Why We Sold To Grammarly | Rahul Vohra

We were acquired last year by Grammarly, and then we renamed the entire company to be Superhuman. Grammarly has 40 million users, makes north of $700 million a year. A multi-billion-dollar company acquiring the small company taking the smaller company's name has almost never happened.

We are now building the AI-native productivity suite of choice, and it's been life-changing. But very deliberately, when we first came out, it was for a rather small set of people. There was a time where even knowing theright people to raise money from was...

difficult and challenging. I came to the Bay Area in 2010. There was no angel list. Obviously, there were no AI chatbots to teach you what a convertible note is, or how a safe works, or how to raise money.

And so, as a result, raising capital itself became a moat. But the biggest moat, though, today that exists is not writing software. In a world where the marginal cost of software tends toward zero, where anyone can build anything increasingly cheaply, what moats are truly left?

Well, I can think of two things. I think it's key to what makes Superhuman Mail a great product. It's key to what makes all of the products that we work on amazing. And it's all because of—

Rahul, welcome to NEW ECONOMIES. This is really another full-circle moment, as I last interviewed you, I think, probably 5 or 6 years ago now, and it feels like, to be honest, it feels like a bit of a blur because so much has happened just in the last 1 or 2 weeks, let alone in the last 5 or 6 years.

So awesome to see you, and welcome back to the show.

Oh, thank you for having me back. It's exciting to be here.

So, so much has been happening in AI in the last few months and in the last few years. But I think let's go back to the early beginnings. Some people might not know, but Superhuman is actually a 10-year-old story.

And when I meet founders, most of them want to disrupt the large incumbents, and most of them want to go after some pretty interesting categories that are fairly easy to disrupt. But email is really the complete opposite because there are so many free email providers out there.

Why did you and your co-founder go down this crazy idea route of thinking and deciding to disrupt email?

Yeah, what a question. And there's quite a lot to untangle because, of course, Superhuman today is not just an email company. We were acquired, the email company was acquired last year by Grammarly, and then we renamed the entire company to be Superhuman.

So we are now building the AI-native productivity suite of choice, and the idea is to bring AI to everywhere you already work. So just to lay the foundation, to begin with, that, of course, includes Grammarly, which is your trusted, always-available writing partner.

It includes Superhuman Mail, which is what you were asking about. That's the fastest, the most productive email app ever made. It includes Coder, which is the all-in-one AI workspace for teams, the best place for teams to work together with AI.

And the new product that we're building called Superhuman Go, which is your proactive AI companion that actually works in every app and tab you already use. But the Superhuman story, as you pointed out, started with email, and that was over 10 years ago now.

And your question was like, why did we set out to do email? Well, in short, it's because I wanted to solve the biggest possible problem. And email is a way bigger problem than most people realize. There are roughly 1 billion professionals in the world, and on average, we spend 3 hours a day reading and writing email.

So that's 3 billion hours every single day, or north of a trillion hours every single year that go into email. And it's actually really hard, I think, to find something that people spend more time on. I did actually find something that people spend more time on, which is sleeping.

We spend more time sleeping than we do reading and writing email, but I had no idea how to fix that. What I did have, though, was a very clear vision on how to fix email. So we set out to rebuild the entire email experience, and now our customers are getting to their email twice as fast as before, they're responding faster to the things that matter, and they're saving 4 hours or more every single week.

It feels we have progressed so much in not only how we communicate, but also the types of tools we use. And the types of tools we use. And honestly, I kind of feel like, and love to get your take, I don't use email as much anymore.

I much prefer the quick communication or using tools such as Voice to help at least write. What is happening in the world of email today, and is it actually still going to be relevant in the next few years?

This is one of my favorite questions because anyone who's been following technology for any period of time will know that roughly every 5, 6, 7 years, a really well-known, respected journalist will come out and write essentially the obituary to email.

Why email is long and done for and should be buried in its grave, and we should pay its last respects. And I've seen that article at least 4 times over the past, I would say, yeah, 20 years or so, actually.

I'm dating myself here. And the thing is, email doesn't die. Now, why doesn't email die? Well, if you think about how companies work, and I want to specifically talk about work email, companies need a way to address a specific person that is owned by the company.

So if you think about my email address, rahul@superhuman.com, it's actually the most efficient, human-readable way you can probably come up with to identify me at my company, superhuman.com, because it's literally my name at the name of my company.

And that is, in essence, what most email addresses are. Except that it is also owned by the company. So technically, the company owns all of the emails that are in the account, and that means that, like, I can't take the stuff that's in my email and then go use it in a way to harm the company and start a competitor, that kind of thing.

All very normal stuff that I think we all agree is appropriate and correct. And that's why, for example, we don't use our mobile phones, our cell phone numbers, as the primary way to address each other at work. Although there are some industries where they do that, and in those industries, lo and behold, those mobile phones and those cell phone numbers are actually provided by the company.

So it turns out that email is the most efficient way to address anyone at any company. It also turns out to be the unit of authentication. If you think about how you sign into every other application than email, you sign into it with email.

