Composability beyond software: building ecosystemic portfolios - with Sangeet Choudary



Composability beyond software: building ecosystemic portfolios - with Sangeet Choudary

Sangeet Paul Choudary returns to the show and unpacks some key insights from his latest work — ‘The Building Blocks Thesis’ — and how it relates to composability and strategic investment in ecosystemic portfolios.

Podcast Notes

Sangeet Paul Choudary is the founder of Platformation Labs and the best-selling author of Platform Revolution and Platform Scale. He has advised the leadership of more than 30 of the Fortune 500 firms and has been selected as a Young Global Leader by the World Economic Forum. Sangeet’s work on platforms has been selected by Harvard Business Review as one of the top 10 ideas in strategy and has been featured thrice in the HBR Top 10 Must Reads compilations. Sangeet is appointed to advisory boards and committees at several Global 2000 firms and government bodies, including the ING Group, the MAS’s ASEAN Financial Innovation Network, Boson Protocol and Standard Bank Group, South Africa. Sangeet is a frequent keynote speaker at leading global forums including the G20 Summit, the World50 Summit, the United Nations, and the World Economic Forum.

In this episode as we delve into Sangeet’s recent report — and despite it still being early days — how the building block approach can play out in the real world economy. We also discuss why the fundamentals in VC will change, the nature of composability, the role of DAOs and empowering the end-stakeholder.


Key highlights from the conversation

We discussed:

  • Markets evolve towards more composability as more and more of the value chain is digitized
  • Traditional extraction models (eg: economies of complements, data lock-ins, etc) are put in trouble
  • Value extraction will move at the venture investing and fund layers
  • Organizations will need to embrace a more DAO-like structure and become more context-rich (3EO comes in handy).


To find out more about Sangeet’s work:

You can get an illustrated copy of the Web3 Bootstrapping Playbook, launching in September 2022. Sign up here to get early access to the playbook as soon as it releases.

Other references and mentions:

Find out more about the show and the research at Boundaryless at

Thanks for the ad-hoc music to Liosound / Walter Mobilio. Find his portfolio here:

Recorded on 9 June 2022.

🌐 Boundaryless Conversations Podcast is about exploring the future of organizing at scale by leveraging on technology, network effects, and shaping narratives. We explore how platforms can help us play with a world in turmoil, change, and transformation: a world that is at the same time more interconnected and interdependent than ever but also more conflictual and rivalrous.


Simone Cicero:
Welcome back, everybody, to the Boundaryless Conversations Podcast. As pretty much always with me there is my usual co-host, Stina Heikkila.

Stina Heikkila:
Hello, everyone.

Simone Cicero:
And today we have a returning guest to our podcast, and a friend, Sangeet Choudary.

Sangeet Choudary:
Thank you, Simone. Always pleasure to be back here.

Simone Cicero:
Well, the pleasure is all on our side because it’s always a great gift to have you back because you bring us so much insights, Sangeet. And your work has always been so pioneering and so interesting that we really thank you for your time that you want to spend with us and with our listeners, with our community. Today, we are lucky to save some time with Sangeet to talk about his latest piece of work, “the building block thesis” – a report that is a culmination of months of research and several blog posts and public posts that you have been sharing that have focused around Web3, the impacts that Web3 and blockchain is having on platforms and ecosystems thinking. So, maybe Sangeet, as a starting point, I think it’s wise to give you some proper space to give an overview of your work, your latest report.

Sangeet Choudary:
Absolutely. So, when I think of the building blocks thesis, I think of it as having a larger perspective than Web3 itself. So, Web3, the big manifestation of the building blocks thesis. But the key idea of the building blocks thesis is that you can create components of value coded in software, and that’s why I call it digital building blocks. But you can create components of value that can be then recombined in different ways to create solutions.

Now, that is something we associate a lot with Web3, because unlike the traditional centralized platform model, where a company would create a platform, and then the market would align itself around it, Web3 allows us to build infrastructure, market infrastructure around a protocol, and then serve it to different types of markets. So, Web3 is certainly an example of the building blocks thesis. But the reason I abstract it a little beyond Web3 is because very often we think of Web3 as having two simultaneous aspects. One is the composability of building blocks, but the other is the financialization through tokens.

And so what I want to focus on with building blocks is specifically the idea around composability of value creation, how that value gets distributed, how that gets captured, that’s a fundamentally independent access. And in Web3, they are combined through composability, combined with token financing. But you could apply the same thing to the rise of digital public goods, where essentially taking the open source software thesis to the phenomenon. Today, countries and foundations are now putting digital public goods out there in the open so that new innovation can be done around these public goods. And that again, follows the building blocks model. The reason I call that out separately is because it’s not financialized in the same way that Web3 innovation is, but it would have its own way of funding and value capture, and so on.

So, my key point over here being that the focus areas around how to drive rapid innovation and solution design and solution creation, by creating a universe of building blocks. And again, then how do you create the incentives or the overall funding mechanism to ensure that this universe is self-sustainable? And so those are two different aspects, which I’m happy to go into deeper. But that’s the fundamental idea of the building blocks thesis.

