Starting & Scaling PillPack To $1B Acquisition With TJ Parker | EM Group Chat #196

Welcome back to Everything Marketplaces, where we talk about founders and leaders from some of today's top marketplaces. So this is episode 196, which is a really great group chat, as you said, with TJ Parker. TJ previously started PillPack, which scaled to become one of the largest online and direct-to-consumer pharmacies. It was acquired by Amazon in 2018 for nearly $1 billion. TJ is now a general partner at Matrix and also helped start General Medicine, which is a one-stop shop for healthcare. So this is a great chat with TJ. We're going back to the founding story and first steps of starting PillPack. Some of the early challenges TJ walks us through, how PillPack evolved and scaled, shared about the acquisition by Amazon, some of his key learnings from PillPack, lots of tips for founders building in healthcare. We got to learn more about what TJ is building with General Medicine now. We also had a great group Q&A. I really enjoyed this conversation. You're going to find it a great watch to the end. TJ, so you've of course been a highly requested group chat guest, you know, for a while now and have paved the way for many of us building healthcare marketplaces. You know, so I want to start off by saying huge thanks for taking the time to join us here today in advance. And we have, you know, a lot that we're going to dive into with all of your experience, you know, starting and scaling PillPack and also now with building General Medicine. But I thought it'd be great if you can start off by briefly sharing a background for those here that might not know you.

So TJ Parker was the co-founder and CEO of PillPack, now one of the founders of a company called General Medicine and also a partner at Matrix Ventures. So for context, kind of grew up in and around pharmacy, trained as a pharmacist. And then started a company called PillPack back in 2013. And the vision of that business was really to make managing pharmacy much easier for customers that took multiple chronic medications. So we would sort and package them based on when they took their meds, so they'd get a box in the mail once a month with all their medications organized by date and time instead of getting them in separate, you know, vials that they had to then sort out and organize. And then, you know, probably the more valuable thing for customers is that we would do all the heavy lifting for them. So we'd manage all their prescriptions and renewals with their doctors. We'd deal with all of the issues with insurance companies. We'd effectively take what was a kludgy, like complicated experience and make it behave like a more traditional consumer subscription business, which took a lot of kind of backend administrative work. And that business was independent from 2013 to 2018. It raised about $120 million in venture. We sold the business to Amazon in 2018 for $1 billion. And then at Amazon, built out the Amazon Pharmacy business, built the Amazon Clinic business, which was effectively a telemedicine marketplace that any provider could participate in. And then left Amazon in 2022, and we started this business, General Medicine, in the fall of '23 and just launched that about 2 months ago. And that very much is a broad-based healthcare marketplace where you can come to us for any product or service in healthcare, whether it's Pharmacy or labs or imaging or seeing a specialist or anything you need in healthcare. We've organized these services as effectively a product catalog and a very e-commerce-centric experience where it's traditionally these kind of bifurcated and complicated services that are hard to navigate for consumers.

That's an incredible background. A lot that we're going to get into here with not only PillPack, but also general medicine. But, you know, kind of going back to PillPack, you know, in the very beginning, you know, what was the founding story and, you know, what were the first steps that you actually took to You know, in many ways it was pretty derivative of a business that my dad had built in the early 2000s.

So he found these machines that did this like sorting and packaging. That was a classic kind of family business. He was selling mostly to assisted living facilities and long-term care facilities, but the same kind of core packaging offering. And so the, the sort of impetus for PillPack is, can we offer something very similar for consumers that live at home? It was sort of a very simple idea. I think, you know, as we built that business, it was me and my co-founder Elliot, who's sort of the technical half of the business, classically trained engineer and built all of the infrastructure, the product, the tooling, all that stuff. And we had met initially at this like entrepreneurship event at MIT. I was at pharmacy school in Boston. He was at Sloan getting his MBA and we met there maybe in 2011 or 2012. And I kept poking him with this idea, like I had the idea and he was kind of like, I don't know, I like, I don't really get it. Like we're starting a pharmacy. That's a weird thing to do. And it was, it was quite odd at the time. But we worked on it at this hackathon for like a weekend together and it was like enough time for him to kind of wrap his arms around it and get excited about it. And we won the hackathon. We applied to Techstars maybe a couple months later. So the fall of 2012, we got into that. We raised our first little amount of seed funding like early 2013 and we're kind of off to the races. So kind of one of these classic stories where we were both sort of in the right place at the right time. And even when we started the company, we were like, we were friends, but we weren't like super close. And it couldn't have been a better partnership. We're like, we're so different, but so kind of symbiotic across every possible vector. So it was an awesome partnership.

