Your Next Move - How to Build a Platform for a Market That Doesn’t Exist Yet

Episode Date: September 1, 2026

Leah Solivan of TaskRabbit and Joanna Strober of Midi Health discuss funding, infrastructure decisions, and AI opportunities that are rewriting the rules.

Okay, I'm going to say something a little controversial here. Right. So a few of my investors, every board meeting, they want to talk about AI. But fundamentally, I need to provide care to hundreds of thousands of women. And that is not AI. So I am pushing back on some of my investors. I'm like, yes, we are using AI. We're going to use AI.

But fundamentally, I am providing health care. And you need to let me continue to provide health care. Today on your next move, the new rules of building a platform, what it really takes to scale a tech business now. We'll be speaking with two founders of pioneering tech platforms. Leah Sullivan built TaskRabbit into the platform that helped define the gig economy. She sold it to IKEA and then started writing checks on her own. Now she's backing underestimated founders who are building the next generation of tech platforms through her firm, Precedent VC. And Joanna Strover was a venture capitalist who became a patient,

and then a founder. Her experience, being dismissed by the healthcare system led her to build MidiHealth in 2021, now a billion-dollar unicorn and the largest virtual care platform focused exclusively on women's health. Hi, I'm Mike Hoffman, editor-in-chief of Inc. And welcome to your next move, produced by Inc. and Capital One business. Sometimes the tools people really need don't exist yet. And today, we discuss what it takes to fund, build, and scale a tech platform when the vision is much bigger than the current market can see. Leah, Joanna, well, Thanks so much for having us. Thank you. We're going to talk a lot about AI and kind of where technology is going.

But first, I actually wanted to start by hearing a little bit about your company's origin stories. So, Leah, let's start with you with TaskRabbit. Yeah, absolutely. So I founded TaskRabbit back in 2008. I was living in Boston at the time. It was cold. It was snowing outside. And I had this 100-pound yellow lab that was out of dog food. And that's what gave me the inspiration, actually, for TaskRabbit. I was an engineer at the time. So I became really obsessed with all these emerging technologies, social, location, and mobile. And I thought, there's got to be someone at the store at this very second. I should be able to connect with them.

And so I quit my job four months later. I was working at IBM at the time, and I built the first version of the TaskRabbit site and got it launched in Boston. Amazing. And I understand you took money out of your pension to be able to fund it. $27,000 pension sitting at IBM. I don't even think they have pensions anymore, actually. But yes, I took it out with a penalty. And that's what I used to fund the original site.

Wow, that's amazing. And Joanna, with Midi Health, you were a venture capitalist and an investor, and then suddenly as you were like going through your own personal health journey, you thought, hey, there's an opportunity here. Can you tell us about that? Yes. So I was in my late 40s and experiencing a lot of symptoms that are now seen as perimenopause symptoms. I was having anxiety, weight gain, a lot of emotional issues and challenges.

and I went to a primary care doctor, and they basically said, take SSRIs, which are antidepressant medications. They told me to go to a therapist. They told me to take sleep medication, but said they were addictive, so I could only take them once in a while. And not once did hormones come up, not once. And so eventually I was referred to a hormone specialist, and I drove to this office over an hour away.

I paid $1,500, got the right care. And what I realized is that everyone needed access to this care and that we needed to, I needed to also quit my job to go start a company that would solve this problem. And the menopause space is very hot right now, but Midi was the first unicorn in the space. Is that right? Not just the first unicorn. I mean, menopause is hot now, yes. But let me tell you, when we first started talking about menopause, no one was talking about menopause. So we went to all these people and a lot of investors and they were, what is, why would you start a menopause company? What is the opportunity? It's really small. It's not interesting.

So, yes, menopause is now hot, but I can tell you, just a very few years ago, no one was talking about it. I love that. It's a small market, 51 plus percent of society, but sure. So I'm curious about that sort of investor education piece and as well, the sort of customer education piece. When you're launching a platform that really changes sort of the way that people behave as consumers, what does it take to get people to understand sort of the early idea? Let's start with you. So it really was both sides of the equation, right?

And when you're talking about menopause, many women actually didn't even know that they're in perimenopause. So we actually had to go out and tell people, look, this is a hormone issue. If you're in your 40s, you are likely in perimenopause. And we had all these women who said, no, I'm not. And it was such an interesting conversation. So we had to start educating the population. This is your hormones. This is a hormonal issue.

Every woman goes through it. So it was actually a very basic education that had to happen to explain to women that they needed specialized care. And that's really what it's about is understanding that women deserve and need specialized care. And what about with TaskRabbit? I mean, similar story, education-wise, because we were, you know, one of the first to help start the gig economy at the time in 2008. It was crazy to think you would jump in a stranger's car. That would be insane. You would never do that.

You would never invite a stranger into your home to help you. Hang show. Who you met over the phone. Yeah, that's crazy. That's insane. And so there was a huge hurdle around trust. But for me, as an engineer, I saw the opportunity to utilize social networking as a way to build trust between users. Because if you knew that Joanna used a task rabbit and had a great experience with them, you're more likely to trust them as well. And so that was kind of the premise of the education building. You know, we were talking before the show, too, just about raising money. And I wonder if you have these stories too,

but it was always difficult to educate investors on what task grab it was because many of the investors typically traditionally were men, we're not in charge of the household, we're not running errands, we're not outsourcing tasks. And so a lot of times I would get the, let me go home and ask my wife about this, or let me ask my EA if she would use a service like this.