Simone Cicero:
Just maybe riff a bit with these initial points. When you say the composability, right, is probably the leading element that has been guiding your research. So, what is new around composability? Why is now the moment to make this transition from traditional ways of organizing markets into this building block highly composable thesis? So, we had composability since a while, right? Open source software, API’s. What is happening? What is the new thing, which I guess it’s related to Web3, but maybe not limited to that. Also, from a systemic perspective, maybe, what are the drivers and the enablers of this building block thesis?

Sangeet Choudary:
So, the idea of composability in software has always been there since we had object-oriented programming and then open source. So, composability has been fundamental to the idea of building software. What we’ve seen is that the incentives for creating and publishing software and then driving its usage that requires some form of value capture. So, while we did have the open source movement, there was no value capture inherent to the open source movement. If you look at a lot of value capture that happened around open source, it was in the complements to open source software, it was not in the usage of the software.

So, you could sell a service around open source and that is what, for example, Red Hat did with Linux. So, the value capture happened in the service. Or you could build critical performance bottlenecks around the usage of open source and you could charge for that. So, think of Android as open source but then Google Maps was created as a complement to Android, if you needed to have an Android phone, and you wanted to use navigation, you had to use Google Maps, and then Google would sell a license to that. So, the value capture for open source has largely been around complements to open source rather than in the core value creation that’s happening around open source.

Now, what’s changed over the last 20 years in particular, because open source came up in the late 80s, early 90s, and so on. What’s changed, especially over the last couple of decades, is that we’re seeing digitization moving across the value chain. It’s not just happening in one corner in something called the software industry, but it’s happening across industries. It’s not happening just close to the consumer, it’s happening all the way up the value chain. So, digitization is happening all across. And as more digitization happens, value flows become digital and so they can be mediated through software.

The way we’ve done that over the last two decades, as this shift in digitization happened, was that we manage these value flows in exchange for the centralization of the capture that happens as a result of the value flows. And so there are a few different ways that centralization happens. When value flows, we capture data and we centralize that data, and then we package it and provide it as products to other third parties. These could be advertising products, targeting products, analytics. And so that’s the Facebook, Google and much maligned data captures, surveillance capitalism model, if you will.

So, we essentially created the incentives for harnessing large scale value flows through software by centralizing the capture using data. Other ways we did that was in the traditional software sense by driving lock in around the software. So, a lot of enterprise software adoption, for instance, a lot of what has driven the digitization of value flows, in general, has been predicated on greater lock in of usage. And over time, once you create that bottleneck into distribution, because you have usage locked in, you then aggregate all the other developers to serve users through you. So, a lot of what happens in SAS, a lot of what happens in enterprise software has been predicated on that capture of value flows through distribution lock in.

What is changing now is the shift towards thinking in terms of building blocks. We’re starting to see if there are other ways in which you can capture value without requiring either centralization of value captured through data centralization, for instance, or through distribution lock in. And that is where a lot of the current focus is on token based financialization, because that provides us a direct route to value capture that has its own set of challenges in the sense that a lot of token design today does not necessarily model the value that is being captured through the value flow -that’s often modelled or is often swayed by speculation. So, as long as the token is swayed more by speculation than by value capture, its price will not necessarily reflect the actual value that’s accruing towards the token. So, there are challenges towards that as well.

But the reason Web3 becomes important in this whole shift towards building blocks is that we now have alternative mechanisms to incentivize and fund this composability. And, to a large extent, if platforms of the previous era Web2 platforms, for instance, if they had other ways of monetizing value capture, which were equally scalable, they would have allowed more composability as well. Technologically, that could have been possible. But the fact that you needed to either lock in distribution, or centralized data capture, essentially incentivized these platforms to stack up all the complimentary so-called building blocks into one central platform. The more you could stack that up, the more you could do cross subsidization, the more you could offer certain value propositions for free and then capture data and exchange. So, essentially, because of all those considerations, we move towards the centralized model. But the more we can determine how to find the decentralized model, the more the approach towards building blocks will become important.

So, to really summarize this whole idea, there are two key things that I’m talking about. One, building blocks are now much more valuable – or composability is now much more relevant across industries because digitization has kind of permeated across industries. Value flows are becoming digital and software building blocks are going to be used for critical business processes. So, thinking about composability not just in the traditional software domain or what we call open source in one corner, but thinking about it across business domains is becoming important. And secondly, the incentives that drove centralization, they start breaking down if we have alternative ways to fund some of this composability. And that is where you either have the Web3 token based model, you have the digital public goods model that is sponsored by different sources, those are the new sources of funding that are emerging to make this happen.