Yeah, that's really cool to hear about. And you mentioned it's, you know, somewhat kind of like basic idea, right? But, you know, really challenging to actually execute on, you know. So what were some of those like early challenges that you initially kind of faced?

Yeah, I mean, in the early days, and I think probably comparable to the sort of average founder, like wrestling through this stuff, like it was not obvious how you acquired customers, right? Like that was the sort of fundamental bet investors were making is like, yeah, we feel like you probably understand the product. Your family's been doing something similar for a while. You like roughly understand the shape of the P&L and the unit economics that like they sort of gave me credit for all of that given the background. And the bet we were all making is, can you actually acquire customers at scale online? Right. That was sort of the fundamental bet. And it's a little hard to put yourself in the place we were in in 2013 because now there's Ro and Hims and like all these, there's GoodRx, there's like all these like quite successful pharmacy businesses. But at the time, like we were one of one, like people hadn't tried to build something similar back in 2013. And more tangibly, like we assumed that our best channel early on was going to be Facebook, right? Like that's where at the time everyone was acquiring customers. It was kind of arbitrage. in acquisition costs from 2013 to 2016 or something where you could pretty efficiently acquire customers on Facebook. And we kicked that off maybe 6 months post-launch and within like 2 weeks our account got shut off and they're like, pharmacies aren't allowed to advertise on Facebook. Like we don't allow pharmacies to advertise. And so we're sort of in this position where like we're supposed to be like a digital, like e-commerce business, but we can't use what at the time was by far like the most effective channel. And I'll save you all the nitty-gritty, but the punchline is 6 to 9 months later, they, we got them to change their policy. We got this accreditation that became kind of their rubber stamp on letting any pharmacy advertise. And we turned Facebook ads back on. Now this is like the fall of 2014, so maybe a year and a half post-launch, and it just started ripping. Like it worked exactly as we expected, but we had this kind of year-long like back and forth with Facebook. And we tried, you know, everything you can imagine. We tried, we tried podcasts, we tried radio, we tried direct mail. Like we tried all the channels that like were available to us, none of which were particularly effective. And turning Facebook on just worked like for, you know, the business was cranking the second we turned it on for the, you know, a number of years after that. So that was definitely the kind of most difficult moment where we thought it would work, but we had no evidence to suggest that it was actually going to work.

Yeah, definitely worth noting that point in time too, as you mentioned, and being, you know, being so early and kind of first in the space there. So, yeah, but could you maybe just briefly kind of, you know, walk us through how the business started to evolve as you scaled?

You know, one interesting— and I'm comparing this a lot in my head now with general medicine, but one interesting thing about PillPack was that we kind of built and launched this pretty idealistic consumer-facing product where it was super easy to sign up. You could track everything really easily. You had like a very simple dashboard to manage your prescriptions and do all the things. And that consumer front end Roughly for the totality of the business as an independent company didn't change. Like it was, we launched it, it was pretty stable. Like obviously like we kept sort of tweaking the visual design and some like minor things, but the front end was like pretty set it and forget it. And almost all of the energy went into backend tooling that enabled that thing to scale, right? So I kind of mentioned this at the beginning, but if you take 7 meds a month and 2 of them run out of refills and we don't get on top of that for you, you get 5 of them in the mail. It's a pretty crappy experience. And you can kind of go down the list of all of the logistics that we had to do to make the product work at scale. And all of our development energy went into building those systems and process and tooling in the backend, which was, was quite invisible to everyone but our team. And so it was really just like trying to build tools fast enough to keep up with the step functions and scale was really the thrust. And, you know, there's a lot of other evolutions of the customer acquisition strategy. We ultimately had a bunch of B2B partners that drove volume and that really evolved. But from a product and tech standpoint, almost all the energy went into kind of administrative and logistics-based tooling.

Yeah, it definitely sounds like it was a big lift as far as on the tech side. What were some of those challenges as, you know, like a, you know, I'd say a formidable kind of player in the space and operating in healthcare?