And so there was an education piece that had to happen I was raising money as well. Did you find that with me? Absolutely. Yeah. And we started with menopause Tupperware parties, as I used to call them. So we actually went city to city to city, and we would go to a house purposely, instead of going to like a more corporate environment, we would go to a house and we would have our providers come and educate a group of 30 women about menopause. And we actually saw our original growth. You can actually see it from those parties, you can actually track.

And actually, I was telling someone earlier that at the original DC party, we had one of the most senior women in the Army there. And now we know that, like, a lot of women in the Army are all using MIDI. And it all stems from that first menopause party in Washington, D.C. That's incredible. Oh, that's amazing. So you have a background as an engineer. I'm curious, like, as you're thinking about, do you have a background as in technology?

I do not. I was not ever in. I was an investor, but not an engineer. So then as you're starting your companies, whether you have a tech background or you're bringing on technologically, you know, smart co-founders or partners, how do you start to think about like what does this platform look like? What shape does it take? Well, you know, it evolved. And what was really hard actually about building task rabid is the consumer behavior and consumer expectation evolved massively as we are building. When I started in 2008, eBay, Etsy, those. were the gold standard marketplaces. And so we originally launched with a bidding platform where you could say, I need help, and taskers would come on and bid for your job. But there was so much friction, it took way too much time. And, you know, very quickly with, you know, Twitter and Instacart and

DoorDash and Uber and all these things that were instant booking platforms, we realized pretty quickly we needed to go mobile first, mobile only. It needed to be an instant booking platform. And so the technology, I literally threw away entire codebases at least three times and started from scratch again at TaskRabbit. And that was really hard. I think we're in the same moment today with AI, to be honest, because technology moves so quickly. And the consumer, you have to keep up with a consumer and their expectations and their mindset. And you have to meet them where they're at. And if they're way over there and they've gone, you know, a hundred steps ahead of you, you have to figure out how to get there quickly.

And let's talk about Midi Health. How does it work as a tech platform and how did you build it? Yeah, it's so interesting. I'm loving listening to you because there's so many similarities. For us, it's about following the women. So when we started, we were a menopause, perimenopause company, and women would come to us to get specifically hormone care. Then we found out that 85% of them want to lose weight.

We're like, oh, we're a weight loss company. And then we find out that actually 50% of them don't have a primary care. care doctor. So we're a primary care doctor company. So essentially what we've learned is that we have to follow the women and follow their needs. And we built a tech platform that basically evolves with them. And we built this AI model essentially that basically enables us to add more and more care concerns on top of the basic system that we've built. And it really is following the women and following the needs that they have. So let's talk about AI then. Obviously, this is a moment when if you have a tech platform that you've built,

AI is changing the game. There's vibe coding. There are people creating new competitive tech platforms all the time or types of software. How do you think that founders who have tech platforms or tech companies, tech forward companies, should be thinking about AI today. Let's start with you, Leah.

Yeah, I mean, because I've lived sort of this product transition, I almost think if I were a founder today, I would prefer to be starting from scratch now versus having to transition my whole platform, right? Because it makes it really difficult when you already have a code base and you have an engineering team and a product team and then all the sudden...

And customers. And customers. Yeah, all of the sudden, AI is enabling you to do many more things much faster with fewer people. Organizations are transforming in front of our eyes. We are seeing it everywhere. But what I'm seeing right now as an investor. So I'm an early stage investor.

I do pre-seed seed investing. And I see so many interesting AI native companies forming that are starting to build on top of these LLMs. They have proprietary LLMs, proprietary data that they're building off of and creating really interesting, unique solutions to consumer problems that it just wasn't possible before. And a lot in health care. I'm seeing so much in healthcare because the data is there to support the intelligence to really provide solutions. And with Midi, how are you layering an AI? So I'm going to say that we are always going to be a human first company. We do not believe that Midi that AI is going to replace the need for caregivers and for expert clinicians.

But what we are finding is that the AI is enabling us to create superclinicians, I guess, is the way to say it. Or they are being given superintelligence with the AI so they can do significantly, better care than they ever could do as a sole provider, using AI to augment them. But we fundamentally do not believe that the providers are all going to be AI. We think it's still going to be people. We think that people is a really important component of this. When I talk to you, I got to look at you and I got to understand how you're, are you really remembering all those medications that you were asked if you were taking? Is it possible you forgot one? Is it possible you forgot that your mother actually had breast cancer. You may have tuned it out and not wanted to remember that.

And you might not have told the AI, but actually she may have had a biopsy that gave you that had breast cancer. You need to know that. So we feel very, very strongly that the human component is not going away. But we are incredibly excited to use AI to augment the providers instead. And both of you are based in Silicon Valley. AI is obviously transforming Silicon Valley in so many ways. It's obviously a gold rush for many companies. I think it's also created some anxiety within organizations about sort of the future of tech jobs. And I'm curious, how do you see that manifesting itself within organizations and how are you managing it? Leigh, let's start with you.

