Pricing Strategy for Startups

Transcript

When I see these, I feel them viscerally because they're so painful. So, one is a worry by founders that what they're building isn't going to perceived as valuable by their early customers. And so, the effect of that is that they really underprice them not appreciating the impact of what they've built. So, they're they're building towards something that is just too small to be worthy of the investment and pain that they're going to go through to create it. And having been on those journeys myself, I wince when I see that.

Hello and welcome to Web3 with a16z. I'm Robert Hackett, and today we discuss pricing strategy for startups from traditional businesses to Web2 to Web3. Topics we cover include unit economics, understanding consumer psychology, using on-chain data to inform pricing decisions, common mistakes and how to avoid them, how to navigate a pricing pivot, and lessons from real-world pricing case studies, including Tesla, Nvidia, and others.

Our experts include a16z crypto's Maggie Hsu, head of our go-to-market team, research partner and Harvard Business School Professor of Economics Scott Kominers, and head of our CSX startup accelerator Jason Rosenthal, who's a tech veteran having spent the last 25 years at various internet companies. The three combine all their different expertise around the theme of this episode. As a reminder, none of the following should be taken as tax, business, legal, or investment advice. See a16zcrypto.com/disclosures for more important information, including a link to a list of our investments.

Maybe we could start with you, Jason. What's the first thing you should do as a founder when you're thinking through pricing strategy? So, I think that for founders, particularly first-time founders, pricing can be this really amorphous, scary, confusing topic because if you've never brought a product to market before, it can be hard to understand where to start. And so, I think that bringing a little bit of order to the chaos is a really helpful thing to do. And so, one place that I like to start is just understanding where your product fits in relation to everything that's come before it. Are you bringing to market

some dramatic, transformative, new capability or innovation that nobody's ever thought about needing or or buying before, or are you introducing a massive improvement on top of an approach to product innovation that people are familiar with? And I I think depending on how you answer that question can be a really good strategy and thread to pull on to figure out how to navigate the pricing journey. So, we're going to talk a lot about how things work in the crypto industry, which kind of makes me wonder then, how do things differ between that versus more traditional businesses or say other tech businesses, Web 2? Like, how does

pricing in that world differ from the Web 3 world? And do things from Web 2 apply in Web 3? I would say two things. One, I I'd love to hear Scott and Maggie's perspective on this because I know they've thought deeply about this question. But the way I think of it is that one of the most exciting things about Web 3 is it really opens up the pricing canvas in a way where the Web 2 landscape was much more, I'd say, constrained and atomic. And I think in Web 3 with things like transactions and tokens and shared governance, you have all these new levers and tools around business models and pricing that just

didn't exist on the internet before. Yeah, I would just add that to Jason's point, pricing is closely tied to business model, and there are a number of different web three business models. So, pricing could refer to more traditional SaaS pricing, which I know Jason's thought a lot about, or it could refer to something like mint pricing, which Scott has thought about. And in some cases, pricing is really set by the company or the entity, and in other cases, it's set by the market. So, there are all these different types of frameworks that could potentially apply given the breadth of business models in web three.

Yeah, totally agreed. And maybe just to riff and build on that a little bit, as an economic theorist, when you're thinking about price, right, what you're thinking about is clearing supply and demand. And a lot of the challenges Jason was talking about around like a completely new product that's never been launched before is you don't actually know what price clears supply and demand, right? You have no idea what total demand for your product is, what people's willingness to pay are, and that makes it really hard to determine price. But one of the things that's really special about web three with all these different instruments that go into sort of the functioning and structure of a business and the relationship between the customer and the business and and its products is that there actually many different levers and dimensions on which you can model pricing, right? And you can price

different components much more granularly than before, and you can price in multiple units, right? Like, you know, often one might sell an initial product in currency, you know, whether it's fiat or cryptocurrency or something of the sort, but then maintain your product ecosystem using some internal unit of account, which makes balancing supply and demand much easier because it's relative only to your own ecosystem, not so much to the outside, you know, sort of world of potential competing products. Now, you talked about this in terms of a balance of supply and demand. Presumably, if you're putting out a product, you have a pretty good sense of the supply side of things, but it's the demand side that is the real question here.

How do you go about getting a sense for what the reception might be for a new product? First of all, some of this goes back to the classical insights of Web 2 tech and tech before that and just ordinary product marketing, right? Like you can experiment, you can smoke test, you can show people versions of the product and try and estimate their willingness to pay, but of course nothing really becomes true demand until your product enters the market. Web3 creates a lot of mechanisms for doing sort of multi-stage sales, for releasing an early product to your earliest adopters, building them in as your core initial community, and then gradually expanding your offering as, you know, you test the product on them and test

out their demand energy. But also, we shouldn't forget that market mechanisms are a big part of how we determine price, right? Sometimes price doesn't just come down from on high. You might sell things in an auction or you might sell a limited supply and see if you sell out and use that as a way of sort of measuring market demand. And I guess the last important footnote is across the board, and I know this is not always how we do product launches, right? We don't always think about it this way, but from the perspective of pricing, it's really important for the value proposition to be very clear, right?

Like market demand responds to an understanding of what your product is and how it creates value. And so if you want to get a good estimate of the market price, you actually have to do a very good job of communicating what your product is so that people can decide how much they're willing to pay for it. And one thing that Scott touched on that I hadn't thought about before, but I think is one of the most powerful and under-discussed aspects of Web3, is that if you look at historically how internet business models have worked, some of the most powerful, exciting companies have been really network effects businesses.

So, Craigslist and eBay going back to the early days and then later Amazon Marketplace and then of course modern social products and multi-sided marketplaces like Airbnb and Uber and Lyft. I think one of the most fascinating aspects of pricing and business models in crypto and web 3 is that for the first time we have really network effect oriented businesses and primitives like token oriented networks with the first truly native pricing mechanisms to go along with them. And so, I think we're seeing early stages of the compounding effects by layering the network effects of protocols on top of tokens and new aspects of IP and ownership around NFTs

and I think that we're only at the start of some of the amazing things that we're going to see as you bring those things together. That's pretty fascinating. Maybe we could like back up just a touch and talk more about the foundational aspects of pricing and more about network effects and how important those are. So, a network effect is this idea that the more people join a network, the more value it creates for everyone. Or is there something you would add, Scott, being the economist here? No, I think that's a pretty good definition. A business with network effects is one in which the more people engage in the business, use the product, are part of the platform, whatever the case, the more value it creates in aggregate. I guess the key thing to