Simone Cicero:
It’s fascinating because I was taking tons of notes in the background, right. And let me try to make some order for our listeners as well. Right? So, you said basically, we come from an era where open source has been essentially used as a go-to-market strategy, right, has been used, for example, through economics of compliments. You spoke about Android, for example, and Google Maps. But we can even say, Android, for example, with the Google Play Store, which is something that you can only use on the original Android, let’s say. Then data lock ins and the capacity that those platforms have accrued in data so that they have locked in users and capture this value. That’s really clear, I think.

So, this new building blocks thesis, it’s somehow related to getting to find new ways to fund these, as you said, because I think it’s really important. Because what I’m seeing now, it’s like protocols and Web3 projects try to somehow replicate the same protection and defensibility mechanisms in this new setting. So, for example, it’s hard to see protocols interoperating or converging. And I think me and you, we both share this idea that over the long-term, instead, we must see and we will see probably some kind of convergence in terms of ontologies between protocols, for example.

I wonder if these elements here kind of brings with it some kind of political elements, some kind of vision of society — related to the vision of society we want to build funding a new thesis where components can be built and financed in a much more plural way, I would say, much more diverse way – it may entail some kind of intentional policy related choice. How do you see this new landscape of financing software for the creation of value, and probably shared value; how should we expect that this kind of more collaborative landscape between this to emerge? I guess it’s also related to mechanisms of investing, right?

Because if we see now, for example, if you look into Web3, and we see investors, and this is something that, for example, Jack Dorsey brought up a lot lately that he said, Web3 is VC owned. If we see protocols coming up with investors investing tons of money into that, defensibility and centralization will stay relevant for these investors and will operate, I would say, adversely, to the trends that you are describing here that seems to be more collaboration, more integration, more modularity, more efficient ways of investing money into building infrastructure and building communication layers. I hope it’s clear enough, but that’s some of the points that I was thinking to.

Sangeet Choudary:
Yeah, absolutely. I think much as we are seeing innovation on the software front with the rise of Web3, I feel we’ll see more innovation around how this is financialized, and how this is funded and how that value is captured. And let’s think of it in a couple of different ways. So, with Web3, what I would posit is that even more than Web2, even more than any previous technology cycle, the winner take all is going to shift away from builders towards funders, or towards those who are financing innovation rather than those who are building innovation.

In Web2, what happened was that the platform centralized the gains and decentralized the risks. So, if I was a seller on Amazon, or maybe Amazon is not the best example, but if I was an early producer on any platform, and I helped the platform grow up. And the reason I don’t say Amazon, as the example over here is because Amazon initially started as the first producer. But let’s say I was an early developer on Twitter and through its adoption, I was an early game developer on Facebook and through its adoption. If I was an early producer on a platform, I invested and built my whole business on that platform, I drove the adoption of the platform, I took all the risks associated with the new platform only to have a bait and switch later on.

Essentially, what that meant was that the venture risk, to a large extent, was radiated to these producers and the venture returns were centralized by the platform. Now, where I believe this becomes even more interesting with Web3 is because I believe that the venture risk will now be radiated to the portfolio and the venture returns will be centralized with the fund. And the reason for that is exactly what I talked about in my recent article, when you have composability, you have the ability at a fund level, to do much more than you have the ability at an individual project level. At the fund level, you can reserve a certain part of your fund, and say this part of the fund is going to go into projects we find interesting.

Here’s another part of the fund that’s going to go into complements of those projects, which are interesting in themselves but even more so because they drive complimentary value creation across our existing portfolio. Here’s another part of the fund that goes into investing in new forks that are created. We will drive the creation of those new forks so that there is even more innovation around our existing projects. And so this part of the fund will integrate those forks back. And there’s a fourth part of the fund that will go into purchasing governance tokens in projects we can’t invest in, but which can enhance the value of our portfolio. And because we can’t invest in them anymore, we’ll just go ahead and buy governance tokens over there and influence the roadmap. So, my point is that with a fund in Web3, you have much more power to invest across the whole landscape of building blocks, rather than taking the risk of putting all your eggs in one basket, which is what the Web2 funding model, or generally, any venture funding model used to rely on.

The other hidden advantage of Web3 is that for all this talk about decentralized innovation that these funds talk about, the fact that you can have an early exit means that you can keep pumping and dumping. You can keep pumping a certain part of your portfolio. And then once you’ve used that to create complementary value in other parts of the portfolio, you can exit it and move those funds into other parts of the portfolio where you created that value. In Web2, the only way you could do that was you could just put all your bets against the winning horse and starve the other competitors and then acquire them cheap. That was the only way to centralize value as a fund.

But in Web3, you have many more interesting, innovative ways to centralize value, which is why I believe a lot of investors who are seeing this potential early on, they are drumming the fact that Web3 is not about data capture, it’s not about centralization, whether it’s behind the scenes, they’re really getting into it, because fundamentally, Web3 has a new way of centralization, which is at the fund level. And so that is what makes this super interesting if you’re an investor and if you’re investing right. If you’re just in it because Web3 is cool, you missed the fact that you can centralize so many more returns in Web3 compared to Web2 funds. But if you get that and that’s why you’re into it, there’s a big game to be played over here.