I mean, probably the biggest challenge we faced as we scaled the business beyond kind of the nuts and bolts of scaling was that all the incumbents hated us. Like if you really unpack the pharmacy business, there are these entities called PBMs, which, you know, I think for this audience, the right way to think about them is they're kind of old school demand aggregators in the same way marketplaces, but they're not consumer facing. And for a very long time, they had their own home delivery pharmacies that had captive audiences. They did not allow competitors in those networks. And so we were sort of fundamentally threatening that. And I think when we started the business, we had zero clue that that was going to be an issue. We were super naive. We just thought we built something great and we'd figure all this stuff out. And we did. So like it ultimately worked out, but not without quite a lot of both public fights and private behind-the-scenes negotiations with the, with the PBMs who effectively controlled access to these insurance networks. And so at one point we had the largest PBM, which is Express Scripts, who at the time was 40% of our revenue. gave us like a 2-week notice that we were out of their networks and all those customers were going to churn out like in 2 weeks. When you had, you know, at that time we probably had 400 employees that were reliant on that volume, kind of all the obvious sort of existential threats of a startup. And we made the difficult decision to run this very public PR campaign where we launched a website called fixpharmacy.com. We had 3,000 customer testimonials. We had all this like wonky stuff for regulators. And we ran this PR blitz where we had kind of 40 articles in one day. We had, you know, 10 congressmen call Express Scripts. We had half a dozen senators call. HHS called, CMS called. We kind of ran this full-court press and ultimately we won that, that battle. I think that was the point where we went from like this potential threat to like, okay, these guys are here to stay. We can keep screwing with them in like more nuanced ways. But like the idea that we're just going to like stomp on them and they're going to go away is probably off the table.

It's always interesting to hear some of the unique challenges that healthcare kind of marketplaces face as we're discussing in the community.

So, yep.

And then another topic that I wanted to discuss is the fundraising journey. And, you know, I believe you raised whatever, $100 million from some awesome investors like Accel and CRV before eventually being acquired by Amazon, which is incredible in itself, you know. So could you share what the fundraising journey was like and eventual acquisition by Amazon?

Yeah, I would say every other fundraise was like frothy and easy and the other ones were rough, were roughly the opposite of the arc we were on. We raised about $4 million kind of off the jump, like during that Techstars program and right after we did Demo day we raised effectively a $4 million seed round. And that round was pretty easy. Like we were like the hot company in the batch and, you know, if you're in one of these accelerators, like that's kind of the trick. And then we raised another $9 million from Accel. That was maybe 18 months after the $4 million. And that was the opposite. That was kind of this moment where Facebook wasn't turned on. We'd proven that people liked the product, like NPS was really high. We had maybe like, you know, 1,000 customers or something that were using it and it was working, but we had no evidence that like we could really acquire customers at the velocity that was necessary. And so that one was, was, would have been very sloggy. We got lucky in the sense that our seed investor left the fund he was at, went to Accel and just liked us and wanted to preempt. And so again, we got pretty lucky in that moment. It felt very unlucky when you have a partner leave a fund because that's typically not a good situation. But the fact that he was pretty jazzed up about it, he ended up just doing the round from the next fund. The round after that was only 9 months later. We literally got Facebook turned on, we turned the crank and it just was ripping. And we raised $50 million at like a 10x step-up in valuation in 9 months or something like that. One was, was really working. And then we raised another $50 million maybe 18 months after that, which was kind of passing the hat to everyone I could possibly find to keep capitalizing the business. So sort of this Yeah, this kind of ping pong between like really easy and like very competitive to like really sloggy or getting lucky with kind of the scenario with, with Accel.

Fundraising journeys are always fun to hear about, right? All across the board themselves. You know, you mentioned as far as, you know, partner kind of leaving, going to another fund, that seems to be somewhat more common in the past few months as well. So yeah, had lots of founders discussing that.

Definitely more common.

Yeah. Well, cool. So we could probably have, you know, keep going on and talking about Uphill Pack, but I want to make sure we could save some time to learn more about what you're building with General Medicine. You know, right before we do, I guess, kind of like looking back now, you know, what are some of your key learnings from PillPack that you've kind of taken with you to start General Medicine?