remember as you're thinking about that is it's in aggregate. So, sometimes network effects are negative for some participants, right? Like newspapers are two-sided platforms that connect readers to advertisers and the more advertisers there are, the less happy the readers might be because they'd rather be reading articles than a lot of ads, but the more readers there are, the more advertisers are happy to reach them. So, you can have negative individual network effects, but the point is that the overall value grows. And so, platform or a network business is one in which the more participants you have, the more users of the product, the more the aggregate value the business creates increases. And Scott, correct me if I'm wrong, but I think typically as you increase the number of endpoints in a

network, you get an exponential scaling of value rather than a linear one. So, the notion of essentially built-in compounding growth to the overall value of the network is a pretty powerful one. Yeah, they often have non-linear value, especially when the network gets very, very large, the value provided in total gets much, much bigger. So, think about your data network effects with a platform like Facebook or your information network effects on a platform like Twitter or your user network effects for a currency, right? Like currencies are not super useful when only five people are willing to exchange them for goods and services, but once you get to 10,000, now suddenly there's like a liquid market to some degree. And once you get to, you know, a million or or a billion, now suddenly this is a real medium of exchange that

many people can use. Exactly right. Now, some of these examples that you raised like a Facebook or a Twitter or even a Google are interesting because in a sense some of the products appear to be free to the user. There are people who are being charged, of course, and those are the advertisers to reach various demographics and segments, but when it comes to a typical regular everyday user, nobody's asking you for a cost. You just log in and you post whatever you want or you type a search into Google. So, how does pricing intersect with some of these biggest Web 2 networks that we think of?

Well, I think this has been one of the topics that's been most controversial and has been one of several compelling and propelling factors for the rise of crypto networks, which is that you have a pretty big disconnect between the value to the end user, i.e. the consumer in the case of social networks or products like Google, and then the interest of the groups that are actually footing the bill, typically the advertisers. And so, one of the more complicating factors is that you can end up with a pretty big misalignment between the interests of the various constituencies. And something that I'm super excited about, I think we're all excited about in crypto and Web3 is through governance, pricing, and

business models, a unique moment in the arc of technology to really align and make transparent the interests and the needs of all the different constituents. I would just add that in these situations, you're essentially at the mercy of the platform fee or the take rate, and that can increase and keep increasing, and you aren't able to extract the value that you've created as as a user, participant, or advertiser, and that's really stuck in this network. To the point around how these networks are monetizing, I feel like we've seen some of these feature-based add-ons or, you know, paying a monthly subscription for a certain type of feature.

But that in my mind sometimes works and sometimes makes sense, and at other times it feels at odds with the purpose of the network. Let me quickly just jump in and say, first of all, 100% agreed that sometimes these monetization strategies are very much at odds with the long-term value growth of the network itself. But also, and again, this is the tension between Web2 and Web3, there's a lot of short-run incentive for these platforms once their network effects have hit that escape velocity, right? They've gone nonlinear, there's so much value to being on the platform that you sort of have to be there.

The platform now often faces strong incentive to extract from its users, right? And so, you see a lot of things that were free while the network was growing become something that's costly, often in a way that's not necessarily total value maximizing, but much more like platform surplus extractive, or you know, say maybe platform surplus maximizing, but but necessarily user surplus maximizing. Chris Dixon talks about this a lot. He talks about the attract extract cycle. You know, at the beginning when a new platform is trying to attract people, it might seem like it's offering a good deal. People are getting a lot of benefit out of it, but then at a certain point, you flip over the S curve, you make your way over the inflection point, and all of a sudden the platform is actually trying to

extract as much as possible out of you. But by then, the network effect has gotten so big that you're kind of trapped in there. Especially people like make it hard to exit. Yeah, you can't afford to to not be there. Yeah. And I think I mean having been crashing at the creation for a number of these network effects businesses, I can tell you for certain the fact that this dynamic exists was never There was never a malicious intent at the beginning. No entrepreneur sat around and thought about, "Okay, well, how can we get people in and keep them here and make them feel trapped?" I think that it's really been an economic limitation in terms of the available business models to companies and entrepreneurs, right? Software engineering and infrastructure and

servers and bandwidth and storage all cost money, so you need a way to bring an audience and then generate revenue around that audience. And again, one of the most exciting aspects of crypto and Web3 is it just really opens up the canvas and the tool set that entrepreneurs can look at and draw from to build these network effects businesses. That's such an important point that there isn't this like insidious, you know, backroom plotting to flip the switch into extract mode. But there's actually a bit of a survivorship bias that goes on there where the networks that last the longest are the ones who are smart about their ability to monetize because that makes them more competitive against other networks. And

so, there's sort of this self-reinforcing prophecy or feedback loop there going on. Only the ones that perform in this way are the ones that we see and dominate today. And I think about the protocols. So, these protocols are open source and the concern is, well, anyone could use this protocol, what's to stop them from creating this new front end or app? But, at the end of the day, the apps that are successful are providing additional features and values over that protocol and and so people are willing to pay something because there's this differentiated benefit. At the same time, they're not necessarily locked in. Yeah. And indeed, Jason, let me push a little bit on this idea that there's sort of nothing

malicious. Like, there's nothing malicious, but there are market forces, right? And your value for a product, once the network effects kick in, the thing becomes really valuable, your value for the product goes up. And so, one's willingness to pay goes up, the platform can afford to extract more, right? It's sort of like price is relative to willingness to pay and the value you get for the service. And then to Maggie's point, you know, that protocols change this dynamic, this is critical. And to me, this is one of the most exciting things about Web3. And actually, Marco Reuter wrote a great piece about this for the a16z crypto service, that if you give your users part ownership and a reason to like want to invest in using your version of the platform, and especially if you give

them some degree of governance, right? They can stop that effect. It's actually like a commitment to not expropriate them when the platform gets really big. And similarly, the opportunity to move from app to app on the same protocol, same thing, right? It It puts some pressure on the app to not expropriate the users. And so, Web3 is creating these new mechanisms of preventing pricing from taking advantage of users. And what's wacky about that, you know, you're like, wait, why would you want to do this, right? Doesn't that mean we make less money, but that's not actually necessarily true because now the users have more incentive to be invested in the product and to use it actively, right? You are more willing to engage and make the thing essential in your

business process or in your day-to-day life or something of the sort if you know that it's not at risk of suddenly costing a ton tomorrow. And so that enables, at least in theory, these platforms to grow much bigger and be much more value-creating overall than the sort of classical walled-garden platforms were. Yeah, Scott, I think a lot about growing the size of the overall proverbial pie or the ecosystem and even though any individual player might have less of that, you're really growing this together. Exactly. Well, and I think that if you look at the history of economics, and Scott, you know this much better than me, but inventions like double-entry accounting and in technology, the notion of giving

