What's Next for Education Startups

Everyone expects schools at all levels -- from pre-school to post-graduate universities -- to change fundamentally as software turbocharges both students and teachers, enables new business models, and brings scale to an industry that often prides itself on its cottage-industry status (think about the debate over classroom sizes). So in this a16z hallway conversation, consumer team partners D'Arcy Coolican and Li Jin chat with deal and research team operating partner Frank Chen about the future of education. How long will it take for a traditional 4-year undergraduate program to not be the main path for students? What should we make of Income Share Agreements (ISAs), where the school owns some of the risk if their curriculum and instructors don't generate employable graduates? Given how hard it is to find a reputable, caring pre-school, what can we do to open up more qualified pre-schools? Beyond STEM courses, what other areas will are well suited for disruptive education? *** The views expressed here are those of the individual AH Capital Management, L. L. C. (“a16z”) personnel quoted and are not the views of a16z or its affiliates. Certain information contained in here has been obtained from third-party sources, including from portfolio companies of funds managed by a16z. While taken from sources believed to be reliable, a16z has not independently verified such information and makes no representations about the enduring accuracy of the information or its appropriateness for a given situation. This content is provided for informational purposes only, and should not be relied upon as legal, business, investment, or tax advice. You should consult your own advisers as to those matters. References to any securities or digital assets are for illustrative purposes only, and do not constitute an investment recommendation or offer to provide investment advisory services. Furthermore, this content is not directed at nor intended for use by any investors or prospective investors, and may not under any circumstances be relied upon when making a decision to invest in any fund managed by a16z. (An offering to invest in an a16z fund will be made only by the private placement memorandum, subscription agreement, and other relevant documentation of any such fund and should be read in their entirety.) Any investments or portfolio companies mentioned, referred to, or described are not representative of all investments in vehicles managed by a16z, and there can be no assurance that the investments will be profitable or that other investments made in the future will have similar characteristics or results. A list of investments made by funds managed by Andreessen Horowitz (excluding investments and certain publicly traded cryptocurrencies/ digital assets for which the issuer has not provided permission for a16z to disclose publicly) is available at https://a16z.com/investments/. Charts and graphs provided within are for informational purposes solely and should not be relied upon when making any investment decision. Past performance is not indicative of future results. The content speaks only as of the date indicated. Any projections, estimates, forecasts, targets, prospects, and/or opinions expressed in these materials are subject to change without notice and may differ or be contrary to opinions expressed by others. Please see https://a16z.com/disclosures for additional important information.

Speaker names marked ? are inferred from speaking time and the show’s known hosts — they are not voice-verified.

• 0:00Speaker 00

  The content here is for informational purposes only, should not be taken as legal, business, tax, or investment advice, or be used to evaluate any investment or security, and is not directed at any investors or potential investors in any A16Z fund. For more details, please see A16Z.com/disclosures.

• 0:19Sonal?

  Hi, everyone. Welcome to the A16Z Podcast. Today's episode was originally recorded as a video, which you can find in our YouTube channel at youtube.com/A16Zvideos, of a hallway cell conversation between Deal and Research Team Operating Partner Frank Chen and Lee Jin, who covers our consumer vertical, and co-wrote with General Partner Andrew Chen a post on the topic of what's next for marketplace startups. So in this video episode, Frank and Lee discuss the past and future of marketplace startups, and why aren't there 100 services marketplaces yet?

• 0:53Frank Chen?

  Well, hello YouTube. My name is Frank Chen, and we're going to have a hallway conversation today. Well, what's a hallway conversation? It's exactly like it sounds. We have tons of conversations here in the office with each other as we're thinking about what kinds of companies should we invest in? And what we're going to do is take you inside the sausage factory so that you can hear a conversation exactly like that. So today, I'm here with Lee Jin, who works on our consumer investing team. Welcome, Lee.

• 1:22Li Jin?

  Hello, everyone.

• 1:23Frank Chen?

  And today, we're going to talk about marketplaces. So, everybody knows eBay. eBay is an awesome place to find anything because it connects buyers and sellers. And ever since eBay, and how it's been 30 years? Something like that. eBay was definitely an Internet 1.0 company. We've been waiting for 100 at-scale marketplaces. There's so many awesome benefits you get when you bring people, buyers and sellers, together. But there haven't been. There haven't been 100 at-scale marketplaces, and Lee has been wondering why. And so, maybe, Lee, let's start talking about why aren't there 100 at-scale eBay-sized marketplaces for buying services?