Yeah, I still have this maybe naive but very stronghold belief that if you build something great for consumers, everything else will roughly work out. And like the whole linchpin of this game is truly being able to build something that's great for consumers. And so I'd say we're, we're very much carrying that through in General Medicine. You know, I think if you really zoom out, when we started PillPack, like consumer was pretty hot, right? It was kind of the era of DTC. There was Warby and Harry's in a way, and like all these companies that that were like hot at the time. That's not the case now. But we still think building for consumers in healthcare is the biggest opportunity. And if you really focus on a differentiated product, that's the best way to win. So I think we're carrying— that's probably the biggest kind of headline thing we're carrying through to this business. I think the other thing more specific to healthcare is that there are definitely more efficient channels, right? Payers, pharma, health systems. But we've always felt like if you don't start with the consumer, you're going to build stuff for those entities and not build a differentiated consumer product. So those become like scaling channels later and they are more efficient, they scale faster. But to me, it's the wrong place to start. Like, I'd rather get volume flowing through that C2C and then ultimately kind of layer in B2B partners as scaling partners rather than the kind of initial go-to-market.

So, you know, so, you know, actually kind of transitioning into General Medicine, you know, so what led you to starting it? And could you share more about what you're building with it now?

I think if you zoom back to like 2018, when we sold PillPack to Amazon, the thing we were the most excited about was this kind of nugget of an idea that if you made healthcare shoppable like other categories, it would be a huge unlock both for the consumer and also starts to unwind a bunch of the inefficient supply chain that exists in healthcare. If we jump back to that kind of PBM comment, they're really like these grand demand aggregators that aren't focused on the consumer. So what does a modern demand aggregator look like that does focus on the consumer? And that's very much a digital marketplace, right? It's not a sort of complicated concept. And so when we sold the business to Amazon, we spent the first couple of years building out Amazon Pharmacy, which was the first truly shoppable pharmacy. So you can search and Browse the category like you shop for anything else on Amazon. You search for metoprolol or whatever medication you're on. You get a normal detail page. It looks like a retail detail page. You see kind of insurance pricing specific to you and cash price, kind of apples to apples in that experience. You add it to cart, you check out, like it just works like everything else works. And while we were at Amazon, we kept pulling on that thread beyond pharmacy, right? So we launched this clinic business, which is telemedicine kind of marketplace. And our, our sort of grand vision both then and now is that if you move all of this healthcare consumption to kind of a traditional retail-like experience, again, it's better for the consumer. It solves a bunch of downstream problems. And so we launched this business about 2 months ago, and it, over the fullness of time, will be mostly a 3P marketplace, but it will remain this kind of hybrid 1P, 3P that looks a lot like Amazon, but in healthcare. And we're super excited about both the breadth of what we can offer and the sort of scale of the opportunity.

Yeah, that's really cool to hear about. And you kind of mentioned as far as like the breadth of the offering, you know, as marketplaces, sometimes we're commonly discussing like, can they, you know, go vertical and kind of go focused, you know, versus kind of more horizontal in nature. So how are you kind of thinking about that?

I think something unique to healthcare is that, I don't know, 7 or 8 years ago when you had all these kind of vertical-focused players that were mostly like EDE players, but there was a variety of them. That was the pitch, right? We start with this narrow vertical and then we go super horizontal over time. My perspective is it doesn't really work in healthcare, and it's very different than retail in the sense that oftentimes you don't know what you need, right? So it's like if you want a book, you know, you would go to Amazon and get a book and like you transact and you're done. You know, if you go to your primary care provider, oftentimes you don't know what the next step is or whether you need to go see a neurologist or some other specialist. You need to get imaging. Like it's a much more amorphous. sort of starting point and shopping experience. And so, you know, our position, and we certainly could be wrong, but our position right now is that you have to have a broad-based sort of horizontal offering in healthcare for it to really start to work. The downside of that is like it's really hard to figure out exactly what is the right kind of wedge. So I think about these things more as like wedges, but like needs to be horizontal versus starting in vertical and then kind of chipping away at other verticals in healthcare because we've seen that playbook and it hasn't really played out in the way that people would have thought.

Definitely. It's a great point and kind of nuance applied to healthcare. So, and it's, of course, an exciting time right now with the AI and lots of kind of new opportunities and possibilities. So how are you potentially leveraging AI with General Medicine?