equity stock options to early employees or in enterprise sales forces, the invention of the uncap sales plan where as sales people sold more and more, there was no upside limit on the amount of commissions they can make. All these things were hugely powerful economic forces in the evolution and development of what we think of as modern companies and products. And again, I think we're on the cusp of perhaps the biggest revolution in this that we've seen today. It's funny to think about the uncap sales opportunity as an invention. Was there a certain company that like pioneered this or So I mean, Oracle's really credited with inventing this around database sales in

the early 1980s, and it used to be the case that as a sales rep, you had your base salary and a commission plan, but at some point, it didn't matter how much you sold, you made the same amount of money. And Oracle introduced the uncap sales comp plan, and what they saw was that the most innovative, most aggressive sales people all wanted to come work at Oracle because there was an opportunity to out-earn what they could make anywhere else. And if you look at the history of competition among databases, this was one of the things that really set Oracle apart from everybody else in the market at the time. And it's now the model that is replicated by just about every modern enterprise software company.

An incredible recruiting tool. I guess I've heard of it from the uh probably less savory name of eat what you kill. Uh I don't know if that's the same concept, but yeah, laws of the jungle at work, the software jungle. Yeah, that's right. I mean, I guess the key, just to bring this back to pricing, the key is that all of this stuff works because of incentive alignment, right? Fundamentally, it's about aligning incentives at every level of the chain. This uncapped sales commission is about aligning the incentive of the sales people with the organization. And similarly, token holding and token-based governance is about aligning the users and the platform, right? The users of the platform have the same overarching goal for the platform to succeed and be

made better as the platform does. And this is all very and it all acts in some sense through price, right? Like pricing and allocation and again, whether it's in fiat or tokens or time or whatever, is really very much about helping the market achieve the socially optimal outcome to the greatest extent possible. So, Warren Buffett has described pricing power as the single most important decision in evaluating a business. What does that mean and why is pricing power so important? So, I don't want to speak for Warren Buffett and I haven't heard the specific quote. He's probably describing a version of market power, like the opportunity to control the price seen by the market for the product would be my guess. Yeah, that's exactly

right. A company's ability to raise prices and yet still retain sales. Yeah, I think I would respectfully disagree with the Wizard of Omaha. I think that that that will be true in established commoditized markets where there's little differentiation between the core underlying value of uh if you take for example an insurance company, if it's life insurance or homeowners insurance, they're all fundamentally selling the same basic value proposition to their customers with maybe a few differences around the edges. But I think in the world that we live in and part of the reason why we're all attracted to what we do is that the history of the technology industry is really about looking forward, trying to bring to

market breakthrough innovation where you're fundamentally changing the kind and type of value that you can deliver to your customer, be that a a business or a consumer. And I think in a market like that, it's less about a competition or ability to influence prices and more about the ability of your product to deliver unexpected upside or value. And when you can do that, it does open up all kinds of flexibility in terms of how you create customer aligned pricing and value in your product or service. Yep.

I've thought a lot about this in terms of pricing assets that are essentially the same but have different brand value. And so different types of wine might be priced completely differently. Are they essentially the same thing to Jason's point, handbags, other luxury goods? And I think in that case, often times it is the brand, but there are all these other things that might go into that purchase because it becomes an identification tool of your affinity for a certain community. In the one example, it would be your taste in handbags, and in this case, you become part of and you're part of that social ecosystem. Yeah, absolutely. I guess just a footnote on all of that is

I think it's really all about value. Like somehow community value, brand value, product value, transformative innovation, these are all about value. And like yes, at some point a company has to be able to capture value as well, but fundamentally, I think all four of us are really interested in companies and products that bring new and very significant forms of value to the table. And yes, you do have to figure out how to capture some of that, but it's much more important that that value number be high than it is that like you have a lot of direct capture power on the margin.

Scott, you talked about this on an episode of the podcast, episode 25 on network effects, moats, and the business of Web3, which people should check out. And you got into all about the defensibility of businesses and how to remain competitive. You also talked about vampire attacks in there where somebody tries to basically steal your market share or your community away for their own product or service, but you didn't touch on pricing. So, how how does pricing factor into this equation? It seems like it could be difficult for a lot of these newer Web3 projects in a world that's largely open source and there's a lot of permeability between these different communities, it could seem difficult to actually have pricing power, that maybe prices would trend

down over time for certain products and services. Absolutely. So, in that episode, we talked about competitive advantage and sort of pressures on your ability to maintain a sustainable competitive advantage as a Web3 business. And vampire attacks, as you mentioned, are a good example. The possibility that an individual user can take their digital assets and just move them to another platform makes it much easier for new entrants or just, you know, incumbent competitors to try and incentivize users to switch. And so, you might see much more pressure, Maggie was talking about this earlier, on an individual platform than we would see in the Web2 world. And of course, that comes with pressure on price as well, right? And we've seen in a couple of cases really significant competition for

customer loyalty in token trading marketplaces, for example. And one of the things that's happened there is that prices have gotten competed down in the midst of that. But by the same token, pun definitely not originally intended, wow. Um, puns encouraged on this podcast. I'll say it again cuz it's a good pun. By the same token, Web3 has, like with open source, this potential for truly embedded protocols and technologies, which do actually have mechanisms for value capture. And similarly, Maggie was talking earlier about applications competing on quality, right? Competing on features and like try to make themselves things users want to use. I know what a wacky idea that like you

know, you're actually going to compete to make your app better and and more user-friendly in some way. But all of these are things that can be sources of competitive advantage and they can be very significant sources of competitive advantage and give you new pricing opportunities, right? If you have a protocol that's embedded in everything, right? If you're a an NFT minting platform or, you know, a chain data processing platform that everyone is embedding into their own ecosystems, there's a native transaction layer on a blockchain and your protocol, like all of the others, can take small shares of everything that's flowing through. And the more embedded you are, the more powerful that becomes, right? The bigger the network you create, the more value you can build around your embeddedness and eventually sort of price that.

Similarly, if you build an app that users really, really like to use some core protocol, if that core protocol creates more value, like we've been talking about, right? Being the lead app or one of the leading apps is again really powerful and is a source of pricing opportunity. The way I think about it is like somehow Web3 requires pricing that is more aligned with users' interest and with the use of these different networks and protocols and platforms, which again is good from an economic and from like a market and user experience perspective, but requires a new type of creativity around where you actually extract that value.