• 2:06Li Jin?

  Yeah. Well, first, I want to back that up with one statistic. So I was doing research about services and how important they are to the economy. So services represents 2/3 of private sector spend by consumers in the economy. But when I looked up the statistics online, I think the Bureau of Labor Statistics has a statistic that only 7% of the services that we purchase are digital, meaning that they have been mediated by the Internet. So despite the huge amount of spend that we have on services, very few of them are actually occurring online.

• 2:46Li Jin?

  And I think there's four reasons why that is the case. I think services are obviously really complex. When we go to a service provider and we're looking for a haircut, or a babysitter, or childcare, or a hotel room, there's so many different attributes that we're looking for. There's so much complexity in our value function of things that we value. For instance, for a babysitter, are they good with children? Do they speak other languages? Do they know CPR? How kind are they? Services are super heterogeneous and difficult to standardize and to skew the way that goods are.

• 3:28Li Jin?

  And that complexity makes it really difficult to capture the relevant information in a marketplace.

• 3:34Frank Chen?

  Hard to skewify.

• 3:35Li Jin?

  Exactly.

• 3:36Frank Chen?

  So many different attributes.

• 3:37Li Jin?

  Yeah. I think another reason why services have sort of been underpenetrated is that there's no clear metric of success or what quality looks like. So what a good service experience has been, for me, might not be viewed the same way by another consumer. So what one consumer thinks of as a five-star experience could be three stars for someone else, going back to my point about the very complex value functions that consumers have.

• 4:08Frank Chen?

  So sort of beauty is in the eye of the beholder.

• 4:12Li Jin?

  Exactly. Yeah, we think another reason why services have been slow to come online is because a lot of services are performed by small business owners and individual owner-operators. And these businesses typically don't have the tools or the resources to really invest in going online, setting up a webpage, making sure that their availability is somehow listed online and that they're bookable. So a lot of it still remains offline for that reason.

• 4:44Frank Chen?

  I think of my favorite neighborhood barber where I get my haircut, Belmont Hair Studio. It's this delightful family, they're immigrant family, and they love cutting hair and they love talking to people. But their sign of the prices and services available, they haven't touched that for three decades.

• 5:01Li Jin?

  Yeah, and often when you call them, it's actually, it's been updated and the prices have gone up versus what's online because that webpage is probably 20 years old.

• 5:09Frank Chen?

  Yeah, it's the absolute last thing they think about.

• 5:11Li Jin?

  Yeah, exactly. And then I think another key aspect that distinguishes goods from services is that goods, the stages of the value chain from manufacturing to production to shipping it to a warehouse to consumption and purchase by a consumer, all of those stages in the value chain can sort of be split apart and broken up and serve and each tackled by a different company in a different marketplace. So for instance, Amazon is responsible for keeping things in stock and making sure that your orders get fulfilled. Services are a

• 5:51Li Jin?

  bit more complex because they involve this very synchronous creation and receiving a value by the consumer. Oftentimes, services happen at the same place, at the same location in an in-person meeting between the service provider and the consumer. So there's difficulty in extracting pieces of the value chain that can actually be brought online. Like if you're going to go in for a haircut and meet with your barber and actually set up a time to do that, then it's probably likely that it might be easier today to just schedule that in person or to do the payment in person while you're getting your haircut,

• 6:36Li Jin?

  rather than breaking up the stages of the value chain into a marketplace.

• 6:39Frank Chen?

  So all of those are reasons that we don't have 100 at-scale marketplaces, right? It's hard. There's lots of complexity in skewifying, right? Products and attributes. There's lots of complexities in delivering. But a couple of companies have managed to emerge, and I think what you did in the blog post was you started looking at the history of the marketplaces that have been successful. So when you dug into the history, what did you find?

• 7:04Li Jin?

  Yeah, so we really zoomed out and thought about the last, say, 30 years of marketplaces and what's been built so far. And when we took that survey of the history of service marketplaces, what we found was that there's pretty clearly been various eras of marketplaces. Specifically, there's been four major paradigms or models for marketplaces that have been trying to tackle services. So in the beginning, in the early days of the internet, in the 90s, the first attempt at bringing services online were what we called the listings era.

• 7:46Li Jin?