I mean, as an investor, I'm certainly seeing sort of the democratization of being able to create, create products, create apps, without having to have a deep technical engineering background. Which I think is a very good thing. I also agree with you, though, Joanna. like keeping the human in the loop, keeping the empathy. And what I am excited about is the opportunity for humans to really be there to provide what we are uniquely good at that the AI cannot replace, right? And that's the power. It's like let's, you know, hand over all the commoditizations and all of the coding and the things that we as humans may not need to do anymore as much. And I think I have a whole theory, too, around, like, what kids should be studying in college right now?

You know. What's the short answer? The short answer is not engineering. I actually think, like, get a liberal arts education, like, study the arts and humanities and English and, like, all the things that humans are unique at and build our empathy and build our trust with each other. And let AI commoditize some of those other things. So as an investor, I'm seeing massive companies being built with under 10 people.

New products, getting to market in, like, record speed. 24 hours, 48 hours, and then evolving from there based on usage. So it feels like the Wild West. It does. And I think that's where the fear comes from because it also feels like we don't have control, right? And we don't have a handle of where this is all going because the technology is so powerful. It sounds like an echo of the gig economy, the beginning of the gig economy, when there were trust issues in different ways. I mean, I got to love that you brought that up because I do feel like this, this.

moment in time feels to me, personally, like it did in 2008, when those three technologies were emerging, social location mobile, created the gig economy, launched, you know, billion-dollar businesses, it feels like that same moment from a technology standpoint today. And Joanna, how about you? How do you sort of balance the AI can be an accelerator for the business, but also it can create anxiety within the organization? How do you manage those two as a Yeah, so, you know, we do have a big company. We have a lot of incredible care providers, and I do. I talk to them all the time about using AI to augment themselves, to become super care providers. And what I think is going to happen, which is insanely exciting, is in the United

States, we have such a lack of caregivers. We don't have enough primary care. We don't have enough doctors. We don't have enough specialists. And what I think AI is enabling is that primary care and specialty care are actually going to collapse. And you're going to have providers who can do so much more than they could ever do before. The collapse is okay. That sounds like a scary thing. Yeah. That's exactly right. That's my point to them is the collapse is enabling you to do your work better. And that the models that we are creating about women's health are going to enable significantly better care than was ever provided before. It's actually a really good job security. It is not going to take away your job. Instead, it's just going to enable you to do significantly better care and less

time, which I think is really exciting. And how do you think about then within an organization of a tech-driven company, the engineering team, the product team? Are there roles that you see becoming more relevant, becoming less relevant? And what's your forecast for that? Lea, let's start with you. Yeah, and I think new roles emerging as well. I mean, I just came from a really interesting conference in upstate New York. And a lot of the conversation was around how. how do product and engineering teams evolve in this environment? And, you know, someone was bold enough to say, I don't think the role of a product manager exists anymore.

Because I think that it combines collapses in a good way with marketing and customer service and engineering and operations. And then all of a sudden, there's one person that can do a lot more. I think new roles, too, that I heard discussed, were around how do you create a role in the organization that holds all of the routines and all of the prompting and all the templates and all the structures that are run as part of the AI, right? Because you still need that human that is creating, you know, new routines,

new operations is checking to make sure they stay in the loop and on track. And that's a role that we haven't had traditionally in product and engineering team. So I think there's both opportunity to do more. There's opportunity for new roles to be formed. And, you know, there is going to. to be this evolution of the traditional org chart in product and engineering that may not exist in the same way any longer.

I'll give you an example exactly what she just said. So we have a marketing team, right? We have an engineering team and we have a product team. We realized that there was a lot of things that could be automated within the marketing team that AI is now enabling. So we didn't need a product person for this. Instead, we sent four deployed engineers

to the marketing team to spend a month basically looking at all the different procedures and policies they were following and figure out where AI could make things more efficient. And it's remarkable, just in looking in the marketing team of all the things we could automate. And so just the marketing team working with these engineers created actually an incredible new system for us that's enabling us to be much more efficient, create ads much quicker, test them much faster and get much faster feedback loops. And no one's losing their job.

Instead, they can all actually do much more exciting, interesting things. That's interesting. You know, tech platforms have long thought about product roadmaps, but that feels like that's, how do you come up with a product roadmap and stick to it in a world that's evolving this quickly? Is that something that you think will evolve or go away in some way? That's a great question. I mean, I remember getting out the big whiteboard and the big conference room and the

Post-it notes and the whole team gathering to set the product roadmap for the next 90 days. And, yeah, now I think, like, wow, you're getting customer feedback so quickly, and you've got to move agilely. I mean, I'd be curious, you're living it right now. I mean, what are you seeing? Yeah, I mean, it's changing much faster, right? That's exactly right. You know, the three-month thing is, like,

we're finishing a much faster period of time. Actually, you have to re-we're re-looking at how fast we can get things done because they can get done so much faster than they were before. So now we can actually look at creating a lot more an agile. I guess an agile might be the right word for the roadmap. And it does iterate. and what we're loving is assigning the engineers to the people who need their work automated and just saying, go for it, and just start collaborating and figure out what parts of your job

can be done faster and more efficiently. Does it change the role then of managers, either in technology departments or non-technology departments? I think a little bit in that, you know, I think, again, managers, I mean, managers have to be great with people, right? Like, that's the key thing, is being able to do. develop and coach a team, particularly if you have a big org, right, and you have younger folks that are working their way up. You want a manager that can really groom those people. But I think with AI right now, too, it's getting people that are able to kind of expand what they're doing

and having managers maybe that can help that development and have that coaching and help with that expansion. You know, when you start a company, you want to find people that can like play all the basis, right, and play all the roles. And it almost feels like now with AI, it's like everybody can do so much more. And so finding those people that feel like they can do all the different roles or at least are comfortable being agile and iterating and leveraging new technologies is really important. But I want to get to what Leah said. In the end, the people stuff is still the hardest, most interesting challenge. And you still need managers for that. That has not been eliminated. I think there's this idea that just all these agents are out there and making everyone get along.