It's interesting you talked about this concept of embeddability and how that can provide a source of competitive advantage. But if we think about a protocol and its ability to capture value, if a group of people or whatever thinks that a protocol is capturing more value than is warranted or maybe it's getting greedy or something, presumably it would be a simple matter to swap that out for another protocol. So, it seems like there's a bit of a balancing act or a tension there when it comes to the protocol trying to deliver value and yet also to capture value but also to deliver that back to the community in a way that people find good.

Well, so I think what you're referring to is a really live active discussion going on within many of our portfolio companies that I think is going to play out in a very powerful way, which is that when I talk to our founders and CEOs, they're all looking within their given network, what are the places where they can add the most value to their customers. So, it could be in the sequencing layer, in the consensus layer, in the messaging layer of a given protocol. And they're thinking very, very carefully about how do they price those different aspects of the architecture in a way that encourages the kind of adoption and embedding that Scott was talking about in a way that helps them build these

very large, very high-scale networks over time versus prematurely rushing to extract value before they achieve takeoff velocity. So again, I think that the search for what I'll call customer-aligned value is one of the most powerful and interesting product development forces happening right now within crypto. And nobody has figured out the ultimate answer or the best way to do this. There's a a ton of Darwinian experimentation happening here that I think are going to lead us to some pretty interesting outcomes that we can't even see from where we are today.

Are any best practices emerging? Well, so I think that there's a lot of excitement and interest. Nobody has done this yet, but two features of blockchains that are very powerful are composability for sure. So, being able to use different components to build on each other and interoperability amongst chains. And nobody's yet figured out a pricing or business model that really capitalizes on those two features, but I think that we're not far away from that and that's going to be very powerful once that emerges. I want to back way, way up and ask how many different ways are there to price? Like what does the landscape of options look like? Jason Maggie, you wrote you wrote a wonderful piece about this, you know, six questions every founder should ask about

pricing in which you enumerated a whole bunch of these things. Yes, I would say it's not infinite, but it's very, very expansive. Maggie and I came up with at least nine different factors that people should look at and consider from user-based pricing to transaction based to value-oriented pricing to subscription-based. And you can look at the post if you want the complete list, but I think that two interesting things that come out of this are for most products, there's going to be one or two ways that feel the most natural and aligned with what you're building. And and I would say founders should follow that intuition and instinct and not try and reinvent the wheel where it doesn't need to be.

But one of the most powerful things that people can do, I think, is come up with a combination or hybrid set of pricing approaches with the goal of being maximally customer aligned with how they come to market with their pricing. And what I mean by that is is a founder, if you can make your customers feel like they're underpaying for the value that you're delivering to them, that's a much, much better way to go to market than if a customer feels like they're being taxed every time they try and use your product or service.

I've been thinking a lot about how Web3 companies can use some of these pricing models in ways that take advantage of the fact that there's additional data. So, one example would be outcome-based pricing where let's say you're running a marketing campaign. If you're running a campaign on a set of wallets, you would not only know if they have your NFT, but you could also see what other NFTs they have. Whereas in Web2, if you're running a similar campaign against a set of customers, you would know that they purchased your product, but you wouldn't necessarily have that accurate data around the competitive products or really the share of their wallet that you're capturing.

And so, there's some really interesting things that you get into here because you have this available data source that you can see the effects of some of the pricing and marketing decisions that you're making, and you can take hopefully more accurate and nuanced actions based on that. So, there's more transparency and data available from which to make some decisions about how you might want to price products or market them to people in Web3 is what you're saying because a lot of these transactions take place on blockchains, and those are default publicly visible to people. Am I getting that right? Exactly. So, not every action is non-chain action that you're driving towards, but in many cases, you are seeing the result of these pricing dynamics and user behaviors on chain. What's an example of

like what that would be in real life in practice? Like what's an example of some data you could take from a blockchain that might inform the way that you strategize around this? So, what I was thinking about is for an insurance product, you could basically take the asset value of a customer's wallet, and then there's some value of the assets they're trying to protect through this insurance. That's one example. Cool. Yeah, Jason and Maggie, I've been trying to sort of like just pop everything you've been saying into frameworks in my head, and I I feel like I don't have a complete answer to the question of like what are all the different ways to price, but putting together the two things you said made me think of the following.

You know, at some level, when you're deciding how to price a product, and we started at the very beginning of this conversation talking about uncertainty, right? Like, what do you know about the market? What do you know about people's values for the good or service that you're providing? At some level, a lot of the pricing decision is about that uncertainty and how much of it you know you know the answers to going in versus how much of it you have to sort of elicit from the market. And so, one dimension as you're thinking about how to price is this, like, how much information do I have? Can I like set a fixed price based on my own knowledge, or do I need to use sort of a dynamic sales mechanism to elicit, you know, people's value for the good or service?

And then another dimension to think about is is the sale one-off, or sort of a single sale or or a single sequence of sales, or is it dynamic, right? Are people going to be showing up I have an on-chain game, people are going to be showing up wanting to play that game at any given time. It's a real problem, you know, somebody finds out about the game by watching someone streaming and they're like, "Oh my gosh, I want to play that." They like install the software and then suddenly there are no units for sale, you know, you you cannot join the game. Sorry, dude. So, many products actually want streaming pricing. You want to be able to sell in real time.

And so, another question is, do you have one-off sales or sort of like big bulky multi-unit sales, or do you have streaming dynamic pricing? And so, combination of like, how do you actually set the price for each individual sale unit, which is some mixture of like how much data you have and how much data the market is sort of bringing in when people buy and sell, and then when do those sales occur? Are all the sales at once? Are they sort of like in real time over time as more people figure out about your product? I'm an HBS professor, so I must draw a two-by-two. That'll be my two-by-two for the day.

And I would add that I think a technique that founders often underestimate is their ability to do some, you know, CIA-level espionage work to understand the state of the art in the market, how other similar projects have priced, and extract learnings from that, and then reconfigure that into their own approach to pricing that makes sense given their particular offering or market. But, I think I see people make two mistakes often. One is to think that they can science their way into the absolute right way of doing it, where I think ultimately all of this is about gathering data, using intuition, and then ultimately taking a leap of faith,

seeing how that works, and then learning from that, and revising it. So, that's one failure mode to avoid, which is believing that you can scientifically arrive at the mathematically correct answer. The other failure mode, I think, is just throwing up your hands and saying, "Well, like this is unknowable, it's impossible." And just throwing a dart at the wall, and hoping that works. When in reality, you want to use every signal available to you, apply a logic and reasoning framework, and then put forth the best initial version that you can in the hopes of finding the Goldilocks zone for your pricing.