And even if you're using a single sign-on solution like Okta, you typically sign into Okta with your email. So it truly is the root identity at most companies. And because of those two things, the email address itself has had such staying power, and I think will continue to stay well into the future.

But the other half of your question is, like, what is the actual email experience and how is that changing? And before the show, you alluded to Voice. I think Voice is a really great example of the way it's changing.

We recently launched a feature in our mobile app that people absolutely love; it blew people away, called Right With Voice. And so if you haven't used this yet, you can literally pull up an email, tap the little icon, the little voice icon, and just start talking.

And this isn't dictation like it is in Gmail or Outlook. You can literally say, "Hey, reply to this email and say I am available for a meeting. Pick 3 times tomorrow, but make sure they're in the afternoon where I'm free."

And then also ask if they can send the pre-read materials ahead of time. So you're having this conversation with the AI, and the AI will literally then go away, do all of those things, check your calendar for tomorrow, see when you're available in the afternoon, write all of this in your own voice and tone.

And all you had to do was speak it. So we launched this recently, and as I say, the results have been phenomenal. I think we now see 10% more people using Right With AI than we did before, and 30% of that is coming from the mobile app.

So we literally went from no one's doing it on mobile at all, really, because it was kind of annoying and painful. You had to type that prompt in, and by the time you're typing that prompt in on the phone, you may as well just have written an email, to tons of people are now doing it, and it is literally changing the way that people email.

So the email experience itself, that is going to keep on changing and change radically, but email addresses and the core infrastructure, that's here to stay.

That's such a great point around the email security, the sign-on. All of these apps and social accounts or new applications we use,right, we always have to use our email. On the Voice side, when did you decide we have to build something in Voice to stay relevant?

I've wanted to build this for such a long time, and I've been a huge fan of a company that you may know, WhisperFlow, or the product is called Whisper, which is this incredible product for desktop. You just hold down a key, and for me, it's the function key, and then you dictate.

And that then takes what you're saying and puts it into Slack or email if you're in Superhuman or Figma, where I also spend a lot of my time, or Superhuman Docs or Coder, which is one of our products.

And it's been life-changing. I can talk reasonably fast. I would estimate I'm talkingright now at around, you know, maybe 200 words per minute, maybe 250 if I'm going quickly. And I'm a decent typist, but I'm not that decent of a typist.

And the faster I go, the more mistakes I make. Whereas with tools like Whisper, and there are a number of competing products in the space, but with tools like that, you can literally talk as fast as you think.

And it only took a few days of using this kind of technology to realize, oh, this is how we're going to interact with technology in the future. So that was one experience. Another was when I first tried advanced voice mode in ChatGPT.

And, you know, when it first came out, the latency was, it was noticeably high. You'd say your piece and then kind of hang around for a while, and then eventually the ChatGPT AI would reply. But it was just fast enough that if you squinted a little bit or just, like, you thought about it, you would realize, okay, something really special is about to happen here.

So I would say it was about a year and a half ago. And then, of course, it's the standard challenge of assembling the resources and making it possible. For a while here on the Superhuman Mail team, we didn't actually have a fully staffed mobile team, which, in retrospect, was something I wish we hadn't done.

But now we have a fully staffed mobile team and we're able to run really hard, not just at the Voice problem, but at all of the interesting things that we can do on mobile.

What's so incredible, I think, about the whole AI stack of applications, if it's vibe coding, if it's Voice, the accuracy has become so much better. What makes these Voice tools so good today, and how are they so accurate?

And what I mean by this is, if you were saying "um" or the "uhs," all the typical things we say in all of these Voice application tools, or just when we're speaking now, the Voice dictation apps or just Voice general apps, they remove all of those.

What makes these tools so good, do you think?

Well, I think on the one hand, the models themselves are rapidly improving, and that's being driven by the super intense competition that's in that space, not just from companies like OpenAI, but you also have 11 Labs, you have many other companies that are going after this problem.

And in the best possible sense, they all drive each other to make the best possible product. And on the flip side, application developers are getting better about managing context and making that available to the model. So what tools like Whisper do, and what tools like Right With Voice, the Superhuman Mail Voice tools do, is assemble context that is useful for the AI when it comes to disambiguating what you're saying.

Whisper, for example, might take a picture of your screen. So this is one of the things I really love about it. If I'm talking to someone whose name has maybe an unusual spelling, or I'm referring to a project or a code name or an acronym that the AI would have no reason to know because it's internal to the company, and why would it have ever come across our weird word?

It can still be transcribed correctly because some context was grabbed, which is what was on the screen at the time when I was dictating. Similarly, when you use Right With Voice on Superhuman Mail on the phone, it has access to your contacts.

It knows how they're spelled. It's able to look at the email thread, the entire contents of the email thread that you're responding to. If there's a weird code name or a project in there, it's able to freely use that.

It can pull context from the rest of your inbox. So this context management, which has now become just a core competency for anyone building an AI application, is the other reason why these tools are now so much better.

You know, so much has changed just in the last few years, and the way tools find product, like the way tools find product growth or, you know, product market fit is becoming so much quicker. What I have found so incredible just about the Superhuman story in general is at the bottom of each email, it says, "Send by Superhuman," and it's still today one of the most popular features, I think, on the email tool.