I'll give 2 examples. I think out of the gate, we're using AI a ton, but mostly in an invisible way to the consumer. So probably the most tangible version is we're the first healthcare marketplace that gives you personalized insurance prices for any service with any provider. So we can tell you exactly what your out-of-pocket is going to be. whether you're getting an X-ray at your local hospital or need to go see a neurologist at the local health system, we can price that for you and you can book the appointment with us. And that would've been impossible pre-AI because under the hood, the way we're doing that is we can automatically pull down your insurance information and we've got clever ways to do that. We then go pull down an 80 to 120 page coverage of benefits summary, which is just a long PDF. And then we're using LLMs to scrub that PDF and turn that into like a useful table that we can then go price things with. It's a perfect, kind of AI LLM problem because previously, like, you would've had to have humans in an operational call center try to like parse this very long kind of amorphous document. And, and to the consumer though, they, they just get like an insurance price, right? So like, wouldn't have been possible without AI, but like, it's not like it feels like an AI feature to the consumer. I think on the opposite end of the spectrum, and I think this is probably applicable to a bunch of the founders here, you know, ChatGPT and similar UI are a huge unlock to service-based marketplaces. Because one of the challenges with service-based marketplaces historically is like, you don't really know what you need. And ChatGPT is really good at taking, like in healthcare, taking like the set of symptoms and narrowing that down to a likely next step. Right. And so for us, like layering in these SKUs downstream of that interaction. So maybe you show up and you say like, I've got this shoulder pain. When I raise my arm, it hurts. I don't really know what's going on. Historically, like you would need to talk to a doctor. to figure out what's next. With ChatGPT, they're going to take you all the way through to, it's like probably this thing. To validate that, you need this X-ray. And that in healthcare can now look like a Shopify SKU getting dropped into the ChatGPT and transacted with. And I think that's probably applicable to like home services and like all these other places that have historically been tricky to, to build marketplaces. I think ChatGPT and those, that sort of, that sort of interaction for consumers is a huge unlock for things that previously were difficult to like shop.

No, definitely. Those are 2 great examples. And yeah, as you mentioned, it kind of has made a big impact on, you know, how marketplaces now can kind of add value. So, and I can imagine, are there some kind of like partnerships that you also kind of have in place to kind of access some of this data?

Yeah, I mean, I would say that on the data side, it's less partnership-oriented. It's more being really good at knowing which vendors can sort of provide this data. And a lot of that's just overhang from the work I pulled back in Amazon where we know we can pull in All your insurance information, we can pull in your entire medical record, like all of that with no kind of data entry from the consumer. But that's really just being smart about vendors. I think on the partnership side, we're looking at a bunch of different opportunities, but in the fullness of time, that'll be a combination of traditional retailers that want to offer similar selection, potentially, you know, these GPT-like interfaces where we could drop in like a Shopify, payers, health systems, like the whole gamut of potential partnerships. And it's really like the tricky bit is like how do you prioritize and stage this stuff? Not whether there's like a good opportunity there.

Yeah, definitely. Sequencing is always, always a challenge, right? Yep. Briefly, before we get to the group Q&A, is, you know, in healthcare, a lot of times like aligning incentives, you know, with the kind of customers and then, you know, who's actually the kind of paying customers. So do you have any tips maybe for founders on this, on how to do so?

I think it's an interesting comparison between PillPack and general medicine where PillPack, you know, we were pretty misaligned with the incumbent, right? So either you're a retail pharmacy and we're stealing share Or you're a PBM and we're sort of encroaching on your captive mail order business. And so it was pretty zero sum and incentives were pretty misaligned, I'd say, with the existing ecosystem. General medicine's very different, right? So we are not zero sum. Like most of our volume is going to be going to incumbent third-party providers and players. And so assuming you can figure out the right flows and prioritizing the volumes that they want, like we're pretty aligned with offline providers and even kind of traditional telemedicine providers. You then look at payers, which are the other kind of major participant in the ecosystem. Ultimately, what a commercial payer wants is you to get to the lowest cost, kind of most effective therapy as fast as possible, right? Like, so if you take this X-ray example, like I had to go get an X-ray for my shoulder, getting it at the Park City Imaging Center was like $600 out of pocket. Getting it in Salt Lake was $200, and that aligned with the reimbursement rates for the, for the insurer. They are actually quite aligned with what we're trying to do at General Medicine because by pulling price forward, we help consumers make better decisions and go to the lowest cost provider. So I think comparatively, General Medicine is a much better alignment of all the different kind of participants in the ecosystem. And I think that just comes with having lived the other experience, which we made it through, but it was tricky and it was existential. And this is tricky for different reasons, but we have definitely it's definitely different from an incentive alignment standpoint.