Absolutely. And critical to convey to founders is you shouldn't be afraid to pivot your pricing strategy even very early on. Right? Sometimes you get it wrong in one dimension or direction or another, and like it is really okay to use what we think of as qualitative feedback, like the you know, the market reacts by saying like, "Wait, what are you talking about? That is not the correct price for this." In either direction, again, maybe you have like 10,000 people who want to buy your 15 units, or you know, you have zero people who want to buy your 500 units or something of the sort, or just like price turns out to not work with your unit economics. We teach in our entrepreneurship classes at HBS this case on the early days of Zipcar when they you know, they didn't think through the that class. Yes, you did.

that. I remember that case, yeah. So, you know the punchline. Zipcar starts selling the opportunity to do short-term car rentals and they pick some price and what they discovered immediately was that people were renting them all day because their price for a full day rental in effect was lower than the price of renting a car at a rental agency. And so this whole idea of this like short-term ride hop sort of model was undercut by their pricing model. And the case deals with like first you discover this, right? It's like how do you recognize this out of the data? Like you notice people are not renting the car the way you were expecting. You thought people were going to be renting them for an hour at a time. Most of the rentals are full day or multi-day. Wait a minute, something's wrong here.

But then, you know, the question is what does that mean for your business? Well, the rental car agencies were renting at pretty competitive prices relative to their unit economics and so this is just wearing your cars out much faster than you expect. And what did they do? They have to pivot. Maggie, what was it like in the classroom when you were doing this? Like was there stress around this pivot decision? Like were people like, you know, oh my gosh, do we do it? Like you know, is this company going to melt on day zero? I think there is that hesitation once you've set a price to change it and I I fully agree with that point of being flexible and being iterative in your approach because often times it's not going to be correct. Whatever the first thing you try is not necessarily what makes most

sense. Exactly. And like just to wrap the Zipcar story, the hilarious thing is that they noticed this very quickly. They just changed their pricing scale and adopted like a daily rate model where, you know, if you were renting for the day, you were paying the same amount as if you were renting a car at a at a rental agency. And it worked fine, right? And that company is still here today. I'm sure there were some unhappy people when that switched over, but Absolutely. And people are unhappy when you switch over like that, but it's still the right decision for the business and they still appreciate the value of your product, right? Zipcar at minimum is like another competitor for rental cars. And so you can't be that unhappy, right? You you can be unhappy that you were getting a,

you know, below-average cost deal, right? Yes, the deal has gotten worse, but it's still actually good to have the product out there. It's better than if the company implodes. So, this is like a perfect transition, Jason, to you. I mean, you have dealt with this issue firsthand as the CEO of an early social media company, Ning. Tell us about your experiences actually pulling this off. Yeah, so Ning was a platform that let people create a vertical social network around any area of interest in their lives. So, things like Indian Premier League cricket or early drone builders or fans of the teen vampire novels, Twilight. We had vertical social networks on all of those and many, many more topics.

And the business started as an advertising-oriented business model where the idea was that we would foster the creation of all these different social networks, which would create a large audience, and then we would run socially relevant and contextual ads over on the interests and passions that people had within those various vertical communities. And what we found was that just simply didn't work at all. The areas of interest and the users and customers were too fragmented for that to be an interesting advertising business model.

But, what we found was that the creators themselves, the people who were building and managing these vertical social networks, felt a tremendous amount of pride of ownership in what they created. They wanted a very specific set of features that we would work to build, and so we flipped the business model to be a subscription-oriented one around our creators. So, we would charge them a low monthly fee to be able to create, operate, and run these vertical communities. And if we would offer new capabilities or ways for our creators to monetize, that turned out to be a much more aligned and sticky business model than trying to show ads to the end

users. And exactly like Scott said, when we first rolled out the plan for this change, we gave our community owners about a four-month heads-up that what they had previously been getting for free we were going to start charging them for and there was at first loud voices of outrage around that. A few people but not a lot ended up leaving the platform because they weren't finding enough value in running their communities. But the vast majority of people ended up staying and choosing between an entry-level, an intermediate, or a pro-level plan and the fact that they could self-select into what pricing model they wanted based on the set of

features that we were offering them just delighted them to no end because for the first time in the history of the platform, our incentives and business model were directly aligned with our most engaged and passionate users. And what we found was that the outrage and the dissatisfaction quickly faded once people saw that we were giving them a better product at an affordable price. You know, a lot of this pricing talk, although it centers on economics, supply and demand, you know, this sort of math equation, it sounds like a lot of it also depends on consumer psychology and that messaging is a key part of that. You know, Jason, you said you gave people like a four-month heads-up before

you were going to make this switch. I don't know how Zipcar handled it when they were going to change their pricing up, but how do you make sure you don't alienate or frustrate your early users when you're going to pull off a change like this? Well, so two things. One, I think you hit on a really important point as it relates to crypto, which is that for crypto networks, I think they are the most community-oriented, community-aligned kind of technology that's existed so far. If you went back and ran the statistics on this, you would see that the most successful projects are the ones that do the best job of engaging with and aligning behind the needs of their unique specific community from the

get-go. I think we've seen this many times in projects in terms of layer one and layer two protocols. And so, as a community manager or owner or entrepreneur it's critically important to be clear and transparent with your community from the get-go in terms of what you're doing and why you're doing it. And that is even more true at the time you decide to make a change. So, if you're going to change your pricing model or change the entitlements and capabilities that your community is able to enjoy in your network or in your project or on your platform it's essential that you communicate the why and the how behind whatever

change it is that you're making. I would also not underestimate the power of support, especially hands-on support. So, I think that some of our companies think about discounting and I try and reframe it in terms of additional value that's being added. And oftentimes that goes to reliability and support levels, so SLAs. And so, I've seen success in even just offering a company offering a Slack channel with the devs to talk through problems. And having that real-time support is really meaningful.

SLAs, Maggie, for those who don't know. Service level agreements. And having that sort of responsibility and access to reach someone at a protocol or a company is huge, especially when this is early and there might be bugs that are identified or something that needs to be updated. Just having that real-time communication is so important. Yeah and the psychology point is really critical here, right? You know, people's value for a service and their willingness to pay for it, which again are not exactly the same, right? They're closely linked, but they're not synonymous, and we often play fast and loose between the two of them, right? The value is the value you get from the service. Willingness to pay is sort of a you know, relates that and

your ability to pay. You know, if you're a kajillionaire, you might you know, have high willingness to pay for lots of services just because money isn't worth that much to you because you have a very large amount of it or something of the sort. And so, it's important to keep track of both of these cuz they both feature into the pricing calculus. But, people's value for a service and their willingness to pay for the service both depend on their perception of the value they're getting from it and their willingness to pay in particular for a a platform service, right? If you think about platform transaction fees and things like that, a lot of these sort of small fees that crop up in many different contexts in Web 2 and Web 3 and beyond, you know, the user is often thinking about the perceived value add of the

platform in assessing, you know, what their willingness to pay should be. So, to the extent the platform can make salient the value it's providing and make the users you know, sort of like understand and internalize that value, the higher a transaction fee or a take rate one can usually achieve. This is really fascinating because the difference in information that's available between traditional businesses and sort of what we're seeing is possible with Web 3, where you have all these on-chain transactions, all this data, and you can use that to inform your strategy, your go-to-market strategy, your pricing strategy.