And that’s why regulation becomes even more important because there needs to be greater regulation around when you can exit a fund in Web3, what kinds of complements you can invest in. There needs to be a way to kind of start looking at monopoly power at the fund level versus just at the individual project level. And so those are key things that will increasingly get regulated. I think regulators still have four to five years before they can catch up on this, because I started seeing this in Web2 in the early 2010s. Regulators caught up with changes only a decade later. And so I see the same thing happening in Web3.

So, it’s important to understand that how we set up funds is going to be critical. Even if we do this from a public good perspective with countries financing this with foundations like the Gates Foundation, and others financing this, they need to think about how to fund the entire building blocks universe rather than how to create just avenues to fund projects. So, how you fund a universe is going to be very different from how you fund individual projects. And venture capital so far, has mastered the art of funding individual projects and assessing this at the project level.

I think with building blocks, with Web3, the advantage is you don’t need to necessarily assess risk at the project level beyond a point. You need to make a few bets and then start figuring out through composability how you would de-risk those bets by inserting them in as part of the larger ecosystem your fund is pushing forward or by creating the right complements, investing in the right complements and so on. So, the fundamental nature of venture capitalist is going to change. And until this is regulated, I would expect the next Facebook’s and Googles of the world are going to be the a16z and the big funds that emerge in Web3, not the individual projects that come out of it.

Stina Heikkila:
That’s interesting. And if we stay a bit on these funds, you mentioned a bit on public funds, potentially countries sponsoring this, and now in the end you mentioned those big players. But do you see public-private partnerships in this space? And could that work at the fund level like you mentioned? Could there be some elements of pushing in certain directions through regulation? You mentioned before, like the digital public goods models. I don’t know if you could link those dots a little bit in terms of the agenda. Do you think that governments will have a role in pushing for this in some way, or they will just be catching up, like you mentioned, and what you see there in terms of societal impact, maybe, a little bit?

Simone Cicero:
Maybe if I can add, Sangeet, before you start, even not just public bodies, but also different constituents that come from the public sphere. So, maybe it can be communities, it can be different types of players not generally related to industrial age governments if you understand what I mean.

Sangeet Choudary:
There’s a lot of money that wants to make the world a better place. And here, I really don’t mean it in the Silicon Valley way of justifying your ambitions. But really, there’s a lot of money that goes into charity each year, that goes into public development, etc. So, a lot of it can be channeled towards this. A lot of it is funding that’s been set aside by capitalistic entities whether it was earlier in the form of CSR – now, it’s ESG and other factors – but there’s a lot of funding that’s available that can be channelized in the right way towards funding these universal building blocks. There are a few different ways in which that could be done.

The first way could be really a version of an impact fund, but essentially creating a whole perspective on what a universal building block is going to be. Say there’s an impact fund that’s focused on solving the water crisis. And their focus would be on that ecosystem of different projects that are solving the water crisis. So, instead of following the traditional VC model of just identifying the right projects, this would look at a more composable view of here’s the overall universe of solutions that is required, here’s how they will interpolate and work together and drive new solution creation, and then start investing with that view. Where they start investing in a few solutions, then start looking at complements, start looking at ways to drive innovation around existing solutions and drive follow-on funding for those new innovations that get created.

So, the funding structures may not be very different from what they used to be in the past, but their approach towards funding would be fundamentally different. Their approach towards funding would be less about picking ventures, and evaluating risk and allocating capital, it would be more about mapping the universe of what those ventures are going to be, picking initial shots, and then figuring out what are all those gaps in that universe have compliments of other capabilities that need to be invested in, and essentially, investing with that mindset. So, it requires a fundamentally different approach to investing. Instead of just looking at it from the perspective of risk, you need to look at it from the perspective of what drives greatest coverage of the universe you’re investing in.

Because ultimately, the returns to your fund are going to be based not on the hedging that you’re doing across different types of ventures, but on the overall growth of the ecosystem you’re investing in. Because your portfolio is going to interpolate, your portfolio is going to interconnect and draw on top of each other. So, the fundamental risk return equation for investing in a building blocks universe is very different from how it would be when you’re investing in traditional ventures. I hope that makes sense. That’s the key difference that I see is going to be important. And we’re going to see a whole new era, a whole new generation of funds coming up that get it versus those that used to work on the venture model.

Simone Cicero:
So, very interesting. I mean, this riffs a lot and connects a lot with this idea of ontological convergence that I’ve been sharing with you a little bit and this idea of being more intentional, as you said, right? So, mapping out the space you want to build for and more intentionally strategic investing, I would say. Because if I think about traditional investors, maybe they are more into building optionality versus building with intentionality, I would say, right. So, they kind of build in many directions and some of the bets will work. Instead, the type of investments you seem to be advocating for are somehow more intentional in what they build in deciding what is salient for them to build. And I think an interesting point that’s raised with this is that for example, I can see the role of DAOs in this.