That is not what's happening. People have feelings. And they want to talk about them. And they also have different priorities and they have different things they want to work out. That has not been eliminated. And so I think it's actually really important to think about what the agents can do and then what people need to do. And yes, there are still need for managers because you're still going to have a disagreement on what should be done tomorrow. And you're going to have to talk about that.

The agents are not going to fight that out. And the agents will do what you decide, but you still need the people to make those decisions. But you know what's interesting, too, is when you think about the development stack, I saw a company the other day that had created this product that was detecting on top of GitHub. So all, you know, developers check in code. It goes to GitHub. Now agents are the ones that are checking in code. And it's like thousands of pieces of code are being.

checked in by AI. And so the managers now need to actually report up the chain on like, what happened? Like, what have we actually created? What is the software we've created? What have we produced today? Are there regressions? Are there problems? And so the company I thought was pretty interesting, which was managing all of the agents that are checking in code, but extrapolating what they have done in a way that any human, you know, not just a technical human, not just a technical engineer, anyone, you know, sort of up the chain to the sea level, can understand what's going on. So maybe you need managers for the agents now, too.

I don't know. It's interesting. Inc has actually done some stories on the notion that the manager of the future, and the future is like tomorrow today, is a person who can manage both people and agents, right? And that you might spend 50% of your time on each. Right, right. Yes.

So I'm curious, like, platform businesses are obviously built on infrastructure choices, and infrastructure choices are changing as a result of AI. Looking back, at least, the beginning of your journey with your companies. Was there an infrastructure choice that you made that was really right on the money and set you up for success? Or was there one that was like a problem that it took you some time to work out of? Well, I mean, I told you, we threw away the code base three times.

So we had a lot of infrastructure choices that, you know, I wouldn't say it went wrong, but it was just the consumer change so quickly. So, you know, very clear one was being web-based, web-first. you know, bidding platform. And then with a million clients, 60,000 taskers open in 44 markets, we have to change everything to mobile first, mobile only. And by the way, the engineering team is building for the web. They're not mobile engineers. So then we have to change the whole org, too, to be able to build mobile. That was painful. It was a painful transition that we had to do, but it was necessary for the business, not only to survive, but actually thrive and get to profitability and we sold to IKEA and it all ended well. But I would say there was 12 to 18 months in there where it was a real grind and it was not fun working at TaskRapid.

Okay. And, you know, I remember having to lay off a big chunk of the product and engineering team because the skill set didn't match any longer. And it was brutal and I felt like a failure and I felt like the choices I had made around the infrastructure from the beginning were not scaling. not holding up. But that is the way businesses grow, and, you know, we made it through in the end. And Joanna, what about with you admitting? Yeah. So we made actually some very good choices at the beginning. We didn't want to build the infrastructure. We decided to build on top of someone else's infrastructure so that we could focus on the customer experience. If we had tried to build our own emergency, you know, our own record system to integrate with all the hospitals. It just would have been this huge engineering team that was not focused on the end user.

So we really put all of our engineering resources into working with the end user and then using someone else's platform. Now, of course, with AI, we're actually able to take a lot more of that and own it ourselves. So we're slowly and surely morphing the company to own more of the technology rather than building on someone else. But our first choice was that we had to integrate with other hospital systems. And if we built that, they would never have trusted us. So the whole company is built on trust.

We needed to be able to integrate with everyone else. And so those were the initial choices that we made. I'm curious, how is the conversation between founders and investors changing as a result of the emergence of AI and all the trends we've just been talking about? Okay. I'm going to say something a little controversial here. Great. So a few of my investors, every board meeting, They want to talk about AI, right?

And they come and they're like, we're using AI. Are you laying off enough people in order to put on more AI engineers and are you doing all with AI? And we are trying and we are. But fundamentally, I need to provide care to hundreds of thousands of women. And that is not AI. And so there's a bit of a disconnect between investors who are like, how fast can you use AI?

And the patients, they don't care if I use AI. They want me to give them really high quality care. So I am pushing back on some of my investors. I'm like, yes, we are using AI. We're going to use AI. But fundamentally, I am providing health care. And you need to let me continue to provide health care. Absolutely.

Well, you say absolutely. But they're like, well, how fast are you using AI? Are you doing AI? Are you doing AI? I mean, I think investors, and you probably know this, they operate out of fear. Yes, correct. They operate out of fear, okay? And I've been a founder and I am now an investor and I've seen it on both sides.

Same for you. And so I think that investors are very scared right now. They're very afraid. They don't know where this is all going. Are they afraid because they've put a lot of money into companies that might not now mature the way they wanted? Or are they afraid because they don't know where to bet? Yes. All of it.