Absolutely. This goes to what Maggie was saying about how seeing people's on-chain data and on-chain identity is really informative for thinking about price and sales strategy in general. It helps you see who your potential customers are, you know, look for a community that's full of like-minded, similarly energized individuals and try and recruit them in. And again, like it lets you see how much they've paid for things, which might give you a sense of how much your product is actually going to be worth in the market. So, I I've got to ask this. We talk about price discrimination, and that's a common tactic. You kind of charge based on who your audience is or your possible customer is.

But, that ability to discriminate based on data that's out there, maybe there's also a possible negative side to that. Absolutely. No question, right? Like, if somebody shows up to buy a ticket to your concert, and you see that they have a bunch of ridiculously expensive NFTs, you know, maybe you say, "Well, you know, the price just doubled." Like, um Yeah, yeah, exactly. We're going to charge extra now. No, it's true. And this is a subject for another day, but I think a big part of how crypto has to evolve, and is currently evolving, is towards giving people more control over which specific data assets are seen by which applications and used in which different ways, right? Like, you know, maybe your holdings as eligibility tokens for a mint, you would reveal

different information than what you're showing to someone who's going to decide on an ad hoc basis how much to charge you for painting your house or something of the sort. But, also, you know, as an economist, I should add the footnote that price discrimination is actually good to a large degree in markets, right? We actually do want goods and services to be able to reach many different people at many different price points. This idea of tiered pricing, you know, sort of different services, service plans as a function of different levels of usage, that's good. And so, there's a fine line that we have to be careful about, but overall, to the extent that markets can involve more personalized pricing, it actually moves us closer to economic efficiency and towards everyone being able to get

access to products, where otherwise, you know, if instead you're pricing house painting in a market where, you know, some people have exceedingly high ability to pay, and and other people don't, you actually might price pretty high to try and extract more rents from the people with the high ability to pay, and then lots of other people just can't afford to paint their house at all. So, there's tension there, and it's not obvious that we don't want some amount of price discrimination and tiered pricing to come out of this, and think of that as a feature. It seems to me that maybe it would be better to lead with the carrot versus the stick here. Absolutely.

Jack it up the prices for somebody, maybe the trick is maybe we'll give you some more discounts or Yeah, they get elite they get a fancier product in one fashion or another. Absolutely. Interesting. And again, I think Web3 with the new competitive forces really supports that, that we're going to see at least I expect much more differentiation and price discrimination on features and quality tiers and various uses of the different products and protocols rather than just sort of purely on like someone's ability to pay for something. So, you're all meeting with founders out there. I'd like to know what are the most common mistakes people are making or common misconceptions they have when it comes to approaching pricing.

It's funny and Jason, Maggie, I'd be super curious if you're seeing this as well, but I see a lot of both too much reinventing the wheel and not reinventing the wheel enough. Right? There's like a lot of teams that go very far out of their way to build a very complex like totally novel pricing architecture when something much simpler that already exists would work just as well. Right? You see a lot of, you know, introduction of like a new very complicated auction mechanism where what you really wanted to do was just run like a standard ascending auction or a standard descending clearing price auction or something of the sort. Right?

So, sometimes there's too much focus on the nuance of the situation of the product or service you're selling in a way that leads you to build something that is ultra customized for that, but really truly like you know, sort of a much simpler and easier to interpret and more tried and true pricing mechanism is just as effective. But on the other hand, they're sort of not reinventing the wheel enough. You see a lot of reasoning by analogy, like, oh, this company sold this similar product. We're going to sell our product the same way, even though your customers are very different. Or, you know, your product has some critical difference that means that those two mechanisms, you know, sales mechanisms should actually be different. That's interesting. So, it's almost like, on the one hand, overly engineering, creating things that are

too complex when something off-the-shelf would work. And by the way, that has the added benefit of your target market would understand the pricing already, presumably, if you picked something simpler than something that's like totally customized to your product. And then, it's funny that the other one, it's a little bit paradoxical to, you know, say that people aren't thinking out of the box enough then. All right, Maggie, let's hear from you. Biggest mistake founders are making. The one that I've seen is pricing something as a static price and not forecasting what it'll look like for the customer and user over time. And so, if you're giving a potential partner customer price based on a certain number of users that they have or a certain number of members

in their community or a certain number of usage and not taking the time to do the work and understand if that number goes up or down, how does the price that they're paying you, whether it's monthly or, you know, some set increment is changing. That's one that I've seen actually a few times. It's kind of thinking of price as this one-time thing versus this ongoing thing that can be actually forecasted and estimated with different scenarios for the customer. Now, in order to do that, you'll have to know your unit economics, you know, pretty well, so that you understand how you scale. And so, you're saying you want to add some flexibility into it from the start, so that you can enable yourself to grow into it or allow your

customers to grow with the product and not just capture yourself early and sort of set their expectations at something that is unreasonable long term. Exactly. Yeah, I see two interrelated challenges that when I see these, I feel them viscerally because they're so painful. So, one is a worry by founders that what they're building isn't going to perceived as valuable by their early customers. Like, you know, maybe the product's not finished yet and so it doesn't have the full feature set that they know they want to build or they're worried about being able to out flank a much larger competitor. And so, the effect of that is that they really underprice the offering in a way where

they're giving too much value to their early customers and not appreciating the impact of what they've built as entrepreneurs. And the the follow-on effect of doing that is that they can find themselves locked into a business model and approach to pricing where if they imagine and extrapolate how large the market that they're in can reasonably get, even if they imagine that grows by a factor of 10 or 100, they'll never end up with a business that justifies the intense amount of work that they and their employees and their investors are going to put into building the company. So, they're they're building towards something that

is just too small to be worthy of the investment and pain that they're going to go through to create it. And having been on those journeys myself, I wince when I see that. Yeah, it's a really great point and uh you wrote about this. You cited Bob Iger, the CEO of Disney, talking about the best business advice he'd received, which is, you don't want to be in the business of selling trombone oil. Right, because even if you gained 100% market share in trombone oil, there's just not enough trombone players in the world that would ever make you want to be in that business. Queue uh sad trombone sound. Womp, womp.