  Basically, webpages that were digitized versions of the Yellow Pages. So in the Yellow Pages, you just had, you know, an entire book full of service providers that you could go down and actually call. Marketplaces from this era were pretty similar to that. So examples of this would be Craigslist or Yelp, which basically gave consumers a huge list of all of the available service providers nearby. But it really put the onus on the consumer to have to do the legwork in clicking on all those links, finding their contact information, reaching out, coordinating with the provider,

• 8:31Li Jin?

  setting up a time to meet offline, and actually transacting offline.

• 8:34Frank Chen?

  Yep. And so there was definitely a set of people who loved browsing through Yellow Pages ads, right, the ones at the back, or they loved looking through classified ads in newspapers and sort of highlighting them and clipping them out. But like, for most of the people, that's a miserable experience. Like, that's the last thing that you want, to find a great hairdresser or to find a great daycare. You don't want to go through lists and lists and lists. And you kind of are suspicious of the big ads because you're like, oh, maybe like, you know, they're spending all their money on advertising and not on actual childcare, right? And so for most, that was a pretty bad experience. Yeah, so browsing listings is a pretty miserable experience.

• 9:08Li Jin?

  Yeah. And I think listings, it's not only tedious and a lot of work for consumers, there were also problems with trust. So for services that required some component of trust, where you were potentially letting someone into your home or letting someone interact with your child, it was really difficult to go onto these listings platforms and have high confidence that the service provider that you were speaking to would actually be someone that you could trust.

• 9:37Frank Chen?

  Yeah. One of the fun facts from this era is if you could find Yellow Pages, go look, you might have to go to a museum to try to find one of these things. But if you look at locksmiths, you will see a lot of locksmiths named A and AAA locksmiths. And the reason for that is they would actually show up first in the listings, right? And so if there was something like that you could do to gain traffic to your business, it's going to create trust problems, right? Which is what makes AAA locksmith any better than ZZZ locksmith. But for some reason, we do have a lot of AAA locksmiths and very few ZZZ. So there was definitely some gaming going on.

• 10:15Li Jin?

  I think the modern day counterpart to the AAA locksmith example is

• 10:21Li Jin?

  Providers who are just listing their service over and over again on Craigslist.

• 10:25Frank Chen?

  Oh, in Craigslist, yeah.

• 10:26Li Jin?

  And you'll see those duplicate listings if you go back a few pages.

• 10:29Frank Chen?

  Yeah.

• 10:30Li Jin?

  Because it's chronologically ordered.

• 10:32Frank Chen?

  Yeah.

• 10:32Li Jin?

  And so the people who rise to the top are the ones that just post a ton of times.

• 10:36Frank Chen?

  Yeah. And they'll post small variants, right? I'm a home theater installer, San Jose. Home theater installer, Cupertino, right?

• 10:43Li Jin?

  Yeah, exactly.

• 10:43Frank Chen?

  So, yeah. You're like, I know that's you. Same you. So that was the listings era. So pretty miserable from an end user experience point of view, right? Like, I'm browsing, I don't know who's there, I don't have trust. Like, what happened next? Like, how did we get past the... Well, Craigslist still exists, but it's not the marketplace for everything.

• 11:03Li Jin?

  Yeah. So what we started seeing next is companies started addressing those issues of trust, of difficulty of finding providers, and tediousness of actually reaching out to these folks by picking off various pieces of Craigslist and creating what we call unbundled Craigslist marketplaces. And so there's that famous graphic that you can probably see online, which is like the Craigslist homepage. And basically, every single category now has its own specialized marketplace that tackles just that particular category. So, for instance, home services, which is still its own Craigslist section.

• 11:47Li Jin?

  Now we have Angie's List, and Thumbtack, and TaskRabbit that are just doing home services. In the childcare section, we have things like Care.com. And there's a bunch of other examples. So basically, the next generation of marketplaces unbundled Craigslist and created verticalized marketplaces around a specific service category so that they could offer richer features, better filtering technology, better ways for users to be able to search and uncover the right provider for them.

• 12:21Frank Chen?

  So by specializing, you could build features tailored to the service providers that you were trying to connect with consumers. So what's a good example of that?

• 12:31Li Jin?

  Yeah, so an example that I like is on Angie's List. Angie's List actually has like a certification process that contractors and home service providers can actually go through so that they get a little badge on their listing, which is like Angie's List Certified.

• 12:51Frank Chen?

  Recommended.

• 12:52Li Jin?