I love it. I love it. Okay. Honestly. Yes. And I think as they're assessing new opportunities, because I say, see this with co-investors and syndicates that I'm doing, if a founder right now does not have a strong AI story, they're not going to get any attention from investors. And in the end, right,

like a business like yours, what matters is the patient care and the human experience. And so I think it's hard to go to investors right now with a story like that without the AI component because they're so afraid of missing up. It's all FOMO. They're so afraid of missing up. Yeah. One thing that one of my investors did that was actually insanely helpful is they set up some show and tell for my engineering and marketing team with AI native companies. Oh, cool. And it was very helpful. So I basically sent my teams to those companies, and they spent half a day learning from what a more AI native team was doing. And then they brought that back into the organization.

So that was actually very helpful. And learning from, you know, learning and learning and business. really seeing what is everyone else doing out there? What are the best practices and then bringing those back to the company is very helpful. And then I have an AI tutor. So I have these. Where'd you find this person? I have this team actually of young 28-year-old business school students at Stanford,

and they are tutoring me. And these are real people, not agents. They're not agents. No, we now meet on the weekends and I get tutoring to make sure that I can build my own agents. I need to be, I need as a leader of my company, I need to be AI forward, I need to be making my own agents, I need to be really ahead of the curve so that I can show the whole company. Can't ask the whole company to be AI native if I'm not. So I hired some real people, however, as tutors, and they came in and they worked with me

and helped me automate my life. So now I have basically a chief of staff that is an AI agent who briefs me every morning. And so I've learned, you know, I need to model this to my company. And so I'm learning so I can show everyone else how to do it. That's great. How are customer or user behaviors kind of for the average user, and who's an average user? That's sort of hard to define.

But how are the average Internet user or the average digital users behaviors changing as a result of AI? And are their expectations rising for companies? I think so. I mean, just like the consumer behavior changed when everything became instant, right? When I first started a task grab it, it was like, I'm going to connect real people

in the real world to get real things done, dot, dot, dot, dot, in real time. And that real time was the piece that really kind of nipped us afterwards. They were like, oh, my gosh, we forgot about that. And I see the same thing happening now because consumers don't only expect to be able to find and search knowledge like they did with Google.

they expect that knowledge then to be able to be extrapolated and summarized and pontificated on and fed to them in a way where they can very quickly understand it within a couple of seconds, right? And think about this next generation. Like our kids are now growing up in this world where everything is not only instant, but knowledge and perspective and insights and just they just have so much more at their fingertips. That changes how consumers interact with any company, with any business and how they expect to live in this world. I think that's true, but we are also living in Silicon Valley, and we have a unique vantage

point there. So we actually have the opposite problem where people are saying to us, was that an AI responding or is it a real person? So you have to get back to health care for a minute, right? People are worried that the AI might be responding as opposed to a person. And so there is fear among consumers of AI. And so there's a balancing act there of getting them to be comfortable with it, explaining that for anything that's actually medical,

it is still a person responding. It is not an AI. So we have to actually balance that for the end user on what they need. I also think sometimes AI can be frustrating. Like you're just looking for a movie on a streaming platform and they give you AI answers. And you're like, no, no, I really just want the movie. Just want to find. Yeah, yeah, yeah.

Yeah. You don't need to explain it to me or things like this or whatever. That's right. That's right. I'm curious, you know, with any tech platform, there are multiple stakeholders. They're the users. There are investors. Often you have multiple types of users, you know, medical systems as well as patients.

How do you balance all those stakeholders in a time, especially when technology is evolving so quickly? I focus on the end user first. So for me, my core, I need to make sure that every woman. who comes to us gets high quality care that has to be first. Everything else has to be enabling that. And so that is my core value, then everything else is around enabling that. So we can use AI all over the place to enable better care, but it helps me to have my north star of what I'm trying to do and everything else falls from that. If you lose that, if you think I'm building a tech platform as opposed to I'm building a care platform that gives women better care, I could get lost really fast,

thinking about all the great technologies that I could build, but are they actually fundamentally creating a better care experience? And that is what we focus on. But are there ever conflicts between what providers, let's say, want? Oh, yeah. And how do you like referee those? No, there are conflicts all the time, right? And what we really need to do is we need to focus on what is really important for our end user.

And so as long as I'm focusing on that. And then honestly, actually, I take that back, focusing on what's important for the end user, But then I have 700 care providers in my platform. I have to make sure that they're able to provide great care to those end users. So we do have two audiences. We have the care providers and we have the end user, and we have to balance their needs also. And with something like TaskRabbit, right, or all gig economy companies, there's the gig worker and then the end orderer.

And how do you manage that? There's clients, there's taskers, there's investors, there's employees. I think in any business, too, if you kind of scope it out a little bit, too, it's a around how are incentives aligned and making sure that incentives are aligned in the right way for the right stakeholders. And I can imagine there's times where incentives are not aligned. And as an investor, too, I've seen companies where they're trying to solve problems where, for instance, incentives may not be aligned. I'll give you an example. IVF. Someone goes to an IVF clinic, right? In those clinics, how do they make money? They make money on the number of times you go

through an IVF process. Those incentives are not aligned with the patient, right? Because they're making more revenue, the more times you fail. And so there are companies out there that are trying to break down those incentives and realign things using AI so that everyone can move in the same direction. I think it becomes a real challenge to solve when incentives are not aligned. Yeah. But I agree with focusing on, like, who is your primary user in general? For us at TaskRabbit, it was like the taskers are making money. They were actually our primary user. They were the ones we were serving. We were enabling them to, you know, find work and make money and have a different way of working. And so we always focused there first, and then everything kind of fell off of that.