So, we are seeing a lot of emergence of investment DAOs, for example. And so far, most of them are asset management DAOs. And of course, we are seeing also DAOs invest in the treasury, something that’s similar to investment arm. So, grants and things like that. And this transition from optionality that is typical of VCs into intentionality and mapping and strategically designing what you want to build as an ecosystem really calls for some more democratic governance process, requires an organizational structure that so far, we don’t see in venture capital firms. Maybe we see that partially in companies that historically have been building ventures more like for example, Flagship Pioneering or Haier.

So, I see the need to also evolve organizational models of corporates, and in general governments, maybe to integrate more of the DAO perspective, right? So, more these permissionless elements and more these democratic elements of governance. So, what is your perspective on the organization that will be a protagonist of this transition? So, for example, DAOs, what do you think about them? What is their role behind this idea of building blocks protocols? What’s the role of the organization around these and how they can evolve and integrate these new perspectives?

Sangeet Choudary:
The role of DAOs becomes even more important at the innovation level rather than the investment level. Because at the innovation level, a lot of innovation is truly going to be decentralized. You’re going to rely on working across projects, you’re going to rely on creating complements to existing projects and scaling up by integrating across many different projects. So, to manage the incentives for such decentralized innovation, you need to have organizational structures that are closer to DAOs rather than the traditional organizational structures. Because developers would, instead of signing up for specific organizations, they would gravitate towards DAOs based on certain aspects. And then they would allocate more of their time and energy and resources towards specific DAOs based on how their participation is set up over there. So, I feel DAOs are very compelling from the perspective of harnessing developer commitment and developer energy.

From the perspective of investment, I still see that as a centralized play in a building blocks world at this point. I mean, investment vehicles, by nature, have always been “open” in the sense that if you want to aggregate funds from different sources you have to be open to those funds. But investment thesis don’t necessarily need to be decentralized. Investment thesis can be fairly centralized. And so you benefit from open access to capital, you benefit from centralization of investment thesis, and you benefit from decentralization of developer energy harnessing through DAOs. So, my view would be that investment funds don’t necessarily need to be super decentralized to map this whole or to participate and win in the Web3 world.

In fact, a smart group of investors working in a very centralized way, promoting the gospel of decentralization to developers and investing across projects could continue to centralize gains in Web3. And in the short term of the next three to five years, I would actually think that’s how investment is going to play out rather than through investment DAOs and other factors. I think those other things will emerge, but most of the funding, most of the institutional funding will still go to the centralized investors.

Simone Cicero:
And what considerations do you have on the typical — how this transition is going to impact the typical investor incentives? Because we have been talking to DIMO, for example, we have been talking to Braintrust, we have been talking with many Web3 projects. And at the moment, it seems like investors are putting a lot of money into these projects, sometimes investing in both the protocol and the platforms, the products. I guess in the future will be the protocol and maybe the building blocks you build on top of the protocol. So, more smaller units, let’s say. But at the moment, it seems like them, the investment perspective is very confusing. So, it seems like they’re putting money here, because they feel like in the long-term, these kinds of user-owned networks or Web3 networks will be more performative and better than Web2 counterparts. But there is no clear idea of how this is going to return them money, basically.

So, I’m curious because most of these tokens essentially are devalued over time. So, of course, maybe investors buy them at lower prices, but I don’t see, I would, say token appreciation necessarily as a very solid perspective for investors to make money by investing in those companies. So, I’m really puzzled now if I think about existing investors with an investment thesis that is essentially return on capital, investing in these more ecosystems of strategic, for example, digital public goods creation, that may have impacts that are not very easy to monetize. Right? It may be something that is very important, you said at the start, right? You said there is a lot of money that wants to go into building a better word. But I don’t see that really connecting with seeing the same type of investors in the short term and the mid-term, invest in those projects. So, I’m a bit puzzled on this aspect at the moment.

Sangeet Choudary:
Yes. So, a lot of investors are just jumping into Web3, because at least till early this year, there was a lot of capital that just needed a new thesis to find a base, and the speculation was driving a lot of returns around tokens. So, a lot of investors were just jumping into Web3, for all of those reasons. I would just say that was momentum investing than anything else. I don’t think we’ve completely outgrown that era. Even though tokens have fallen, people still haven’t really sat back and thought through where is value really going to be created in Web3. Because it’s one thing to say that value is going to be created at the developer level because of composability, etc. But where is value going to be created and captured for the investors? And I think that is investing in a universe of projects, identifying the key building blocks to invest in, identifying complements to that, and buying governance tokens in others, and so on. So, they’re those specific kinds of approaches that sophisticated investors will eventually develop in terms of where they want to invest in Web3.