Have you been surprised just quickly how quickly, but also the reasoning people keep that on their email?

If your question is, have I been surprised, the answer is no, not really, because it was by design. Superhuman as a brand has changed over time, but very deliberately, when we first came out, it was for a rather small set of people.

It was initially designed, and again, this is deliberate, for CEOs, for founders, for executives, for managers of large organizations, for leaders who are insanely busy, have historically been ignored when it comes to their productivity and their software needs, who command outsized budget relative to just being one person, and for many of whom their work is email and email is their work.

So it was the perfect storm to create something that they themselves really loved. Now, there is this phenomenon that I think we've all experienced, which is, to some degree, we kind of want to be like our boss. We want to be our manager someday.

Or there are people in the industry who we aspire to be, if we don't happen to respect, for whatever reason, the people in our company. There are people that we look up to, and we put Superhuman Mail in the hands of those people.

And that's how it started to spread. So it became an aspirational thing. Partly also, it's simply the name itself. Of course, the first company to do this famously well was Hotmail, and that really popularized the idea that email signatures were a way to go viral.

I think the iPhone did it to great success. Sent from my iPhone for a while was like this badge of, like, status and even wealth to begin with, and of course, now half the population has an iPhone. But we deliberately tried to ride that same curve, and it worked really well for us.

The name Superhuman is interesting. When you see "Sent via Superhuman," if you don't know what that is, that is so much more interesting than "Sent via I'll pick on mailbox," an old email app. I forget what exactly their email signature was, but let's say it's "Sent from mailbox."

It's so much more boring than "Sent via Superhuman." And as a result, people click on it, and they come to the website. They come to the website in droves, in fact. I checked yesterday because I knew that this was going to come up, and 20% of site traffic is still from the "Sent via Superhuman" signature.

Interestingly, it's actually been at 20% of site traffic all the way from the early days when we first launched that feature. And so as our user base has grown, that percentage has stayed steady because email has, of all the things we do, has one of the most massive fan-outs.

Fan-out is a technical term when you're building viral software. Basically, how many people does a user interact with and expose the brand to? And if you think about how many people email uses email, it's a lot of people.

So a lot of people are seeing the "Sent via Superhuman" signature. And because it's interesting and because it looks cool, and even today still has this sort of badge of, yeah, I care about my tools, and I have chosen to use Superhuman, and that says something about me and my relationship with my work and how productive I want to be and the person who I want to be.

And I'm actually proud to share that in the way I might wear some Under Armour clothing or some other relatively premium piece of clothing. So those are the reasons why people keep it on, and that's the reason why it's super effective, and as I said, still drives 20% of site traffic today.

That's insane. 20%. And if you think of how much the company's compounded in the last 10 years, that's still a huge amount of traffic. And I feel also, you know, I've got quite a few friends who use the product and tool.

It does feel like a badge of honor when they have Superhuman at the end. What do you think is the biggest modes software companies need to have today when they're creating these virality type of software tools? Is it having some form of community, some form of privilege around the tool in order to make it

viral, but also, you know, scalable?

Modes are hard. Modes are, in fact, increasingly hard. It used to be, let's do a history of modes. We won't start with castles, that would take far too long. We'll do a history of modes as it relates to venture-backed software companies.

It used to be that there was a mode around simply having theright connections,right? That there was a time, it's a long time ago now, but there was a time where even knowing theright people to raise money from was difficult and challenging.

And that was sort of just before my generation of founder. I came to the Bay Area in 2010, and I did my last company in 2010 called Reportive, went through Y Combinator, sold it to LinkedIn. We built the first Gmail extension to scale to millions of users.

Basically, on theright-hand side of Gmail, we showed you what people look like, where they're based, facts about them, which, by the way, is now a feature in Superhuman Mail. And back then, there was no angel list. There were, obviously, there were no AI chatbots to teach you what a convertible note is or how a safe works or how to raise money.

There was a little bit of help online. Naval and Nivi from Angel List had just started this thing called Venture Hacks. And for the first time, people were writing about the mechanics of how option pools work and the difference between pre-money valuations and post-money valuations and how dilution can really hurt and why you shouldn't raise too much money and how liquidation preference can kill a company and all of these things that kids today know inside out because

they grew up watching my generation of founder and the next generation of founder actually build their companies. But there was a time when no one knew this stuff or very few people knew this stuff. And that was a moat.

And so as a result, raising capital itself became a moat. Even before then, being able to acquire hardware for colocated servers was a moat. Today, we have a version of that moat, which is being able to acquire enough GPUs.

So I suppose that part has come full circle. The biggest moat, though, today that exists is not writing software. It's not, to some degree, it's hiring a team. But the biggest moat that exists is the one that you mentioned, which is distribution.

In a world where the marginal cost of software tends towards zero, where anyone can build anything increasingly cheaply and maybe one day nearly free, what moats are truly left? Well, I can think of two things. One is the degree that you have built-in distribution.

Can you get your product in front of theright people in theright way with theright message so that they use it, love it, stick around, and tell the people about it? That's one. And then the other thing is, this isn't really a moat, but it's a competency that every company needs to have, is restraint.