I guess that's my point, though. It's not AI for AI's sake, right? Like, I think that we have to remember what you're trying to accomplish, and then you're using AI. It's a tool to get there. But it's not just AI for AI sake, it's because you're able to do something better, more efficiently with higher quality using AI. Now, as founders who've started billion plus dollar companies, what would your advice be to a young startup company with a founder who's just getting started in the tech space?

What mistakes do startup founders often make, whether it's building their technology or articulating their value proposition or selling it to customers or selling it to investors? I feel like I made so many, mistakes. It was starting at 28. I mean, I look back and I'm like, I don't even know how I got this company off the ground. I was an engineer. I did not have an MBA. I didn't know how to fundraise. I don't know. As I look back, I think, you know, storytelling, being a great storyteller because

you need to raise money, you need to hire people. I mean, basically your two jobs as CEO is don't run out of money, hire people better than you. That's like how I thought about it over the time that I was running TaskRabbit. But when I think back early on, like I really have. I really have. had to nail that vision, that storytelling, and then figure out how to fundraise, how to pitch investors. And there's so many dynamics, right? In venture and investors and now being on the other side of the table, there's so many things I understand now that I wish I had then. You were an investor first. I'm curious if that, you feel like that helped you as a founder, right? Like build maybe a better business faster with investors. I think I understood, every

investor has their own objectives, and you need to understand when you're talking to them, what are those objectives? And it's both objective as a person, and it's objective within their fund and where the power dynamics are within their fund, and then also where they are in their fundraising path, right? There's so many parts to the investor story. So once you understand that, you kind of understand that they're all people too, right? They're not scary. They're just people who are trying to do their job. And their job is to make money for their investors, and my job is to show that they can make money investing in me. So like that that actually becomes very clear

in when you understand those dynamics. Yeah. That's great. Can we just talk about when you were raising for Midi the first time? What did you then look for to match up with like, you thought like this is a good match for me? So I definitely wanted people who cared about the problem that we were trying to solve.

And the problem is better health care for women. And so I was looking for people who cared about that and who understood that this could be a really big business if we got it right. So I actually turned down money from someone who said to me very nicely, I don't think your business could be more than, you know, two or three hundred million dollars. So, you know, we're kind of underwriting it to that. And I said, no, this can be a multi-billion dollar business. So you are the wrong investor for me. So it really was understanding their priorities and my priorities and making sure that they understood the really the big exciting company that I thought I was

building and did they want to jump on for that ride? That's great. So a question about things you're seeing out there in the world. What's a feature or a piece of technology or even a platform company that you've seen that you think, oh, that's really cool. I'm interested in what they're doing. I'm currently obsessed with the new town assistant. Have you used this? No. Oh, my God. It has changed my life. I've been using it for two weeks. Yuri Kim, led the round, their series A. Yeah, I've completely, replaced my current admin, and I'm doing so much more. I'm so much more productive. It's like everything I wanted when I went to chat GPT, but like just couldn't wrap enough. There weren't enough connectors already, so I'm really loving the town assistant. So honestly, I am just

incredibly excited about what AI is doing for health care, because we really are creating providers who can give so much better care than they were able to before. The amount of information they can get using just the standard LM models is so good that they can give so much better care than they've been able to give in the past. And so when you think about going to a doctor now who only has their own knowledge, that's going to be so antiquated, right? Instead, you're going to have access to the whole world of knowledge. And so I just think every person in this country, and honestly in the world,

is going to get access to higher quality care. And the care you get is no longer going to be geographically restricted. Historically, the care you got was restricted to whether you could drive to a doctor who was going to give you their own knowledge. But now they have access to so much information. So we are all going to get so much higher quality care, and that is really exciting to me. Well, that's great. So obviously the name of the show is your next move. So I'm curious for you and your business, what is your next move?

Joanne, let's start with you. So for Midia really is about that collapsing of specialty and primary care. What I believe is that we're going to create providers who can give so much more information to patients. They are going to be able to give you information about cardiology and endocrinology and your thyroid and your migraine. They're going to be able to have one person who can take care of all of these issues, and that is very exciting to us.

Great. And Leah, what's your next move? Well, I am running a fund called Precedent VC. I've become sort of obsessed with the idea of breaking precedent. So I started a podcast called Breaking Precedent, and I have a book coming out next summer under the same name where I'm interviewing leaders across industries like Joanna here. We are interviewing next week.

We are. About how they're shifting their industries. I mean, we're living it right here, right now. And then giving readers a framework, if they're trying to break through something, how to do it. There'll be some task rabbit stories in there, too. Great, great. Well, Leah, Joanna, this has been a great conversation, and we'll rejoin it in a minute.