  Exactly, recommended. And a lot of these verticalized marketplaces have some sort of similar credentialing or verification or recommended process because what they're trying to do is they're trying to help consumers figure out what's good quality versus what's poor quality. Who are the service providers that have a lot of good customer satisfaction versus the ones that are more average? And so this is trying to tackle that problem of search and discovery.

• 13:30Frank Chen?

  Yeah, so the badge of awesomeness helps service providers stand out from each other. I remember this era well because I was actually working at a consumer company called Respond.com, which was a participant in this. We were trying to be a marketplace that would let you find local service providers. And the way it worked is you'd describe what you want in your own words and you'd fill out a form. And then we would take that and send it to a list of local service providers and they would bid on your business. So kind of a reverse auction. You would say, "I will change that lock for $50." Or you'd get another one saying, "No, I'll change the lock for $75," or something like that. And one of the things we ended up having to do is that for every service provider, they wanted to know different things. The locksmith wants to know, "Do you need me now?

• 14:15Frank Chen?

  Do you need to change locks? What do you need me to do?" Whereas the window covering person needed to know, "How many windows? And do you want blinds or do you want curtains? And do you want short or fuzzy?"

• 14:26Li Jin?

  Right.

• 14:27Frank Chen?

  And so we ended up having to build these specialized forms for every service provider so that when they got the request, they could respond to it intelligently. So the specialization helped sort of Streamline the communication Yeah, exactly. Between somebody who wants something And then somebody who was providing something.

• 14:44Li Jin?

  Right.

• 14:45Frank Chen?

  Got it. So that was sort of era two, which is we moved out of straight listings and we moved to an era where there were specialized feature sets for the marketplace. And then what happened? What was next?

• 14:57Li Jin?

  Yeah, so then, starting in the late 2000s, early half of the 2010s, we saw the rise of a ton of on-demand marketplaces, or what we call in the blog post Uber-for-X marketplaces. So these, I think, the why now for those, that generation of marketplaces was really the advent of mobile and the rapid adoption of smartphones by consumers and service providers. All of a sudden, people could know where various providers were located relative to them. So you had location that you could do matching on.

• 15:40Li Jin?

  People could respond to requests immediately, rather than waiting until they got home to their computer and could check their email. And so this was really conducive to an on-demand model where you could receive a service very quickly, almost immediately in real time, without having to wait and to have a back-and-forth messaging process over potentially multiple days. And so the on-demand model of marketplaces was really conducive to simple services that were very atomic and simple transactions that could, you could do matching for in a very quick algorithmic way

• 16:20Li Jin?

  without, you know, matching on various different complex dimensions.

• 16:26Frank Chen?

  You didn't have to ask 27 questions. It's kind of push-button and something useful happens, right? Car arrives, a, somebody who will pick up your stuff at Costco arrives, or whatever.

• 16:37Li Jin?

  Exactly. And I think another thing that we saw happen with this generation of marketplaces is that because the matching happened in an algorithmic centralized way, what happened was that the supply side sort of became more obscured and sort of abstracted away for consumers. So as a consumer, you would say something like, "My Uber driver picked me up," or "I'm getting food delivered through DoorDash." You wouldn't think of the counterparty of the transaction as being that actual end provider, the way that you did in previous generations of marketplaces.

• 17:16Frank Chen?

  Yeah, you didn't know their name, you just wanted to know that like, I don't know, they had four stars or more or something like that, right?

• 17:21Li Jin?

  Exactly.

• 17:22Frank Chen?

  And that was good enough.

• 17:23Li Jin?

  Yeah.

• 17:24Frank Chen?

  Because otherwise they were fungible, right?

• 17:25Li Jin?

  Right.

• 17:26Frank Chen?

  They were sort of replaceable.

• 17:27Li Jin?

  Yeah, and that's because these on-demand marketplaces really nailed in on a singular use case and took hold of that entire end-to-end transaction from discovery to matching to payment and actually a transaction. And so because that entire service delivery process was captured by those marketplaces, people didn't really think of, the end provider wasn't top of mind for the consumer.

• 17:54Frank Chen?

  Mm-hmm. They were sort of more or less replaceable.

• 17:56Li Jin?

  Exactly.

• 17:57Frank Chen?

  So long as you're reputable and strong enough to carry my boxes, I don't really care who shows up to carry my boxes.

• 18:03Li Jin?