Do you have the restraint to say no? Do you have the taste to know what not to build as well as what to build? And I see this, by the way, in increasingly many products. It's just like, you've got this feature and that feature and this feature and that feature.

And

that clearly, no one is saying no because it's too easy to say yes. And I am reminded of, this is a classic example, but it is so relevant. Instagram, before it was Instagram, was in the era of social, it was like Solo Mo, social, local, mobile, I think, is what we used to call it.

But it was all the things. It was like a check-in app. It was a photo app. It was a friend-finder app. It was a social network, all at the same time. And it had a different name. It went by the name of Burbn.

And it was fine. People didn't care that much about it. It was one amongst a number of those apps. But when they focused on purely the photography piece and they did all of the other things to make that experience excellent, as is well documented, that's when it took off.

That's when it accelerated. And so focus is a core part of what makes products great. I think it's key to what makes Superhuman Mail a great product, for example. It's key to what makes all of the products that we work on amazing.

It's why Grammarly, by the way, is this incredible product that has 40 million users, makes north of $700 million a year, is beloved today, is, by the way, an AI product, although people don't think of it as an AI product, all because of its focus.

So those are what moats look like today. On the one hand, distribution. On the other hand, taste, which leads to focus.

I feel, you know, especially in the

era we live in and building, it's so easy to build, as you said, you know, so many different tools. Having the taste and having the desire to focus is super, super important. How does Superhuman as a group today stay focused when the abundance of opportunities is unlimited?

There's macro focus and micro focus. And at a macro scale, you actually want to have multiple bets. One of the most exciting things for me about Superhuman is we are a multi-product company. When we were in the process of selling Superhuman Mail to Grammarly, which is now Superhuman, Shashia, who's the CEO, and I obviously had many conversations.

And in one of them, I said, you know, Shashia, one of the things that's most exciting for me is we get to do this roadmap slide. And I pulled up a slide from our Series C fundraising pitch deck that had all different kinds of products on it: documents, an agent platform, calendar.

I'm being careful what products I name because I can't reveal too much of the roadmap, but like a lot of really cool products. And I said, you know, the incredible thing is we now get to pursue all of this in parallel.

And that's amazing. You know, as a founder, and Shashia had, I'm sure, had this experience himself as the founder of Coder. He was also acquired into Grammarly. You tell these stories of like, yeah, this company is going to be a multi-billion dollar company.

It's going to have tens of millions of users, hundreds of millions of users one day. It'll be a household name. And this roadmap that today we have to contemplate doing in sequence that might take 10, 15, 20 years, will one day build that thing.

And it's incredible to now be on the other side of those facts. We are a company at that scale. We are a company that's growing quickly. We do have that order of magnitude of users. We are working on all those products and more that I can't yet mention in parallel.

So is that focus? You could argue it's not particularly focused. In fact, it is a deliberate portfolio approach that has theright amount of risk. We think we're taking, we're applying a lot of ambition to the problem because the opportunity today to build these incredible products and to build this AI-native productivity suite of choice is bigger than it ever has been.

So a lot of words to say, we have a lot of fuel in the tank and we're throwing a lot of gas at the problem. Then how do you run a company well when you have such a large amount of ambition and so many things going on?

That's where the compound startup comes in. So I run essentially the mail or the communications startupright now that has the mail product, that has the calendar product. There will be future communication products that are part of that business unit.

But we have a number of other business units as well. One that you will know well is Coder, which is our document editor. We have another business unit that contains Grammarly, which is the core business, the core legacy business, as well as new agents that we are building.

One that's doing particularly well is an AI detector agent and a humanizer agent. So the ability to, as you're writing and reading, help you detect when what you're reading was probably written by AI, or if you are using AI to help you write pieces of your documents, just to help you write that in a more natural, humanistic way.

There's that word human, by the way. It's part of the company name, one of the reasons why we chose that company name. So we have these business units that we run the company with. We are a multi-product company, and that's how we maintain focus.

Each leader of each business unit is very focused on the set of things that that business unit does.

It's quite a unique approach. It's similar to this company called Bending Spoons, which I think what they're doing is really interesting. They're obviously buying companies,right? And there are some similarities. If we were to go inside Superhuman Group today as a whole, you have all of these different products.

How are you actually all working as a team in parallel?

Yeah, yeah, great question. And I love that you mentioned Bending Spoons because I actually think whilst it superficially looks similar, there are so many differences in how Bending Spoons operates and how Superhuman operates. And I have huge respect for Bending Spoons.

I think, I don't think we could do what they are doing. I equally don't think they could do what we are doing. So my understanding of the Bending Spoons playbook is

acquire companies. These are often very legacy companies, companies that maybe folks would have written off from the outside. Of course, they just acquired AOL, and there are a number of others over time, Evernote, which I still use and actually really quite like.

And I'm glad to see that someone is finally maintaining it. And they often strip out a lot of the cost from those companies in a way that resembles classic private equity. They restaff it using their own personnel. I believe they're located, is it in Italy, if I'm remembering correctly?

And there is not, they do

reutilize technology components between the products. That's where they get leverage. But they're not necessarily trying to advertise one product to another or to build an integrated suite of products. That's not the goal. That's very, very different to what we're doing.