Yeah. And then kind of back off the top of that is monetization alignment as well, right?

So yeah, totally.

Cool. So we're going to get into the group Q&A, but, you know, right before we do, do you have a few tips maybe for early-stage marketplace founders that you could share with us?

Yeah, I mean, it's— I'm guessing it's super dependent on exactly what folks are working on. I do think this kind of like very pithy comment is still quite important, which is like it all comes down to like how good is the product experience for the consumer and like how How much have you wrestled with like the true nuance of your customer journey and how you compare it to other options for consumers? And like, is it something that's truly, you know, novel and really seamless for the consumer? I think if you can nail that, like there's a lot of tactics that are still tricky, right? Customer acquisition and finding your wedge and all that stuff. But like none of that matters if you haven't built something great for, for the consumer.

That's going to be a helpful tip. So we're going to get into that group Q&A. I'm actually going to ask a question for a founder that they added to the Google Doc in advance, and then we'll get to some others. And this is actually from Al, and He was asking, at what point did you feel like you had a product market fit for PillPack? And what were the signs that showed this to you for the business?

This is a funny one with PillPack because it was so derivative of a product that my dad was doing previously. And so if you really kind of unpack that story, we had like a couple hundred consumers that really wanted PillPack, like wanted what he was offering, but didn't live at a nursing home or assisted living facility. And so we were like sort of jankily putting that together. Like we were jamming it in a Ziploc bag and dropping it in like a USPS flat rate box. And like you had this pull from the market that really insinuated that people wanted what we were building. And then, you know, I think more quantitatively, like we maniacally tracked NPS and retention and all the kind of downstream stuff. And from the jump, like our NPS kicked around 75, 85, like somewhere in there for the totality of the business. And we felt like we had product market fit off the jump because of the kind of precursor to PillPack. The first year when we had like 1,000 customers, they loved the product. So we were feeling pretty good about it. Once we got Facebook turned on and people were signing up at a higher rate and stayed on the product, we were really confident. So compared to general medicine, which is definitely less obvious on day one because you didn't have the precursor like that, we're really wrestling with MPS and where the most efficient wedge is, like all the stuff we've talked about. And that'll be, I think, a more traditional journey. But PillPack, was wacky in the sense that I don't think we ever felt like we were wandering in the wilderness, like trying to find product-market fit.

And I'll go ahead and ask another question around, you know, competition and kind of the competitive nature of not only marketplaces, but, you know, some in healthcare as well. And so, you know, how do you think about competition and what also seems like lots of moats in the healthcare space with larger organizations?

Yeah, we didn't think about it a ton. I think we always felt like the existing experience was pretty bad for consumers and had mostly been designed around other constituents. Right. It was like almost all about what the payer wants and what the provider wants. And very few folks have thought about what the consumer wants in healthcare. That's quite unique, obviously, to healthcare. I think as, you know, some of these super frothy, like, AI categories where there's 20 people trying to do the same thing, it's probably pretty pertinent to think a lot about competition. But, you know, I think for us, and I think in a, probably for a lot of the marketplace founders, it's mostly a distraction, like heads down, are we building something great? Kind of ignore the competition has always served me well.

No, I love it. That's, that's what we need to hear. So, and then, and then another kind of fun question that was added was, you know, what are some of the biggest mistakes that you see with founders building in healthcare today?

I think this vertical one tends to be a mistake. I don't, I think consumers don't think about healthcare in like a condition way, a condition-specific way. This is probably more controversial, but I think like not Taking insurance and only doing cash is a crazy thing to do. Like if you ignore healthcare and imagine that like every consumer signing up for your thing has a coupon in their pocket that will make your service like free half the time and you don't do the work to make that possible, that just feels like a really shortsighted, like crazy thing to me. But people take the opposite side of that and that's fine. We're long insurance integrated with cash, not just cash.

That's great. And then another question is around team building and hiring. So of course, I mentioned, you know, your awesome relationship, you know, with your co-founder, Pass, and being key. But do you have any tips maybe for founders as far as when it comes to team building and hiring?