Alternatively, who's doing things right? Who are you impressed by? Do you want me to jump in or Yeah, yeah, yeah. Scott, hit us. Okay, let me quickly give an an example of reasoning by analogy that's not quite right. So, for example, we often see people selling membership tokens, right? In an online community with some purpose, right? You know, it's a collector collective or a you know, creator hub or something of the sort. And often the sellers of those memberships choose to sell them as if they were, you know, limited edition collectibles or artworks. When in fact, membership is something that's often more ongoing and has the property that people like will discover the product and and say, "Oh, I want to join this collective." and go looking for a membership.

And so, you'll see teams reasoning by analogy. They're like, "Okay, well, the last time this artist sold their own images in an art drop, here's how they did it." and then they copy that for their membership sale and this doesn't work at all. You're selling a completely different product. Even if there is this image attached, you know, your ideal buyers are not the ones who are buying it for the sake of the image. They're the ones who are buying it because they want to be part of your membership club. And so, maybe knowing that it has an image, you know, that's attached to an artist who's had some sale history is useful for reasoning about what the eventual price might be, potentially.

But really, you want to be thinking about the value of the membership package and selling it a way that's consistent with that being the product. This goes to an earlier point you made about big ticket one-off sales type items. Something that would be kind of more equivalent to what you see in like a hardware-based product, you know, like you sell a car or something or a fridge. And it's this big item, you sell it one-off, somebody buys it. And sure, that's a type of business model, but what's even better is if you can get a subscription going where you get continual revenue from people. Absolutely. And it's funny to kind of see this transition happen in something like the NFT world where, you know, for a while when this was really capturing people's attention, you know, there were

these one-of-ones going for big money and selling, but now maybe NFTs can be all different sorts of things. They can be memberships in clubs and perhaps that hardware style business model of selling the big ticket item is not the way of the future for many of these products. It reminds me of printer cartridges and razor blades, which I think might have been another case study. Absolutely. Oh man, printer cartridges are the worst though. Those things last like a day. I mean, it's a great business, I think, cuz I'm constantly having to buy them.

To answer your question on who does pricing well, I'll actually not respond to that part, but I think it's who's communicating well because to Jason's point, a lot of this relates to how you're communicating this and a lot of projects will do something and forget that they actually need to over-communicate the same point about the change or the update or whatever that pricing modification is more times and in more channels than they would imagine to the point in the advertising adage if you have to see something seven or eight times before you start to process it. I think over-communicating in Web3 is typically a good thing because people want that sort of access

to understanding what's going on behind the scenes because that's part of why they're there. And so having that element to me is really one of the best practices of pricing. It's not actually about your pricing model, but it's about how you're communicating it. That makes sense for something where community is so important that like actually there is no such thing as over-communication. You know, you want to communicate as much as possible since community is really the foundation. Exactly. An example that really sticks out in my mind is Yuga Labs with Bored Ape Yacht Club where they did the initial mint, it went better than they ever expected and they followed it up with the Bored Ape Kennel Club where they essentially gave every existing

BAYC holder their own dog. And the I mean the story behind it is really interesting. They did it in part because they thought, well, you know, if you actually were a Bored Ape and you were hanging out at this Yacht Club and you had everything, what might you want to have? And one answer was, oh, you'd want a pet. So, you'd want a dog. And so, that's what they went ahead and created. But, the way that they did it was that they gave 100% of the royalties from the follow-on sale of that. So, every Bored Ape holder got a Bored Ape Kennel Club for free, but then the royalties from secondary sales of those all went to support animal shelters and other, you know, good causes around animal welfare.

So, they had the double effect of rewarding their community for being part of the early journey and using this momentum as a way to raise visibility and raise funds for a worthy cause. And I think that had an impact of just really increasing the loyalty that the community felt to what they were doing. That's cool. I didn't realize that there was also that giving back element to the Kennel Club. You know, you raised Bored Ape Yacht Club also in your post about this where you use it as an example of a good better best pricing framework. You know, there are ways where you can price a more basic level of your product or perhaps a less premium version of it lower than the more prestigious or fully feature set version of it. I think one

of the classic examples of that would be GM selling Chevrolet brand as the sort of lower end, Buick as the middle tier, and then going all the way up to uh the Cadillac as the high-end version. Yeah, I think that good better best strategy is one that applies across industries. Actually, going way back when I was at uh AOL, the whole mentality of the company was that there were brand buyers, so people who would pay a premium to be associated with what they perceived as a high value brand.

And in that case, it was AOL. And then there were people who were price-conscious shoppers, and that's what the CompuServe offering was for. And it turned out that the way the company ran them was it was exactly the same technology and infrastructure underneath both services down to, you know, just about every single line of code. But by applying different branding and pricing models to them, they found that there were two non-overlapping customer sets that they could go after independently. And I think same thing, you know, I'm dating myself here, but with MTV and VH1, you know, same production infrastructure, two different brands going after different parts of the marketplace. And I think we're

increasingly seeing that in crypto with BAYC being a good early example of that. That's funny. Maybe a more recent example, I don't know if this is exactly equivalent to what you're saying, but I know some of these new vehicles that come out, you know, they're stocked with all sorts of sensors and stuff. It's not the hardware itself that you're paying for. It comes like that, but you might pay more to get access to some of those sensors and extra features and AI self-driving whatever. For sure, right. To get a full self-driving Tesla, it's not even a software upgrade, it's a software unlock. They used to charge a big upfront $10,000 additional fee for it.

Now they've actually moved to the subscription model for that since such a high percentage of Tesla owners lease their car and don't necessarily shell out all that money upfront. Yeah, that's a powerful good market model when it works with your product. I feel like this gets a little bit into pricing psychology, and the economist Dan Ariely has done some really interesting experiments with pricing subscriptions of The Economist magazine and having print and web plus print and having a third pricing option and simply by adding a third decoy price, he was able to influence the choices of the breakdown of people using or purchasing one or the other. So, there's some interesting schools of

thought around just pricing psychology in general. Something that I don't know just reminded me of the former CEO of JCPenney. When he took over, he was in charge of Apple's retail stores. Yeah, he's a good friend of mine. Okay, yeah. So, he developed the whole strategy for Apple's retail stores. He went to JCPenney and he was like, "We're going to simplify things. You know, if they have all these coupons and discounts." He was like, "We're going to do away with that and just make things like list it at the price that it should be." You know, a very Apple philosophy. What you see is what you get. And the thing is that the customers kind of revolted. Turned out that like the discounting and couponing was a feature not a bug for them. It's like they totally loved that sort of gaming aspect of the shopping experience there.