When we buy companies, and we're very acquisitive. We just recently announced the acquisition of roads.com, which adds incredible spreadsheeting and table functionality to Coder, as an example. We usually acquire the whole team, the whole company. And the goal is to buy real businesses that can have a meaningful, ideally a meaningful impact to our business at the same time as buying technology, at the same time as buying incredible teams.

We also then want to put those products in front of the users that we already have. Grammarly is one of the very few products that are at massive consumer internet scale. And one of the biggest opportunities we have is to leverage all of the audiences of all of the products that we have into each other so that a user of Superhuman Mail can become a user of our document editor, can, if they want to, start using Grammarly across the surfaces where they write and then maybe encounter Superhuman Go, which is our upcoming proactive AI assistant.

And so there's this, that's just one possible journey. There are many journeys through our products, but we want to get the users of our products to experience the whole bundle. And we call it the Superhuman bundle. In fact, if you go to superhuman.com and look at our pricing page, it articulates a bundle.

It articulates that for $15 or $30 a month, you're going to get these many products. And our goal is to keep adding products to that subscription so that over time, it's increasingly a no-brainer that you should buy this thing.

Today, there's three, four, five products in the $15 or $30 tier. Over time, that list is going to get increasingly long. And the value for money of that subscription is, it's great today. It's going to become truly excellent over time because each of these products are going to be the best in the category they're in, in the way that Superhuman Mail is by far and away the best email application that's ever been made.

So that's a very different strategy to Bending Spoons because we're building towards an integrated product suite. We're trying to cross-pollinate our users where it makes sense. We're trying to build a bundle. And we're very interested in the acquiring teams having a huge impact at the company.

And we want the CEOs, the founders, the leaders of those teams to become executives at the company in the way that I have, to expand their remit and to be part of this growing company.

I love that so much. As soon as you have the distribution, there are so many amazing and incredible opportunities that lead to. What do you think is missing from the bundle that you would love to see at some point in the future?

We are really open to acquiring companies of any size, small all the way up to really quite large. We have a lot of financial power. We actually raised a billion dollars of capital last summer to fuel go-to-market as well as M&A.

And the acquisitions can be anything from

accelerating the roadmap of our

products. And an example would be roads.com and how that helps Coder or our document editor, all the way through to meaningful businesses that are at large scale and are growing quickly in their ownright and gets us from zero to 100 miles an hour in a whole new product area.

The one product I think is missing and potentially an opportunity. Slack is so bad.

I agree. Slack itself is probably a little too expensive, and I doubt that Salesforce would be selling it anytime soon. But I agree there is a huge opportunity in the chat space. I think they're doing good work recently.

I like their new activity feed. It's decent in dense mode. They're starting to fix some of the information hierarchical problems. And of course, there's the new headless mode if you're going to use AI and agents with Slack. But for a long time, I have thought that Slack and more generally chat is ripe for a full redo in exactly the same way that Slack did to the generation of products that came before it, like Yammer and Campfire and even IRC.

Yeah, I agree. Why do you think chat is so hard to build and execute? And if we think about it, there are not actually that many players. There's Slack for enterprise. There's Discord, which is more for consumer gaming.

No one's really cracked the enterprise space yet. There haven't been that many new players.

Well, arguably, Teams is significantly larger than Slack. And the reason they sold is growth was stalling out. Teams was skyrocketing. And Salesforce really needed a play in the space. So it's easy to forget this as startup practitioners, but Slack kind of lost.

They're the second place, and that's ultimately why they ended up selling. That said, they also did fantastically well. And their success, I think, is the reason why we have not seen new startups enter the space. I think what Slack did uniquely well that Yammer never did, that Campfire never did, and many of the other companies whose names we don't even remember anymore never did, is they built a network effect between companies.

The thing that was originally called Slack Connect, but the idea that you can

connect with people outside of your company is one of the things that adds an extra level of network effect that we previously didn't see in chat applications.

When acquisitions come about, very few acquisitions then rebrand to the acquirer. How did the whole acquisition come about, first of all? And why did Grammarly, who was acquiring Superhuman, then rebrand the whole group to Superhuman? It's a bit of a tongue twister, but how did the whole story come about?

Okay, so there's a lot there. Let's talk about how the acquisition came about, and then let's talk about the name. So I met Shashia, who, as I mentioned, is the CEO now, many years ago, back in, I think it was 2017.

And back then, he was the co-founder and CEO of Coder. And we were at a conference. It was the Lobby Conference, which takes place in Hawaii, run by David Hornick and his team now at Lobby Capital. And it was one of those special moments when nobody else was really around.

It was just him and me. I think we were both early. And so I went to hang out by the pool, literally in my swimming trunks. I think he was also in his trunks. We both had margaritas. We were chilling.

I was like, "Hey, who are you? What do you do?" He starts telling me his life story. And it turns out one of his first jobs was to work at Microsoft. And he worked on Microsoft Outlook. And I was like, "Well, boy, do I have a product for you."