Yeah, I mean, I think the only way to be really good at building a great team is to be good at firing people that weren't great, to be honest, as like a— especially as like a first-time founder. Like, if you look at the statistics, like, great hiring managers are right like 65% of the time and terrible ones are right like 45% of the time. So it was like, you have to be really good at understanding if someone's good after the fact and moving on quickly. You just don't have a ton of time in these games. I think as you get a little bit older and you have a little more experience, hopefully your hit rate trends closer to the 65 than the 45. But for me, it was like being really good at sort of acknowledging I made a mistake and moving on. You know, I think I went through marketing leadership like 3 times over in 3 years just trying to figure it out. And I think that's like totally okay. I think it's much better than— kind of sticking with something that's not working. So I think that's like, and it's like really hard. I think like for me, firing people was the most painful part of being a founder. Like it's the only time that I like really like lost sleep and was super stressed out. But every time you do it, you realize you made the right decision. You probably should have done it faster. And I kind of think that's the only way to build a super high caliber team as a first-time founder.

I can definitely relate with that myself as a founder.

Yeah, it sucks. It's like the worst, but you have to do it.

Yeah.

Yeah.

And another kind of related topic to this that we're typically discussing in the community is, you know, with when building a market, say for instance, like kind of healthcare, you know, hiring for, you know, as far as specific experience, maybe like in the market when like, you know, relationships kind of might be key and/or, you know, the business model in itself as well.

The only place where I think the value of expertise and experience and relationships outstrips the downsides of it are like business development, sales, roles. I think we have a strong bias to like great generalists that especially in like product and tech, like we're not looking for people that have built stuff in healthcare per se previously. We're looking for great technical talent that sort of backfires in places where like relationships matter, which is mostly BD and sales. And then I think like more generally, you know, like at Copac, like at one point, like it was very clear the operations were imploding, like financial, like the financial org was imploding. Like when that happens, like you need an adult. Like we hired like a really killer COO, CFO that had been there, done that, like very accomplished. And at that point, like we really needed that expertise. That one I think is really tricky to get right. Like you either hire too early and then they want all this process that's like slows everything down and like it's just unproductive, or you do what we did and you hire that person when it's so far gone that it was like not a question whether you needed them. I think getting that right is really important, but really, really hard. And I'd bias towards doing it too late rather than too early, because again, you preemptively hire those adults in the room. It can kind of really backfire if you don't truly need them yet, if that makes sense.

And I was going to ask at what point in time that was, but it sounds like when you really knew that you needed them, right?

Yeah. The funniest was we had this board meeting where we'd hired her, but she hadn't yet started. And you could tell all the board members wanted to just They were ready to like lose it, but if they lost it, she would've not like actually joined the company. So it was just like a really funny dynamic. We're in the boardroom like, I know you all want to say it, but you can't say it because otherwise you like burn the bridge with the person that could fix it. So here we are.

That's a funny, funny story. So yeah, and I guess, you know, related to, you know, and jumping into kind of what you're building with General Medicine now, you know, what is the team like today?

Team today is about 30 folks, very kind of product and engineering heavy because it's super early. You know, we're just building out the kind of marketing and BD functions, very kind of nascent. So a lot of product and eng. There's obviously clinical, like we have our own medical group and our own docs and like some operations. But I'd say the thrust of the current team is very like product tech because it's just launched 2 months ago.

Yeah, definitely early. So we'll be following along and we'll have to make sure that we can share some roles in the community. Yeah, I guess lots of people coming your way.

So yeah, if there's any great marketers listening, we are looking for great marketers, so feel free to reach out.

So if you had any questions, yeah, feel free to jump on here.

Yeah. So first off, love the conversation, especially around fundraising. I think that's a funny story and the focus on just building something that has a huge impact first. So love that. From what I've seen, like marketplaces have undergone a few iterations. Right on the back of technical innovation. So internet equal to Amazon, eBay, smartphones, Uber, DoorDash, all of those. One, do you think that AI's going to have a similar impact on the way that marketplaces operate? And if you do believe that, how do you think that's going to materialize?