But next. Now, today, the People Playbook tackles leadership at scale. As a business grows, leaders must shift from doing things themselves to building a bench of talented, capable managers who can help the organization scale. But when you've had such an important role in growing your business, it can be hard to let go and delegate key roles to someone else. For more on this, I'm joined by Caitlin Hargrove, head of operating office and chief of staff Capital One Business. Welcome, Caitlin. Hi, thanks for having me. So why is it such a challenge for leaders to build strong management teams?

Often the best performers are elevated to leadership roles without any direct conversation about whether they'd like to manage people. People leadership is a craft that can be mastered, but you have to provide coaching and development to new leaders. Each leadership team requires a different skill set. So be thoughtful about balancing individual contributors. and strong organizational leaders. Now, delegation seems to be one of the hardest skills for senior leaders to master. Why is that? Yeah, many executives came into their roles because they were really excellent at execution.

So delegation could be seen as a threat to their identity. Letting go of that can feel like lowering standards. But refusing to delegate CAPS organizational growth. The discipline is not in doing less. It's in building others to do it well. Delegation also takes time. It's a go-slow-to-go-fast principle that can be hard to prioritize when you're also running 100 miles per hour.

And what do leaders typically get wrong about delegation? Delegation is often mistaken for simply task transfer when it's actually capability transfer. You can't hand off work without context, authority, or boundaries because that just creates confusion. Effective delegation includes information and autonomy. It helps the manager build judgment and expertise rather than just checking the boxes on completing tasks.

Okay, let's break down the elements of effective delegation. How can you be sure you're giving your managers the right guidance? This is where there is definitely an art in a science delegation. The science consists of explaining the strategic why behind the ask. At the same time, you must make the desired outcome clear while also providing support along the way. The art isn't ensuring you resist the urge to control the how. Managers are more engaged and empowered when they can take ownership of how a problem is solved.

In other words, focus on the outcomes, not the process. And are there some practical steps that? leaders can take to get their leadership team up to speed quick? Yeah, this is a great question. Clear role, scope, and expectations are of course table stakes. Assuming you've got those defined, the next step is ensuring you can articulate your strategic vision. This helps quickly calibrate new leaders and point them towards the goal. Caitlin, thanks so much for joining us and for talking about delegation and effective leadership.

Thanks, Mike. Thanks for having me. Now it's time for the debrief, where we'll take a closer look at the strategies our guests used to build and scale their tech platforms. Thanks, Mike. I'm Rebecca Dyshinsky, Inc. Senior Editor, and I'm here with Angela Lee for a conversation about the strategies behind platform building. Welcome, Angela. Excited for the conversation. So both Leah and Joanna had to educate the market before they could successfully launch their companies. What's the most important thing that a founder needs to know about building a market that doesn't know it needs you yet? I always tell founders to ask themselves two questions. The first is, are you solving a real problem? And the second is, how do you? how quickly are you able to educate them that your solution is the right one? So if you take TaskRabbit and Middy Health, they have different answers to those.

In the case of TaskRabbit, the problem is very easily apparent. I don't want to build this furniture on a Saturday. I don't want to go run this errand. And the solution is also very easy to immediately understand. In the case of Midi Health, it's a little bit different. Yes, a woman might know she's having symptoms, but she might not know that it's due to hormonal imbalances. And so not only does Midi Health have to educate the patient,

they have to educate the provider, the insurance companies, in some cases the regulators. And so her journey was much more difficult in building that marketplace. But the good news is once you figure it out, you have a really good competitive edge against other people trying to come into the space. And both founders talked about trust as a foundational challenge. How should founders build trust at scale? Yeah, so trust is something you really have to be thinking about from day one.

And not only for the TaskRabbit, are they trusting a stranger to do something, but maybe even coming into their home. And for Midi Health, they are trusting somebody with their bodily health. A company that not everybody might remember is called Friendster. And Frontster was a social media platform. And their product was really good, right? They were actually before Facebook. And even though their product was really good, they actually grew too fast and didn't think about trust. And what happened is their server started breaking down and people would go on the site and it wouldn't work.

And every single time that happened, they lost a little bit of trust. So we always ask founders don't prioritize growth over trust building. So as far as infrastructure goes, Leo was candid. She rebuilt the Codex three different times. Joanna took a very different approach, and she decided to build on top of someone else's existing infrastructure. What are the advantages and disadvantages of each strategy? So I think the first question you have to ask yourself is what is critical to your company? I think a mistake that some founders make is they think they have to build everything.

So if you are building a clothing company, you do not need to build the shopping cart. Shopify has done that for you. And then the second thing is in my class, I talk about this term technical debt. And technical debt is where you build your initial product really quickly because you want to get it out to market. The problem is the longer you wait to build the good version of your product, the more technical debt you occur because you have more customers, you have more features that you've built. So I want to talk about AI. Joanna used the phrase superclinicians, meaning people augmented by AI. rather than being replaced by it. I'm curious in a platform business,

where does AI help people and where does it start to hurt the organization? You really mailed on its head the idea of replacement versus augmentation. And so I think you have to ask yourself, does the human overseeing the AI know that function really well? And so an example might be me as an investor.