So I show him Superhuman Mail. Turns out he loves email. When he was at Google, he became a hardcore Gmail power user. And as you might know, back then, we only did one-on-one VIP concierge onboardings. So I onboarded himright there and then,right by the pool.

And as those who've done the onboardings know, one of the very last steps is when I ask people to close Gmail. And as his mouse was moving over to the Gmail tab and he was about to click the X icon, another tab caught my eye, which is a fav icon for an app called Krypton.

And I asked him, "Hey, what's that?" And Shashia then proceeded to give me the best product demo that I'd seen in years. It was a document, but it was also a spreadsheet and a database. It was a collaboration tool as well as a mini app builder.

And if that sounds familiar, of course, we have a number of tools that have tried to do some of those things over the years. But back then, this was truly mind-blowing. It was the first time I'd ever seen anything like that.

And so Krypton, of course, then renamed to become Coder. And Grammarly then acquired Coder about six or seven months before we acquired Superhuman. And in their acquisition announcement, Shashia wrote, "As I wrote the foundation, sorry, as I watched the foundational capabilities of AI change just about how every tool and surface operates, I started drafting my 2025 planning memo for the team.

I titled it the AI Native Productivity Suite." So I read his acquisition announcement, and I thought, "Wow, the AI Native Productivity Suite. That sure sounds like what we're building at Superhuman." Because our vision has always been to build the AI Native Productivity Suite of choice.

As we talked about email as a critical part of that suite, there's a billion professionals in the world. We spend over a trillion hours on email every year. In fact, we spend more time in email than any other work app.

So Shashia and I caught up, and over several conversations, it became very clear that we share the same vision, which is to build the AI Native Productivity Suite of choice, a suite that is made up of both application surfaces like email, like documents, like calendar, like maybe one-day chat, as well as agents like Grammarly, which is actually one of the first agents ever made it.

It follows you around and helps you be brilliant at writing wherever you are. But you'll see a lot more of that kind of agent very shortly. It turns out email also sits at the heart of where Grammarly is used today.

For professionals, email is actually the number one use case for Grammarly. And Grammarly helps write more than 50 million emails every single week. So that's why the acquisition happens. Email is clearly a large part of any AI Native Productivity Suite, and email is also a key part of what Grammarly is used for.

Let's talk about the name. In fact, in that first conversation, when I read Shashia's blog post and I pinged him, I think I asked him, I was like, "Obviously, Grammarly isn't really a viable name going forwards. Are you going to use Coder?

Are you going to pick another name?" And he shared, "We're probably going to kick off a naming search. We're going to hire an agency. We're going to look at our assets. We're going to figure out what's possible." And over the course of the acquisition, both, and as I mentioned, these conversations took six or seven months, which is a little bit longer than usual.

And we're able to get through these deals faster now. But that's because when we started these conversations, Shashia wasn't even the CEO yet. We had this conversation at the end of December. He wasn't the CEO until January 10th.

And then, of course, he had to spend a month or two actually understanding the business before we could truly engage in M&A discussions. So we had time. We had time to have some of these naming conversations. And we both agreed that whilst the Superhuman naming asset was interesting, we shouldn't have it be a primary reason for the acquisition.

And so I told my team that, he told his team that. And we were really just focused on the fundamentals of things like growth rates and technology synergy and team quality and roadmap synergy and so forth. At the same time, Grammarly hired an identity agency, a naming and branding agency.

And unbeknownst to me, they independently suggested the name Superhuman. They had no idea that there was an M&A discussion going on. Of course, these things are very confidential. But they independently were like, "Look, Superhuman's a great name. If you could somehow make Superhuman happen, you should go do Superhuman."

And lo and behold, we happened to be in M&A discussions at the same time. So it was one of those very rare moments where all of the stars happened to align at just the same time. And that's when you know that you have to walk down that road.

And then, of course, there's the actual name itself, which is such a good name for what we're trying to do. It speaks to making people the best versions of who they are, to giving you superpowers everywhere you work, to unleashing your Superhuman potential.

And at the same time, what we like most about it is that it has human in the name, where so many companies are working to automate, to remove humans from the equation, where so many companies are butting up against what should a machine do versus what should a human do.

We really want to keep humans at the center of what we think is most important: taste, creativity, strategy. Of course, AI is going to help and become excellent at all of those things. But we want to keep humans at the center for what humans can uniquely do.

And what we believe is that's where teams of people come together to do their best work, where I combine my lived experience with your lived experience. And there's a frisson, to use the French term. There's an interesting friction or a coming together that can only happen when human beings get together and do their best work.

And that's what we're most interested in making possible. And that is Superhuman.

Such a cool story. Like I said, it's very rare for the acquirer to then rebrand their entire brand to the acquirer. I don't think it's really happened that much, actually.

Maybe once or twice. Someone gave me an example the other day. DataVent, I believe, is a multi-billion-dollar company acquiring the small company, taking the smaller company's name. But you'reright. It has almost never happened.

It really happens. One of the things we also love to ask on our show is getting inside the actual brain and day-to-days of all the amazing guests we have. So if I was to join as your chief of staff for a day or for a week, what does that actually look like?