Yeah, I think that was a big example of assuming that the tastes and preferences of a customer base from one business i.e. Apple applied universally across the spectrum and just really misunderstanding the psychology and the needs of the JCPenney customer base and ultimately ended up costing him the job. Yeah. So, it's a good thing to study some of this pricing strategy stuff up front to save yourself some pain later. That's interesting. I've got to ask all of you Well, Jason, because one of the things that you and Maggie talked about in your post was that most people actually underprice their products. Like generally speaking, people are selling themselves short. And you could be charging more for what

you're selling. How do you overcome that barrier to know actually what your product is worth? So, I think that's the beauty of the internet age, right? Is that there are all different kinds of ways to cohortize what you're doing and run a set of experiments to do price discovery. So, Scott talked about different auction models that enable you to do that and we're seeing that more and more. If you're doing enterprise sales, which tends to be more direct to your end customer and more of a hand-to-hand combat style of go-to-market, you can literally divide your prospective customers into two different cohorts. Think of it as cohort

A and cohort B. Show one set of pricing to the first cohort and then do, for example, a 5x uplift on that same pricing and show that to a different cohort of customers and just see what happens. And likely both pricing model A and pricing model B are not quite correct, but you'll likely zero in on your optimal pricing based on the feedback that you get from each of your cohort tests and and you know, bring it out to the consumer web. You know, companies like Netflix and Facebook and Google have done you know, what's classically known as AB testing for many, many years, where you show different offers to different prospects and see what works best. So,

lots of modern data-driven tools for arriving at the optimal price. So, you're actually charging five times more in one cohort. That's an example, but I think that to get the most learning out of it, you want to actually make the difference between the two or more options fairly extreme, so you can understand if you're way, way too low or way, way too high and probably ultimately end up meeting somewhere in the middle. And just quickly before we go to you, Scott, cuz I see you want to say something. How do you make that decision of that trade-off of perhaps more limited market share, but the ability to get higher prices and get more money from the people you're selling into? Typically, you're doing this very early in the evolution of your

company. So, the biggest value of your early customers is to understand what you're doing that's working and what you might be doing or thinking of doing that's just not going to work. So, you're valuing the learning and the data much more than you are the ultimate revenue from that early group of customers. Got it, Scott. You wanted to say something. Yeah, and just also reflecting on that same question, the other thing to remember and companies make this mistake all the time, it is really scary to raise your price because in expectation you're going to lose some customers. Right? Like the supply and demand curve, you know, is such that unless you're pricing on the completely inelastic part of the demand curve, if you raise your price, you're

going to lose some customers. But the goal from a revenue maximization perspective is to find an opportunity to raise your price where you don't lose so many customers that you don't make more money on net from all the customers who stay. And so, one has to not just think like, "Oh my gosh, like if I raise prices people are going to leave." You know, they will. The question is, what fraction of them leave? And that's part of what you're trying to learn from these tests. When should you cut price? Like when would it be justified to slash prices? Since we've talked about raising them. I think there are a couple different scenarios. So, there's a big opportunity to reduce prices to capture more market share if you can take advantage of a market dislocation that other companies

may not be able to. So, we actually saw this with Tesla where they, due to their supply chain advantages and due to some regulatory changes around what income levels could have access to different federal and state rebates on electric vehicles, they proactively went out and cut their prices just as a bunch of new entrants were coming to market with their EVs as a way to maintain and grow their market share and they were able to force a situation where just about everybody else had to respond to their move. I should probably out myself and just say that yeah, the Model Y was slashed 20% and uh and they got me.

I was like, "Oh, that's a competitive price. All right." There you go, right, it worked. I think you know, Maggie spent a good amount of time at Amazon and early in the history of AWS, Amazon was famous for at their big customer events, they would call out the number of times that they had reduced prices on their core cloud services in a given quarter both as a way of demonstrating how customer aligned they were, but also as a as an opportunity to seize additional market share. So, I think those are great proactive and aggressive reasons to cut price. I think sometimes you're forced to do it if the market changes on you unexpectedly. If you're in a physical

goods business and you suddenly find yourself with excess inventory and you know that you need to sell through it before a new generation of product comes to market. This happened to Nvidia not with their highest end GPUs which faced, you know, huge demand for AI training and inference, but for some of their gaming cards when the economy softened a little bit. They were just about to come out with a new generation, but they had massively overbuilt their prior generation. So, to clear their channels, they did a big price cut. Wow. So, I've got to ask, what product out there would you pay five times more than it's listed price right now? Like what product do you love so much that

you would pay five times the price? I have a jacket that I have worn basically continually all seasons for almost 10 years at this point. Wow. I've had it warranty like heavily repaired four times. I have gotten far, far more value than the original price and the original expected lifetime of this jacket. I mean, like I I sent it back to the company again to have the lining re-stitched and they they write to me and say, "Don't you want us to just give you a new jacket in place of this one?" I'm like, "No, this one fits and the new ones you guys make do not." It's a little bit It's a hindsight bias because, you know, I I didn't know it was going to be this great 10 years ago when I bought it, but it's easily something that would have been worth

five or even 10 times as much to me. Like, you know, sort of because I've just used it non-stop and it's it's been perfect. Wow, subscription jackets it sounds like. You're You're sending it back to get it tailored and refurbished. Continuing in that vein, I think that products where you just get so much more joy or utility out of them than you ever would have expected are the ones that you'd pay more more for. For me, I'm a big cyclist. I have a number of different bikes and I have this old mountain bike that I've had for over 20 years that I just love and have had, you know, so many great rides on it. And at the time I bought it, it felt expensive, but if I knew at the time

that I would get as much joy out of it as I have, I would have, you know, paid whatever price they would have wanted to charge for it. Wow, that's great. My example of this is when I was formerly a journalist, I used an AI transcription tool and that was the best. I remember being like, "Whatever these people are charging, I will pay 10 times whatever it is. This is like changed my life so much for the better." That's actually a really good example. Like, a lot of the products in this category are actually things that are saving you time or enabling you to be more successful, right? That they pay for themselves more than five or 10 times over in whatever it is that they're doing that that enables you to go forward, you know, whether I guess to go forward either in

your career or on your bike. That is until the AI takes over and replaces the writers entirely. That's a subject for a for podcast. Okay. All right. Well, thank you all for joining us. Great to be with you all. Thanks for the opportunity. Thanks for having us. Yeah, thanks so much.