  Yeah, and we also saw, I think, a large reckoning with this generation of marketplaces as well, where a lot of the startups that were trying to become Uber for X didn't ultimately end up lasting and surviving as standalone companies. And we think that that's because the model got applied probably a bit too liberally to service categories that weren't as conducive to the on-demand model. So there's a lot of services out there that actually don't need to be delivered immediately that people are willing to wait for. There's other service categories that are more complex, where consumers actually want

• 18:44Li Jin?

  to know who they're interacting with at the end of the day, rather than just the platform itself.

• 18:50Frank Chen?

  Mm-hmm. And that gave way to the next era of marketplaces.

• 18:54Li Jin?

  Exactly.

• 18:55Frank Chen?

  So where the person actually mattered, they weren't fungible. And so what do you guys call this era?

• 19:00Li Jin?

  Yeah, so the most recent era of marketplaces is something we call the managed marketplace era. So managed marketplaces went a step beyond all of the previous eras of service marketplaces in really providing a ton of value add to the consumer as well as to the service provider in tackling even more complex services and doing a heavier operational lift.

• 19:27Li Jin?

  And so these managed marketplaces take on some additional value add or functionality that consumers and suppliers really value that they previously had probably been doing themselves. So examples of this would be Honor, which is one of our portfolio companies that provides elder care. Honor actually interviews vets, recruits caretakers who they employ as W2 employees so that they meet some sort of minimum standard bar and they can be trained and they can provide really high-quality service. And previously, in the Care.com era of care marketplaces,

• 20:12Li Jin?

  the end user, the end consumer would have had to reach out to a bunch of different professionals, interview all of them and do all of this work. Now, a platform like Honor is taking on that work and that operational heavier lift on their behalf.

• 20:27Frank Chen?

  Yeah, Seth and crew are doing such a good job. And he was explaining to me the last time I was chatting with him about how complicated it is to find you the perfect person to do the perfect task and interact with just the right person. There's so many variables, like what language do they speak? And can they lift heavy things? And are they punctual or not? And all of these things are very complicated and lead you to either have an okay or a great experience.

• 20:52Li Jin?

  Right.

• 20:53Frank Chen?

  And so they actually have to use machine learning algorithms to match service providers to maximize the chances that you'll actually have a great experience because it is complicated.

• 21:02Li Jin?

  Yeah, exactly. I think of the managed marketplace era and this managed model as being a really good model for services that are complex, that require higher trust, that maybe have higher stakes involved or potentially involving higher-value transactions. So for instance, Open Door is another example of a managed marketplace for buying and selling real estate. This is another one of our recent investments. But rather than following the existing model of real estate in which most real estate agents actually are independent contractors

• 21:46Li Jin?

  and they work on commission. Open Door actually, I think, employs people to go show these homes to help homeowners sell their homes. And they actually, they're super managed because they actually purchase the home from you if it doesn't sell within a certain timeframe.

• 22:04Frank Chen?

  Yeah. One of Alex, who's the general partner who led the investments thesis, is that 10 or 20 years from now, we'll be buying homes from corporations rather than people because corporations have processes, technology they can bring to bear to sort of make sure every experience is awesome, right? They're pricing it right, they're staging your home right, they're showing you the right houses. Whereas individual operators, individual agents, may or may not have the experience to do that.

• 22:33Li Jin?

  Yeah, exactly.

• 22:34Frank Chen?

  It makes total sense that, like, we would professionalize.

• 22:36Li Jin?

  Right.

• 22:37Frank Chen?

  Things like that. And for the most important things in our lives, right? What house do we live in? Who takes care of our parents when they can't take care of themselves? Or who watches my kids while I'm at work, right? All of these things really need this era of management, right? You need to trust the service provider, you need to trust the company bringing you just the right set of people.

• 22:57Li Jin?

  Yeah, exactly. I think of managed marketplaces as a model that works really well if the existing customer experience is really suboptimal in like a less managed version. Managed marketplaces can often create an extremely wow customer experience that you would have never been able to accomplish before had the customer still been looking for a service provider on, for instance, a listings platform.

• 23:25Frank Chen?

  Right. And it's a lot easier for them, right? You don't get a list of six people that you then have to call and interview and schedule and figure out how to pay and all that.

• 23:34Li Jin?

  Yeah, exactly.

• 23:35Frank Chen?

  Yeah. Awesome. So that was sort of the managed marketplace era. Well, that was a fun romp through history. We like to do these historical walks through all of the types of investments we made. So we talked about listings from the 90s, we talked about the unbundled Craigslist era, we talked about the Uber for X, and then we sort of have just been talking about managed marketplaces where you need higher trust for the service provider on the other end. What are you excited about now?