I think even more so investing is going to be not just about picking ventures, but also about picking investors. Because if I invest in five projects in a certain building blocks universe, right, and all of them, I believe, are going to be the key projects around which new innovation will happen. The sourcing of new innovation or identifying where to invest next is not just my prerogative, it’s also something I would ask those projects to look at. So, I would allocate a certain amount of fund from the actual venture fund into a project fund because these projects are closer to the innovation that’s happening around them. And they will be able to better identify what’s growing, what’s not growing, what are the compliments that are taking off versus not. And the more you invest across a universe that you are targeting, a thematic universe in this case, the more you would want to ensure that it’s not just all the money that’s being allocated centrally, but you’re creating funds at each end of the project level to continue allocating to complements. And through that, the investment would permeate further.

So, I think that would be one fundamental difference in how investments happen in a Web3 world once this gets sophisticated rather than how it has happened in the Web2 world. But at the same time, to your point, there are limits to token appreciation. That is why early except from a token is in the best interest of investors. So, at the same time, you need to have the right regulation, especially because of how narrative based Web3 investing has been and will continue to be for some time. And because of how the post pandemic cohort of investors invest more than narratives rather than on actual financial fundamentals. Regulation of narratives, regulation of when you can get in and get out of an investment will become even more important. Otherwise, you could easily sway public opinion in a certain direction, drive speculation, exit at a higher value, and then essentially see that whole product collapsing and it would not hurt you as a fund because you are still invested across a larger universe. And there are other thesis that that capital can then be deployed to. So, there needs to be some clear push towards regulating this. Otherwise, we will see a lot more of that happening from the investor side.

Stina Heikkila:
If I can, I wanted to ask you also about — I mean, leaving this investment for a bit. When we have talked to different Web3 projects, I mean, a lot comes down also to the role of data, right, and ownership of data. So, I don’t know what your thoughts are on how that looks like in the building block model. To what extent do these different building blocks — are they, so to speak, sovereign in their data ownership? And let’s say the level of openness will be determined at that level or…

Sangeet Choudary:
This whole idea about data centralization, data decentralization is kind of distracting us from the new centralization. Yes, data centralization was where value was captured for the last 10 to 12 years. I don’t think you can continue doing that in Web3. If you centralize somebody else will decentralize and so you will be out-competed, and so that’s not going to happen. Data centralization is not going to be the way you extract returns. However, investment centralization of the kind that I just mentioned, is still going to be how you extract returns. And so data decentralization, or just calling something altruistic and democratized, etc, just because data has not been centralized, it’s too naive. Because you have to look for where the new centralization is happening. So, there is a lot of talk about data sovereignty and data decentralization, etc, portability. And there is a lot of that talk because it doesn’t matter anymore. You don’t necessarily have to centralize data in order to centralize returns, you will be able to centralize returns at the fund level instead.

Simone Cicero:
In recent conversations with Web3 players, and especially with DIMO, we had this conversation around “data unions”. So, there could be a pattern maybe that can work in these new worlds, so basically users entitling some new entities that can emerge in the ecosystem, maybe even intentionally being invested by the funds that have created the protocol and so on, like in the case of DIMO, right? Because for example, DIMO has been created by Digital Infrastructure Inc, that has kick started the protocol, and is also going to be one of the first data unions. So, what do you think about this perspective? Maybe it’s a good way to think about how data can be used, even if not as a competitive advantage, but at least to make value out of the data in our building block landscape.

Sangeet Choudary:
Yeah, absolutely. I think we will see emergence of data unions at the “universe level” rather than at the project level. So, a lot of the comparison that we make today’s between decentralized projects and centralized platforms, or decentralized Web3 initiatives versus the centralized Web2 platform initiatives that were there. And if we just forget it at that level, it seems like a data union is spanning many different projects, eschewing power back to the users, which is all of it is true. The incentives to make all of this happen would be that some of these data unions would also be backed by investors working across a portfolio that they believe in. And so all I’m trying to say is that the centralization is not going to be at the individual project level. There are many different ways in which the returns can accrue at a much higher level for those who are investing across the universe. So, that is why again, having that universe lens rather than the venture lens becomes really important.

But having said all of that, it will create a lot of beneficial value for the user in different ways. Because users will have much less lock-in, they will be able to move between different projects, they will be able to move between different data unions and so on. And so the traditional challenges of lock in, are they going away? That’s definitely an additional driver of agency back to the user. But are we completely going away from centralization to decentralization? That is where I believe that that is more a narrative, which is being pushed. But it kind of obfuscates what’s changing and what’s not changing. Certain things are becoming decentralized but centralization is happening somewhere else. So, you need to kind of trace that as well, as you think about how Web2 versus Web3 will play out, and Jack Dorsey’s comment about the term centralizing with the VCs and all of that. So, you need to kind of play out those nuances to see how victory will really work and where decentralization will happen versus not.

Simone Cicero:
So, maybe Sangeet as a kind of closing phase of this conversation, I think it may be very worth for us to give some kind of picture overview, a bird’s eye view of how a market enabled by these building blocks thesis may look like. I am talking about things such as, for example, how competition will play out from a perspective of conservation of attractive profits, essentially. How could we look like and I’m also very much interested, and maybe this can be the last conversation. I’m very much interested in what are also the implications from our perspective of let me say, real world economies, not just digital economies, right? When we started, you said, we are seeing a continuous evolution where digital is mediating more value changes in the real world.