Yeah, I think there's 2 pretty obvious vectors. One is the services dynamic I mentioned where I think things that were previously tough to turn into marketplaces that didn't have a clear product catalog with like clear pricing, I think it's a big unlock there. I think consumers can get much farther down the buying process without involving a human than they could before. So I think it is a huge unlock to do things that previously have been resistant to become kind of digital marketplaces. And I think services are the biggest bucket of stuff that are going to be impacted. And then the second, which also probably quite obvious, is that I think these GPT-like interfaces are going to be the main driver of incremental demand for marketplaces. Like, I think you're going to see more and more of the shopping kind of decision-making process happen offsite, happen inside of ChatGPT and elsewhere. And the more you can either design your experience to be consumed by GPT or integrate in the ways that like Shopify already has, where like it drops a SKU straight into the chat and you can transact right there. I think that's going to be critical. And the folks that are like the fastest off the jump on those consumer flows are going to be a lot like the companies that understood like Facebook arbitrage in 2015, right? It's just like the new, the sort of new demand arbitrage. And at some point it'll normalize to where they're capturing proportional economics, and that's going to happen obviously. But there's always this kind of like couple-year window where if you're really good at it, it'll be like very beneficial to kind of new entrants, if that makes sense.

We'll get in another question here that, that Danny added to the doc. And it was around, you know, how's your founder experience shaped how you evaluate founders and, you know, startups specifically building in healthcare when you're investing at Matrix today?

I mean, I think in early-stage investing, it's like the founder is really the only thing that matters. And every time I kind of wind myself up about TAM or like exact business model, or it's just like a useless use of energy, like it always comes back to like, how good is the founding team? And like, do I believe they can figure this out? And, you know, I think if you look at my background, like, I'm not like a traditional founder. I went to like a very mediocre school. I trained as a pharmacist. I didn't go to Stanford. I'm not like technical. And so I like have a bias to people that kind of, that have a more non-traditional background personally. And then, you know, if you, if you take us as a founding team with me and Elliot, like, that to me is a pretty ideal kind of pattern matching exercise because like if you have a great, if you're someone that's great at selling and like has really good product instincts and someone that's great at building, like that is like the quintessential founding team. And you don't always get it. People have been very successful just being technical and there's been just non-technical founders. But if I can find that kind of symbiotic relationship, that's sort of the best for me. And like healthcare, I think some amount of understanding of the actual like realities of healthcare is relevant. I think in some other categories, like coming in super hot with like no relevant expertise is like, is probably better in healthcare and probably fintech and some of these other very kind of technical specific categories. Like having some like domain expertise is pretty useful in healthcare for sure.

That's going to be really helpful here for, for founders. So, you know, really appreciate, you know, taking time to join us here today as a, as a special guest. This is definitely a real treat and to have you join us here today and, you know, sharing all of your experience and, and learnings with us. You know, it's really cool to hear about that general medicine as well. And I actually had one last question for you before we wrap things up. And this is my usual kind of parting question, which is that if you could go right back to before you started PillPack, what would you tell yourself about the journey?

I probably would tell myself to like really like appreciate the journey, like really enjoy the experience. Like I think in the moment it's like so existential. And I think for me it was like, I want to make sure I can get this thing to a good outcome for me and my team and my investors. And like, I was so wound up about That. And in hindsight, like, those are like the most, the most fond memories I have was like those really existential, really hard moments. And I think part of going back and doing it again in general medicine is being able to like, it's less about like, we want to sell this thing for $1 billion again or $10 billion or whatever. It's like, I want to do that again. Like, I want to be in the shit. Like, I want to have like, be really be in it and be able to sort of really enjoy it for what it is. I kind of, I kind of think about it like the way a grandparent views grandkids versus like you're a parent and you have a 3-year-old and it's like, it's a lot, you know, and try and sort of like, if you've already done it and you've been through it, like you realize how special it is. And so I think if I could go back and tell myself something, it'd be just to like really enjoy the, enjoy the ride because it's pretty special.

That's gonna be great parting advice for us all here. So, and definitely relatable. So we're all in the, in the thick of it right now and everything seems like a fire, right?

Yep. Yeah, but it's fun. You'll look back on it fondly for sure.

Awesome. Well, that's encouraging as well because it's something to look forward to. So, you know, we once again, I really appreciate, you know, taking the time to join us here today. And last but not least, time for a plug. Where can we follow along?

I'm on Twitter @TJParker. I'm on Instagram, TJ Parker as well. So, and I guess I'm on LinkedIn. I don't even know what my handle would be on LinkedIn, but I'm on the internet. Pretty easy to find.

Awesome. We'll include links and I'm sure you'll have lots of founders coming your way and also include links to general medicine as well so we can send some awesome talent your way. So Once again, really appreciate it, and thanks everyone for joining and the great questions.