• 24:02Li Jin?

  Yeah, so we ask ourselves that every day in thinking about what investments we should make.

• 24:09Li Jin?

  And so, zooming out of this entire historical picture, I think our, one of our observations about the best marketplace investments and the best marketplace companies are that they tend to be marketplaces that have more constrained supply than demand. So the biggest marketplace outcomes that we've seen, or the biggest marketplace companies that we've seen, are ones in which they've always struggled to keep up with the demand on the platform. So really good examples of that would be Airbnb and Uber. So during the entire lifetimes of both of those companies, there's always been an excess of demand and supply has always struggled to keep up.

• 24:53Li Jin?

  There's never been enough homeowners who are willing to open up their homes because there's just so many consumers who want that kind of authentic travel experience. Similar with Uber, ever since that business model was pioneered, they've always had more trouble, relatively more trouble finding drivers than finding consumers who actually want to ride. And so we think that this is actually a really key insight because we think that the best marketplace companies, the ones that can grow to be the biggest and really scale, are going to be the ones that tap into some sort of latent demand that already exists in consumers, that consumers already really want but are unable to access today.

• 25:34Li Jin?

  And so, in other words, we're looking for categories where there's a lot of demand that is being unfulfilled today. And so that leads us to what we think is next, which is regulated services and marketplaces that address regulated services. So we, the reason why we zeroed in on regulated services is because by definition, there is an artificial constraint on supply in those service categories.

• 26:07Li Jin?

  There's regulation that exists to basically be able to say who is able and who is not able to provide this service to consumers. And so we think those are really promising categories for people to build companies in.

• 26:21Frank Chen?

  What are some examples of services, the service industries that are regulated?

• 26:26Li Jin?

  Yeah, so you can go onto the Bureau of Labor Statistics and it'll actually give you like a huge, long list of all of these professions. But the major ones would be law, accounting, medicine, and healthcare. Those are the major professions where a huge majority of those service workers are actually regulated and licensed.

• 26:51Frank Chen?

  A member of the bar or certified by the state of California.

• 26:54Li Jin?

  Exactly, or exactly has passed the medical board of California, et cetera. And then you get down to actually like professions where there's not as much risk to public health and wellness if the provider is not certified. So for instance, in some markets and some geographies, florists actually need to be certified in order to operate a flower shop and to practice as a florist.

• 27:20Frank Chen?

  I didn't know that.

• 27:21Li Jin?

  And people have been driven out of business because they can't afford to go through that licensing process. Another example of this is in many geographies, interior designers actually need to be licensed in some way in order to decorate other people's homes. So there's categories where we think licensing makes a lot of sense to protect the public health. And then there's categories where we think that there could actually potentially be an unlicensed supply pool that meets a lot of the latent demand that exists out there.

• 27:58Frank Chen?

  So a little like Lyft when they thought about gee, do you need a taxi medallion in order to drive your car around? Their argument was, no, you have a car.

• 28:06Li Jin?

  Precisely.

• 28:07Frank Chen?

  You have a proven track record of safety because you haven't gotten into any accidents. So why don't we let you drive your car around?

• 28:13Li Jin?

  Exactly. And without that kind of unlock, without expanding it to the unlicensed pool of potential drivers, there's no way that Uber or Lyft could be anywhere near the size they are today.

• 28:27Frank Chen?

  So it sounds like the first thing that we're looking for is we don't necessarily have to have consumers learn a new habit, right? It's something that they do already. Get your haircut or whatever, or go get a medical procedure, right? So it's something that you're doing already, but we think that a crop of startups will make this experience better. And why don't you provide some examples? How will regulated service marketplaces make your life better first as a consumer of these services?

• 28:58Li Jin?

  Yeah, I think we've all had experiences where we have tried to get access to a service provider in a regulated industry and we come upon challenges in accessing that service. So for instance, it might take months to get an appointment with a particular specialist in a medical field. Or perhaps it's cost prohibitive for you to receive interior design services because those people need to be licensed. So this shows up in the consumers' lives in the form of like decrease in access, higher prices that you have to pay, longer waiting lists, or just generally having difficulty accessing that type of service.

• 29:42Frank Chen?

  Yeah, so that definitely sounds better, right? Ready supply, I can meet with somebody sooner, I can have the same service at a lower price.

• 29:50Li Jin?

  Yep.

• 29:51Frank Chen?