At this point, so I'm 11 years into the Superhuman journey. And I'm so fortunate and blessed to be surrounded by a team of people who are incredible at what they do. And as it relates to me, the things that they do tend to be the things I'm not good at doing.

So my super skills as a founder are around product design and marketing. And when it comes to marketing,

not the science of it, not paid acquisition, but more brand marketing and product marketing and narrative and storytelling. And I think that that's visible in Superhuman. Everything from the name to the brand to the movement to

it being an aspirational product in the early days and its evolution now as we're crossing the chasm. And now it is a mass market product, or it's becoming more a mass market product. We have very large enterprise deployments.

That whole evolution is storytelling. That's where I spend most of my day. So I would say maybe 5, 10 percent of it is outwards facing like this. I'm recording a podcast, or I'm traveling to a conference. I'm telling the story.

One of the stories in particular I'm telling, like we spent a lot of time on, is our acquisition story. We're genuinely interested in meeting great companies and great teams that can accelerate our roadmap. So that's some of my work.

But most of my work, actually, is product design and marketing. And I would say that the remainder of my time is a quarter on each of those things and then a quarter on strategy.

There are so many amazing opportunities now for founders to build. The cost of software is pretty much zero. Anyone can now become a builder today. And there are so many categories that are yet to be disrupted. Looking outside of the Superhuman bundle, what type of companies are you most excited about today?

And if you weren't building Superhuman, what category would you go after to go and build something in?

So I should point out, by the way, I very actively angel invest. And that's another way to get involved if an acquisition doesn't make sense. I do this investing with my investing partner, Todd Goldberg. We've called it Todd and Rahul Capital.

And we've invested in over 120 companies at this point. Thank you. We decided to keep it really simple. Todd and Rahul Capital. I guess it'll make succession planning difficult in 20 years or whatever, but whatever. Yeah, so we've invested in over 120 companies, nine unicorns now, over 100 million AUM.

And that's been going really, really well. So we love investing in really gritty founders who are highly persistent, who have a technical edge, who have a distribution advantage. It could be a channel that they've figured out or contacts from a previous role or anything in between those two that move at extremely high velocity and where we can look at that founder and say, "Yeah, this person, they're legit one of the best people in the world to be running this company."

Which, by the way, is the bar I would set for myself if I do another thing, which I'll come back to that in a second. So that's the bar that we have for investing. And every founder we've ever invested in meets that bar.

One of my favorite examples, I'll give you two examples recently. One is Loyal, which is a biotechnology company. And the other is Supabase, which is a well-known AI company. So let's talk about Loyal first. Loyal is a longevity company.

They are building a lifespan health span drug for dogs. As you probably know, dogs die sadly early. I'm a dog lover myself. Fortunately, I have a small dog. So the life expectancy of my dog should be around 15, 16, 17 years, fingers crossed.

But large dogs often die at 6, 7, 8, 9 years old. And they do so for entirely predictable genetic reasons. And that set of reasons is susceptible to therapy. With theright medication, you can theoretically extend the health span of these dogs.

And Loyal is the first company in the history of companies to ever receive FDA efficacy approval for their drug. And what that means is the government agrees that their drug extends the health span and the lifespan of large dogs and soon-to-be dogs in general.

And that is absolutely incredible. And think about what that means. We truly are living in the golden age. A lot of the conversation is around AI. There's a lot of conversation, of course, around peptides and GLPs. But then on the longevity side of the equation, I'm super excited because this works.

This drug works. And

it is no accident that Loyal are working on dogs first. I mean, apart from being man's best friend, dogs are incredible. We don't deserve them. It's truly unconditional love because of the fact that the life cycle is 5 to 15 years, the petri dish of experiments can run quickly.

So as a company, we can evolve our science and our commercialization and our go-to-market fast, become a very fast-growing entity, become self-sustaining, and then apply what we learn to other animals, to ultimately humans. And I think that within our lifetimes, you and me, we will see significant not just lifespan extension, but health span extension.

Not just due to peptides and all of the great stuff that's happening there, but also due to therapies that are going to come from Loyal and other companies in that space. So super excited about Loyal. Another example is Supabase.

Supabase, of course, is not a biotech company. It is an infrastructure company. It is also now an AI company. If you are vibe coding, if you're building an application from scratch, then you will need to store data. And chances are you're going to store it in Supabase because it is the best and the easiest tool for the job.

And I think one of the most interesting things about Supabase is that inflection point was really when vibe coding and AI coding tools took off. Because what would happen is you would come to a coding tool and say, "Hey, I want to build an email app for iOS," which, by the way, don't do that.

It'll take you 11 years to build a really, really good one. And it will be like, "Cool, you need to store something. You need a database. OK, let's use Supabase." And the agent will go away and sign up for it all by itself.

And that is incredible. The fact that AI tools cannot just recommend, AI systems cannot just recommend what tools you should sign up for, that they can actually also go and sign up for it. And we are seeing this ourselves, by the way, with Superhuman Mail.

We recently launched our MCP, which is incredible. If you use Superhuman, please go check it out. It is so great. If you don't use Superhuman, this might be one of the reasons why you want to start using it.

