserve that for a long time. But, yeah, that's part of, I think, what makes our team special. Well, we thought we'd end on a fun question. And it's sort of this theme. of gamma-ramas, which we didn't even, you know, the first time we heard that, we should call it that internally. But this is actually going to be a question for all of us, so we all have to answer this. But what was the last gamma that you made, or pick an interesting one?

And how do you personally use gamma? Yeah, that's a great one. Oh, for an upcoming, yeah, Gamma-Rama, the one of the topics I was thinking about, I think there's a lot of talk around like the longevity movement right now. And so I did want to kind of demystify some of that. And also just, like, what are the common themes around, like, you know, what are people saying that are important? I think definitely as I get older, you know, I have two young kids. Just even trying to keep up with them just means I have to, you know, invest in things like just being healthier.

And, you know, physical activity, all that stuff matters a ton. So I'm going to try to put something together there. And I think it's a, yeah, kind of a fun topic. Totally. I have a slightly more boring topic. That's why. I actually have two, I'm going to cheat a little. I have two use cases that I love Gamma for.

One of them is, as VCs do, like, we'll go down. deep into a space for months and kind of form a thesis around it. So like voice agents is a space I spent a lot of time in. And so I'll kind of publish a very long gamma, like, you know, dozens of slides. And I think what I really like about using gamma for those is the analytics. Well, first, the ease of publication. Like, it's so easy to open and check out. I can put it in my Twitter bio. I think that's very easy. But then I can actually click in and see, like, what part of this extremely long thesis are people engaging with? Like, what should I write more about? What should I publish more about?

Great signal. Other end of the spectrum, much smaller audience. I use gamas a lot for when we're actually trying to talk to other startups and share why we're excited about them, what we've learned about them, all of that. And for that, going back to the 16 by 9 insight, I think the flexible size and slides and the fact that you can almost package and send something that feels more structured in a good way than a website, but feels as kind of content rich is like very, very special. Super cool. I use it for some of those same use cases, so I won't mention those. but maybe a personal one. I do an annual retreat with my best friends from college,

and we have an agenda for this retreat because we're total nerds. And this year we put it in gamma, and this is kind of our ongoing tradition of, you know, here are the cards, here are the agenda topics for, you know, motherhood, career, et cetera. And it's so interesting because you can kind of see it when you use it for personal reasons, the tie to work and, like, how this is just sort of the ideal way

of putting ideas onto, into reality so we had a user create a graduation like scrapbook like as a gift for the one of their family members and it was it was really cute like there's so many pictures it was one that got shared across generations like grandparents parents kids and you know it feels like it's it's pretty special to be part of some of those moments too even if they're not core business use cases but the fact that they shows gamma for like a pretty special moment and something that's to live on and like share the across the family. It was definitely a fun one to see. Well, thank you so much for being here with us, Grant. This was such a fun conversation and

we're thrilled to have you here. Thank you so much. It was great. Thanks for listening to this episode of the A16Z podcast. If you like this episode, be sure to like, comment, subscribe, leave us a rating or review and share it with your friends and family. For more episodes, go to YouTube, Apple Podcast, and Spotify. Follow us on X at A16Z and subscribe to our Substack at A16Z.substack.com. Thanks again for listening, and I'll see you in the next episode. As a reminder, the content here is for informational purposes only. Should not be taken as legal business, tax, or investment advice,

or be used to evaluate any investment or security and is not directed at any investors or potential investors in any A16Z fund. Please note that A16Z and its affiliates may also maintain investments in the companies discussed in this podcast. For more details, including a link to our investments, please see a16z.com forward slash disclosures.