  So what are the challenges startups have in making all that magic true, right? Because they need to unlock the supply side for that to be true.

• 29:58Li Jin?

  Right. So there's a few strategies that we walk through in the blog post about how can marketplaces actually unlock the supply side in these regulated industries. One of them is in some industries, it's enough to just make discovery easier of these licensed providers. If today there's no resource for people to turn to in finding licensed providers, potentially just having a place where you can go and be connected to them.

• 30:30Frank Chen?

  Back to listings.

• 30:31Li Jin?

  Exactly.

• 30:32Frank Chen?

  Or good search.

• 30:33Li Jin?

  Right, good search. So an example of this is like Zocdoc, which helps consumers find doctors. Previously where consumers had to call and leave voicemails and try to make appointments, they can now do that online. So that's one strategy. Another strategy for unlocking supply is

• 30:55Li Jin?

  Is hiring and managing the providers themselves as a platform?

• 30:59Frank Chen?

  Go full stack.

• 31:00Li Jin?

  Exactly.

• 31:01Frank Chen?

  Make them your employees.

• 31:01Li Jin?

  Right, so the managed model. And under the managed model, the outcome of that is that companies can really raise the level of service and quality and accessibility and make these service providers a lot easier to access and have a consistent standards bar for them to really elevate the consumer experience.

• 31:25Frank Chen?

  Got it. So down this path, you're likely raising more money because having them on your books is more expensive rather than having them as contractors. But your argument is that in some of these marketplaces, you have to have that extra increment of quality for the service to work.

• 31:42Li Jin?

  Right. Yeah, potentially, there is just huge variance among the licensed supply pool such that a managed marketplace can really come in and relieve consumers of that problem of finding the high-quality supply. The next strategy that we identified in unlocking supply is companies that are augmenting or expanding the licensed supply pool. So a perfect example of this is Lyft or Uber. So expanding beyond the current pool of professional livery drivers, who are other people in the population that could provide a similar service, perhaps not the same one

• 32:22Li Jin?

  that consumers had experienced before, but something that met the same consumer need. And you can apply this model to other industries as well. So a startup that we recently met with is building a marketplace for buying a dog as a pet. So I don't have any dogs, but what I've heard is that it's just really difficult for consumers to find a good-

• 32:50Frank Chen?

  A perfect dog?

• 32:50Li Jin?

  Responsible dog breeder and a healthy dog. Exactly. And a lot of the dog breeders live in other geographies or other states, and it's not easy to go and visit them. And so what Good Dog is doing is it's actually finding these breeders in the middle of the country that otherwise you couldn't get connected to and creating its own set of standards as an alternative to the licensing that exists today, which doesn't necessarily map to what consumers actually value. And expanding the supply pool there.

• 33:23Frank Chen?

  Got it. So rather than the traditional license, you trust Good Dog to find a good breeder. And you trust that because they're doing all the vetting and all of the standards and all the training, that the breeder you get connected to will be first rate.

• 33:37Li Jin?

  Yeah. And then the last strategy to unlock supply in these regulated industries is using AI and automation.

• 33:47Frank Chen?

  Yay, machine learning.

• 33:48Li Jin?

  So probably your favorite category.

• 33:50Frank Chen?

  My favorite.

• 33:52Li Jin?

  So today, in order to be a truck driver in the US, you have to be licensed. And this is the most common profession in most states in America. And yet, we're still extremely supply constrained when it comes to the truck driving industry. There just needs to be a lot more drivers on the road than there are today. And so there's a number of companies working on self-driving trucks to relieve the industry of that supply challenge. I've also seen startups that are applying AI to dermatology and skincare. So there was a company I met with recently called MG Acne in which you take a selfie of your face and it uses AI to detect pimples or discoloration.

• 34:36Li Jin?

  And instead of sending that to a dermatologist to review, they use AI to review it and to determine the correct product, the correct ingredients that you should be using on your skin.

• 34:46Frank Chen?

  That's a clever way to get around the constrained supply of board-certified dermatologists.

• 34:52Li Jin?

  Exactly.

• 34:53Frank Chen?

  You can sort of replace the basics of what they do, right? It's not like an advanced thing. It's sort of the very basics, what creams and what should you do with your skin.

• 35:03Li Jin?

  Precisely. And for companies that are not issuing prescription ingredients and asking consumers to go get a prescription, I think that kind of solution is perfect because otherwise, there's a lot of people out there who would never have the time or the resources to go to an actual dermatologist for help.