I can leverage AI today to say, tell me about the market size of this company, tell me about the competitive landscape. And that's great. The AI can give me an answer. in seconds. However, behind it, I have looked at 20,000 pitch decks to, for me to judge, are these answers right? Are they wrong? Are they missing a category of competitors? And so artificial intelligence augments the skill set of the human. And so what I would really encourage

founders to do is to put people in those departments who really have that experience, that lived experience, and then use artificial intelligence to support them, not replace them. We heard that investors operate out of fear of missing out right now, especially when it comes to AI, does every founder need to have an AI story? Venture capitalists are incentivized in a really weird way. If we invest a million dollars into a company and we get it wrong, the worst case scenario is that we lose a million dollars. However, if we miss out on the next Open AI or the next Anthropic, we can lose out on $100 million. And so there is so much investor FOMO, and AI only amplifies that. That said, a lot of founders are like, well, then I'm just going to put AI into my pitch deck,

and I really encourage founders to solve the problems that they know well. And if AI does support that, go ahead and put it in. But if you're just throwing an AI there as a buzzword, we're going to see right through that. What should a founder who wants to build a platform company know right now? I think something to recognize is that going forward technology might not be the differentiator, but what is never going to go away is deep, deep customer understanding. And so people keep talking about the pace of change coming with AI. But the thing that I think cannot be replaced is living the customer's problem. We don't know what technology is going to look like in five years. we don't even know what customers might want in five years.

And the really great founders, like Joanna, like Leah, they are able to anticipate what might happen. And then even if they don't exactly get it right, they can very quickly adapt to whatever the future might bring. Thank you so much for being here, Angela. Thanks for having me. Back to you, Mike. Now, one of the goals of the shows

to connect our audience with founders like Leah and Joanna, and so it's time for audience questions. And our first audience question is this. Obviously, when you're building a technology company, it scales online or in the cloud, but actually the business expands geographically. So are there any learnings that you have from when your company was expanding into new markets?

Let's start with you, Leah. Yeah, we were very much a geographic expansion, city by city by city. Every city was different. It was actually an interesting challenge every time we opened a new market, particularly when we got to international markets. So imagine a market like London in the UK,

very different types of jobs, very different clients that are posting there than, say, you know, San Francisco, California. So, yeah, we had to basically learn new categories, find different taskers that had those skills and backgrounds in every single market we went to. So it's kind of a fun challenge. I once used a tasker to stand in line for a restaurant. Really? I love that. Yes. Joanna, what about you when you move into new markets? Yeah, it's actually really interesting. So women in the United States are actually not all the same.

It turns out that a woman looking for care in New Hampshire actually has very different interests in what they're looking for than someone in Los Angeles. So there's a lot of cultural competencies required on race, making sure that the providers understand the differences in race and how you might need to treat differently. But it is also the case for geography. And so people in L.A. are asking us about peptides. People in New Hampshire are saying, I only want to use supplements. There's a lot of differences that we need to train that, too.

So our next audience question is when you're raising money for the first time, what are two or three things you should keep an eye on, which you make sure is not in the term sheet or in the final deal that might cause problems down the road? I'm going to say it's about understanding their objectives and what their personal objectives are and what their fund objectives are and making sure that you're aligned. I'm really fortunate to have long-term investors who view this as a long-term opportunity as opposed to a short-term opportunity, and that really does change how we work together. I think a question a founder could ask to that point that I wouldn't have known to ask at the time is when you're talking to investors, of course, align on values, vision, et cetera. But ask them where they are in their fund cycles, right? Where are you in your fund cycle? You know, what percentage of the fund do you have left to deploy? How much do you have in reserves? Meaning, do you have money that you're going to invest in me later? Or is this one check? And are we done? So I think to Joanna's point, like, those are some questions you can. can ask an investor to kind of get at that understanding as well. Yeah, it's really interesting because I didn't even think about understanding that. Right? There was one really big fund that invested in us.

And then it was the next round. And I just assumed they're a big fund. Of course, they're going to do their piranha. They're like, no, our way of doing things is we only invest in the first round. And I'm like, maybe I should have asked you about before. Right. It leaves a big gap for me to know, right? It did.

And I didn't even think to ask. And then we got into the fundraising process. So it's like, really? you're not investing. So I should have been smarter about asking that earlier. And one more audience question. When you're creating consumer-facing technology, what's the hardest part to get right? The closer you are to the consumer and talking to them all the time and understanding what they're looking for is just you just have to do that every day. And I think if you don't, you're going to lose them. Do you still have the Tupperware parties? We do. We still have Tupperware parties. I still

help some of my house, actually. We invite neighbors, literally neighbors, to come talk to us about menopause. Leah? Yeah, I mean, I would say along the same lines, you know, doing that customer development on what does the product roadmap look like, don't make it up yourself. Don't, you know, be the product or engineering team or person that says, like, I know what to build and I think this is what customers want. And if you can't sit in those Tupperware parties with your customers, which is amazing,

you know, bring in your customer service team. sit, put the product and engineering team with the customer service team for a day and just have them listen to what sort of calls come in, what sort of feedback they're getting. And that's how you build and follow your customer. Leah, Joanna, thanks so much for being with us today. Thanks for having us. Thank you. And thank you for watching.

We hope today's conversation helped you to better understand how to grow a tech platform into a true launching pad for your business. And join us next time for more industry experts, candid conversations, and the strategies you need to make your next move. For now, I'm Mike Hoffman at Air and Chief of Bank. Thanks for watching, and we'll see you next time.