So, I know that you care about infrastructure building. I know that you have been researching about trade, for example, and you have been connecting these topics with geopolitical forces and more countries and regions exerting their influence through standards and protocols. So, first, maybe, we can have, first, a quick birds’ eye view to the market appearance from the perspective of competition, cooperation, conservation of profits, and maybe connected these more like geopolitical aspects. And again, let me stress that I would like you to maybe double click on real economies. And when I talk about real economies, I’m talking about economies of fundamentals like food, energy, trade, transportation, telcos; all the things that are so important and we are seeing disrupted so often in the very complex world that we are living in now.

Sangeet Choudary:
If you look at some of the biggest challenges in real-world economies, the biggest challenges are around, effectively deploying these sources to points of need when it is not cost efficient to deploy those resources. And that would mean providing access to things like sanitization, healthcare, microcredit, education, to parts of the economy where all of that is not going to be very profitable to engage in. If you look at a lot of companies that are doing FinTech, Health Tech, Ed Tech, etc, they’re just trying to skim the cream of the crop. They’re just trying to figure out how to create sales engines where you can more easily in a data driven way deliver certain services to the top of the population and charge for the extra convenience, and so on.

So, a lot of traditional venture funding is still backing projects that are doing that. If you really want to solve problems at the grassroots level in these real economy that you’re saying, you need to have an alternate approach to driving that innovation. Also, because as I write in the building blocks thesis, context matters a lot in terms of delivering some of these services and solutions to the end user across the economy, and traditional centralized platforms were not context rich. You would have a one-size-fits all marketplace, and irrespective of which category you were selling in, the marketplace would work in the same way. Or you would have a one-size-fits-all Edtech platform, for instance. But if you want to drive, for example, primary education, highly diverse, and customer environments, you have to think about it more from a building blocks perspective, as somebody who actually participates in the end user environment can create the final solution using building blocks that already exist.

So, my key point being that if we shift away from the financialization and centralization topic we’re talking about, even if we just go back to solution design and innovation, which is what the building blocks thesis has been primarily about diversity of solution designed to solve these myriad end user problems, which are not cost effective to solve today. That is possible only through the creation of this kind of a universal building blocks so that you empower the end stakeholder closest to the user to create a solution that is most closely relevant to the user’s need. The example that I mentioned that was the example of EkStep, which I’ve been an advisor to, and who I co-wrote that whole report with.

But essentially, EkStep tried to look at the problem of primary education, and essentially, they identified that every child has a different context. But every child has a mobile phone. And so while textbooks are one size fits all, if we can understand the child’s context, and if we can map that to the time of the year, the location and the kind of learning journey they have been on, we can provide highly customized content. And so they started putting up QR codes in textbooks so that when a child is reading a static textbook, they can still scan a QR code. And the content that they get access to can keep on changing at the backend based on who the child is, where the child is scanning from, what time of the academic year that scan is happening, etc.

So, that then becomes a digital building block complementing a physical building block in the form of a textbook. And that digital building block then can be customized and delivered in different contexts for different types of users. So, if you look at something of that sort, where you are essentially driving customization of solution or driving solution design, to solve highly contextual problems, the building blocks approach becomes really important. And this is where the traditional view of building centralized solutions may not necessarily work, or building blocks approach becomes much more compelling to create solutions that are highly targeted to the end user’s context. So, that is one key aspect in terms of how this could play out in terms of real world economies.

And, again, as I mentioned, there’s a lot of funding that can be channelized towards encouraging such as distributed innovation to solve such distributed problems. And as long as that funding can be appropriately channeled, that funding is already channeled towards these problems today, but it’s not effectively solving these problems. But with the building blocks approach, where most of the building blocks for, say education delivery have already been created at a global level, at a national level. Doing that at a level of a particular, say, county or district, for instance, would be much easier, customizing at that level would be much easier because of the building blocks that already exist. So, that’s how I see this playing out in the development space, in the change across the real economy space, if you will.

Simone Cicero:
How does this play out from a geopolitical perspective, reconnecting with your earlier work on platforms and standards being used in this matter?

Sangeet Choudary:
Just extending what I just talked about, geopolitically, this is going to be about export of standards and protocols and soft power that can be created on the basis of that. In the Cold War era, the world was split into two blocks, right? And increasingly, we’ll see that happen through standards. We already see a lot of efforts towards it, where different countries are creating digital public goods and exporting that to other countries. And I’ve talked about this extensively in terms of how China has developed its entire trade export strategy, export of building blocks around trade, that it’s exporting to different African countries, different Asian countries; exporting urban city infrastructure, smart city infrastructure, smart city AI, exporting, payments, capabilities, exporting national identity systems. These are all fundamentally the idea of building blocks approach.