• 35:25Frank Chen?

  Got it. So of all of the companies that you've sort of met in this regulated services arena, what are some of your favorite categories? Like, what are you personally excited about?

• 35:36Li Jin?

  Yeah, so personally, I am really excited about the mental health and wellness space. So, therapy today is a product that is really what I think of as a luxury product. So, mental health is something that people can really only access if they are really wealthy and have extra time and money to pay for it. Most psychiatrists are out of network, and so they can charge prices of, you know, even $600 an hour. And so, this is really a product that most people can't access.

• 36:15Li Jin?

  And so, I'm excited about the potential for a marketplace that perhaps leverages AI and automation, perhaps make some of the service delivery asynchronous rather than synchronous, for a marketplace to really come in and improve access to that kind of service. And I've seen a few attempts at this so far that I think are pretty clever and interesting. So, one startup based in San Francisco is called Basis, and they're using unlicensed but trained people in the community to basically serve as therapists.

• 36:55Li Jin?

  And so, the concept is that you have an hour-long conversation with someone who isn't actually a therapist, but who knows the techniques to help you identify problems or actually tackle issues in your life that can really improve your mental health and your happiness.

• 37:14Frank Chen?

  And so, let me, let me play a board-certified psychotherapist and give you the emoji, which is, "Oh my goodness, we're really gonna let untrained, unboard-certified psychotherapists out into the wild and like, coach people? Like, what could possibly go wrong?" So, how do we answer that question?

• 37:34Li Jin?

  Yeah, I think that's going to be something that they need to tackle and really figure out how to address because I think there is definitely a limit of the types of cases and the types of patients that the unlicensed but trained model can be applied to. I think for a large portion of the population, that's just going through everyday stresses and anxiety, having someone to talk to is just a great resource in and of itself.

• 38:03Frank Chen?

  Right, so maybe we pick patients that aren't most at risk.

• 38:07Li Jin?

  Exactly, yeah, and perhaps for the patients that are most at risk, they can be then routed to or sent to, you know, an actual licensed therapist.

• 38:19Frank Chen?

  So, mental health is one category you're interested in, what else?

• 38:24Li Jin?

  Yeah, another vertical that I think this kind of regulated service marketplace could be really applicable to, one that is very near and dear to my heart, is something in the beauty space. So, a lot of makeup artists today have to be licensed as cosmetologists and they have to go and receive that licensing and pay the fees associated with that. And so makeup artistry is actually also a luxury good today. So, we think of celebrities as people with, you know, makeup artists that they have following them around, doing their makeup every day. Maybe for special events, women will splurge for a makeup artist, but on a day-to-day

• 39:05Li Jin?

  basis, we have to do our own makeup and we have to figure it out ourselves. And so, how can we give every woman that experience of having a personal makeup artist who helps them look their best and feel confident every day? I think that would be a really interesting application of AI and a lot of the themes that we talked about in this conversation, to really improve access to that kind of service.

• 39:37Frank Chen?

  Yeah, I love the idea that you're going to democratize the service by bringing the prices down, and I love the idea of sort of maybe makeup artists building their brand by YouTube, and so like you're kind of following them anyway, and then now they could actually do your makeup. I think that's a great idea. And then if you could offer encouragement to entrepreneurs who are starting companies in this space, what would you say to them?

• 39:59Li Jin?

  Yeah, I would just say that there's so many opportunities that are still left. There's a table in our blog post with the various regulated professions and examples of startups that are tackling them. And you'll probably notice that many of them are actually blank and don't have any names listed next to them. And that's because we couldn't find any startups that are doing it. And so I would just say to entrepreneurs that there's so many service categories that are still left to build marketplaces in. Marketplaces for services are a huge opportunity because services represent 80% of the US GDP.

• 40:42Li Jin?

  And so there's a ton of opportunities left, and if you're working on any of them, we'd love to hear from you.

• 40:47Frank Chen?

  All right, Startup Land. If one of those empty spaces in the blog post has your name on it because you can't stop thinking about it, you're at the breakfast table and you're like, "Oh, I can't stop thinking about this," then think about starting a company and come see Lee. Well, thanks, Lee, for joining us in this hallway conversation.

• 41:06Li Jin?

  Thank you.

• 41:07Frank Chen?

  We hope to do lots more of these. And so if you liked this conversation, leave us comments, subscribe to the channel. And if you have thoughts about what other hallway conversations you'd like to hear about, let us know. So, see you next time.