Unlike the Gmail MCP, unlike the Outlook MCP, you can actually send emails. The emails are written in your own voice and tone. You can just give it vague direction. It will write the emails in your own voice and tone.

It will use the context from the rest of your inbox. We've already embedded and indexed all of your inbox. So instead of using, if you search by the Gmail MCP, it's literally just doing bog standard Gmail searches, not very powerful.

You can actually do semantic search via our MCP. You can set reminders on emails. You can send tracked emails so you know when people open your emails, all of those things. Really, really powerful tool. And we worked closely with OpenAI and with Anthropic to build those integrations with ChatGPT and Claude, respectively.

And those applications are massively taking off. In fact, 11% of our new users per week are coming from Claude, which is crazy,right? So you go back two years ago, these AI assistants were not

growth channels. But now they are significant growth channels. I gave you the example of Superbase. It's happening with Superhuman Mail. We're seeing it in more of the investment portfolio. So

this kind of goes back to your moat question, by the way. The other thing I would work on if I were or I advise companies to do this all the time, or if I were building something new, is how can you get the AI assistants of choice to recommend your tool as one of the core tools that people should be using for a certain problem?

If you go into Claude today and say, "Yo, I'm really behind on my email. I would like help with email." And you can try it. It will say, "OK, for email automation, you can connect to the Gmail MCP, or you can connect to the Superhuman MCP."

And it's so much better in all of these ways. And that's just driving a ton of traffic, which is incredible.

Yeah, that is incredible. I was going to say there are now these new SEO tools out there, which will help companies get citedright. And that's going to become potentially a new level of growth for a lot of these AI and software companies.

Yeah. I don't know if I believe that those companies work. Frankly, it is a lot of how

Claude and Anthropic and LLMs in general decide what is viable. Well, there's two ways that they decide what is viable, what's a good tool. One is, of course, they just simply read the internet. So it's all the usual stuff about building a great product and getting it out there.

Do people like it on social? What are people on Reddit saying? What are your G2 reviews? And so on, and so on, and so on. That will then feed into what those tools know. But at the same time, business development also works.

You can go to OpenAI and Anthropic. And if there's good reason on both sides, like if the same it's kind of like the Apple playbook all over again. If you can help them promote the thing they're most interested in promoting, which is usually the thing they're working onright now, that might be enough reason for them to feature you.

And so founders and teams should work on both sides of that equation. Building a great product, but also don't forget business development. It is still a very viable and real strategy. We invest heavily in our relationships with the foundation model companies.

So that's on the investing side. And we do $350,000 to $500,000 checks, usually pre-seed and seed. Let me know if you're interested. We help a lot with product-market fit and growth and raising subsequent rounds and all kinds of weird and wacky stuff that I can't get into publicly, but the crazy stuff that happens when you're building a company.

And then your other piece is, what would I build if I were not building this? In short, I don't know because I'm so tremendously focused on making Superhuman work. But I can describe the process I would do. I wouldn't immediately jump into building something.

I think that was maybe a little bit of not really a mistake because Superhuman turned out fantastic. But

after I sold Rapportive to LinkedIn, I took about nine months off just kind of messing. But you know what I didn't do? I didn't cleanly separate vacation from messing around with ideas. I was kind of bouncing between having a lot of fun and then actually thinking really hard about what I wanted to build.

If I were building something new, I might just deliberately go and spend time with some venture firms I really respect. A16C did our Series B, for example. First Round did our Series C at Superhuman. I might go hang out with them and help them do whatever they need help with.

But at the same time, meet lots of interesting founders and just sort of start to get smart about where the opportunities are. I think deliberate learning is a phase that a lot of founders often don't do before they start a thing because they're interested in starting a specific thing.

But it is interesting

to sort of just open your aperture a little bit wider. I'll give you a specific example. In 2014, I was hanging out with Balaji at A16C. He's now left and working on his network stage project. But back then, he was a general partner at A16C.

And this was 2014. I was pitching in Superhuman. And he was like, "Yeah, this is really, really cool. But you should look at Bitcoin. I think that Bitcoin is going to be really, really big." I look it up and I'm like, "Yeah, whatever.

Crypt, distributed, finance. I don't really care," which was clearly my mistake. 2014 would have been an amazing time to follow his advice and to actually double-click on what he was saying. So maybe a culpa. I should have paid more attention.

But that's the kind of thing I mean. That's an example of go hang out with the smartest people you know. What are they interested in? What are they seeing? And just learn. Fully open mind, see what's cool, see what's going on.

I'd also say if anyone's looking to spend time learning, actually go and hang out with creators who are covering the space they are most interested in. They're normally at the forefront of innovation. They're covering a lot of these tech trends.

And normally, they know a lot of really cool people as well. So go and spend time in maybe a VC or a tech company and also go and hang out with some cool creators.

Yeah, go volunteer your time. Go help a VC with deal sourcing or analysis or vetting. And in return, you can get really smart very quickly.

Exactly. Well, hey, I'm a super fan of Superhuman. I should say super fan, Superhuman fan, one of the ways. And super excited to see where Superhuman as a group and the bundle goes next. And what a cool story.

And thank you so much again for coming on the show.

Absolutely. Thank you, Ollie.