You create a standard, you create complimentary building blocks, you subsidize the building blocks and provide it for the developmental needs of other countries. And through that you propagate your standards. And that creates a new kind of geopolitical power compared to what we’ve been used to in the past which have been more based around treaties and agreements, standards now create a much better way of organizing geopolitics. And we’ve seen this happen in business ecosystems, right? Because business ecosystems are managed through bilateral contracts, and now they’re managed through standards and platforms. And similarly, geopolitical negotiations have been managed through these bilateral agreements, for instance. But increasingly, we won’t need those agreements, we will have those implicitly propagated through the adoption of standards and protocols. So, that’s my view on how this plays out geopolitically.

And in terms of defensibility, conservation of attractive profits and so on, it’s important to know that ever since we’ve moved towards a modular world – because if you take out the vertical integration of the 20th century, ever since we’ve gone towards greater modularity – business models have always been about taking the right module, sticking them together, cross subsidizing different modules, and destroying the economics of those who were not able to create that combination of modules. So, typically, somebody who was just providing software for the price would find it difficult to compete with somebody providing software for free and charging for the data, or charging for analytics around that data, and so on.

So, the nature of competition in a modular world just gets accelerated with the building blocks model. What will be different is how much control you have in terms of which modules can be integrated and protected, and link fenced so that you can drive that integration and push it out into the market in a way that is cross subsidized. So, there will be some pressures over there that come up, because especially in what we now call Web2. But essentially, in the digital economy, with the rise of modularity, we’ve seen a lot of scenarios where business model innovation has just been about taking two modules that would not be traditionally combined, and combining them and cross-subsidizing them, charging for something that was independent earlier, and providing for free, something that would be charged if just was provided independently, and hence creating a new business model.

So, essentially, that’s how we made money through modularity over the last several years. What happens with decentralization is some of those, what can be combined and can be link fenced towards creating a new business model might be brought into question. Whether that can be captured as a profit, whether that can be captured at the level of an individual project or at the level of the fund, those are things that will change. But fundamentally, from first principles, the idea of modularity, combining different modules and cross-subsidizing, that will continue to be what drives how we charge and how we identify where value gets captured in a decentralized world as well.

Simone Cicero:
Yeah, probably will make it harder to centralize only in one place, and maybe across the value chain, there will be several plays where this accrues, right? Because maybe you can have different pieces and building blocks connected into kind of modular products and probably defensibilities and capital accrual can happen at several layers of the value chain. But of course, it’s an open question.

Stina Heikkila:
What I notice a lot from the conversations that we are having is that these are early days and there will be a lot of things happening. So, I think maybe unless we have any other points we want to talk about, maybe Sangeet, you can close by telling our listeners, what should they look at right now, and maybe a little bit peek into what they should expect coming up from your side?

Sangeet Choudary:
Yeah, I think a lot of what we talked about was around two or three themes, right. The first is that thinking about innovation in terms of building blocks, provides you a fundamentally new way to solve problems instead of centralizing and creating the whole solution yourself, you can now leverage a distributed ecosystem of solution creators and contribute to that universe your set of building blocks benefit from other building blocks that others are contributing. But then it goes back to the point that centralization will not happen in the traditional way, it will not happen only at one point, it may happen at multiple points in the value chain. More importantly, the best approaches or the best returns to centralization would be at the level of funds rather than at the level of projects in this kind of a model. So, even more so than before, we would see investing play out in a new way than it used to in the past.

So, a lot of my work right now, especially what’s coming up as well, is more on understanding how to think about ecosystems. And think about where value will accrue in modular ecosystems. I mean, this is independent of web to Web3, it’s more about how to think about value accrual in ecosystems. And there are different factors that drive where value gets accrued, depending on whether we’re talking about decentralized versus centralized ecosystems, vectors of centralization, whether we’re talking about singularly owned or co-owned business models within the ecosystem.

So, there are different factors that go into explaining where value will be accumulated and captured within the ecosystems. But fundamentally understanding or thinking of your investing landscape or business landscape as an ecosystem, understanding what are the rules of where valuable flow, where value will be centralized? And then determining what your approach should be. Should you build something over here? Should you set up a fund to invest in complements? Should you create standards and drive the adoption of those standards? These are the core questions that we’ll be confronted with as we move further into this world of ecosystems. And that’s where most of my work is going to be.

Simone Cicero:
Sounds like something more difficult than what companies have been used to, right. So, they need to probably be more strategic, more attentive, more intentional, more sophisticated, right, in thinking about markets coming up. So, Sangeet, it was as always an amazing conversation. And I’m sure that I have to relisten to that, to really properly extract all the insights and the wisdom that you gave us tonight. So, thank you so much for coming again on the podcast. It was really a pleasure.

Sangeet Choudary:
Thank you so much. It’s always a pleasure. And as always, it’s new ideas riffing, and so always a pleasure to come in and be challenged and think about these things more deeply.

Simone Cicero:
Right. To our listeners, we’ll catch up soon.