Superstructure
As web3 matures, so does its infrastructure and tooling, and one of the most interesting developments lately has been the emergent stack forming around an embedded wallet of one kind or another. Dozens of companies and projects are vying to arm the dapp developers of the world with a seamless onboarding experience with (mostly) non-custodial guarantees. And while I’ve been vocal about the fact that improving the UX of crypto alone won’t solve mass adoption when value risk looms so large, tools which let developers provide a seamless experience on the wallet interaction side of things are absolutely critical for dapp developers to be able to de-risk with the least amount of confounding variables (i.e. the scraggly edges of unpolished web3 UX). The opportunity for what I’ll call the wallet-centric customer experience stack is enormous, and stands as an in-progress category creation moment with an uncertain resolution.
One of the biggest prizes in b2b software across different tech paradigms has been what I’ll call the "experience stack." It means different things in different contexts, and there are overlapping and coexisting experience stacks in many cases, but broadly we can identify platforms like Salesforce and later Intercom, Twilio, and Shopify, companies who have, from diverse product strategy starting points and target markets, converged around the same privileged position of being a company's "vendor of record" with which they primarily manage their end-to-end customer relationship. The “experience stack” is not the sole component that companies need to integrate, but it is the sun around which everything else in the value chain orbits and enjoys outsized strategic optionality, pricing power and value extraction. This usually coincides with a kind of "customer record" innovation that reflects the centrality of that vendor's complete solution for managing the customer relationship, such as the cloud CRM, the e-commerce order history or, more recently, the "customer conversation" history as chat-based (and very soon, the long promised AI chatbot-based) customer engagement begins to become more and more central.
As technology is always changing and evolving, and new S curves spring up and advance as old ones peter out, we have a few overlapping "experience stack" paradigms coming out of web2. Salesforce's major victory in SaaS CRM laid the groundwork (and also the inertia) for its fight with Intercom, Twilio et al for the "conversational" CRM of the future, while the e-commerce stack that Shopify used to build a complete merchant platform for small and medium sized businesses has begun to see competition from below from the cloud providers ever increasingly large "lego block" services. While AI will pull forward and emphasis of conversation on this "experience stack," in the web3 corner of the world, we are watching a very interesting contender for this category - and unsurprisingly, it's completely oriented around wallets and the on-chain transactions they help users execute. This wallet-centric experience stack is being approached from multiple angles, but we can see it converging around the same chokepoints in terms of user flows to enable:
Allow users to onboard easily, regardless of whether they are an existing web3 user or a new user, with some kind of seedless or equivalent experience
Manage wallet connection, transactions and switching
Allow users to make gasless transactions, pay with any gas token, authorize actions with session keys
Manage users, sessions and all the edge cases around accounts connected to wallets
Make it easy for users to on and offramp into crypto with their payment method of choice
Make it easy for users to make more complex transactions and accomplish higher-level wallet-aware workflows, like depositing into a defi yield-bearing vault or gaining access to token-gated content
While there is much more to it than the wallets themselves (i.e. linking web2 credentials to web3 wallets for simple sign in, session management pre/post transaction), the wallet is the primary and most important customer relationship paradigm, one which flips the web2 paradigm on its head. In web3's wallet-centric world, the wallet is a self-contained payment and identity solution that the user has some kind of custody over and which natively lets them express their preferences as blockchain transactions. Fundamentally, the dapp or protocol is using the wallet (in whatever form) to communicate with the customer and engage in a rich, ongoing "conversation" - the history of which is (at least for now) public and onchain. Fundamentally what something like WalletConnect's 2.0 protocol is doing is allowing dapps and wallets to communicate with one another through cryptographic means, initially ferrying value and data related to transactions, but soon even messages themselves. Establishing a wallet-mediated relationship with the consumer is the name of the game.
The crypto downturn has emphasized the need for clarity on exactly what “consumers” we are talking about, and so I’ll try to lay out in a handy 2x2 what the potential buyer/integrator demographics look like for this emerging experience stack category.
First, we have Web3 native apps trying to reach web3 natives, i.e. Defilama and Bungie. These are crypto-native builders who are trying to reach the niche (but valuable on an ARPU basis) segment of existing web3 users, whether that be hardcore NFT or hardcore DeFi or hardcore technical users. This segment is definitely likely to grow, and I'm very bullish on the on-chain web3 native economy’s absolute growth prospects in the medium and long run, but this is definitely the most challenging segment for experience stack providers to be focused on at the moment (and is generously maybe 5 or 10 million MAUs in size in terms of active smart contract platform users). Some of these companies can roll their own stack for some of these needs and are more comfortable using lower-level technologies themselves like wagmi and ethers. And since many of these interactions for web3 natives are very transactional (thin-ish UIs on top of mint, swap, deposit, etc), the free + OSS Wallet Connection libraries for free from the likes of Rainbow and WalletConnect are likely sufficient for a lot of their use cases. The preferences of this segment are the lumpiest and hardest to nail down from what I can tell, and so providers competing here are going to face an uphill battle.
We also have web3 native builders who are taking swings directly at the mass market from an upstart position What comes to mind here are the likes of Stepin, Axie, Polymarket, OpenSea. These are app developers who have come from the crypto world, have crypto bona fides, but are firmly trying to swing for the fences and target the mass market with their products. They are highly experience-sensitive because their users are much more likely to be new to crypto, and have more of a need for an end-to-end customer experience and onboarding management platform. Still, these are startups, they're likely going to want to have a very product-led sales experience where they are getting to use the product for free up to some point, their developers are able to kick the tires and weigh in, and only talk to sales when it's directly necessary. Other examples of this kind of company are what I’ll call “noncustodial neobanks” like Cenoa in Turkey, which are using DeFi rails to bring a differentiated financial service (in this case, inflation resistance and dollar savings yield) to a mass market audience. Some of the first category are confused and think they are in this second bucket (“our TAM is everybody!”), and some may well make it to the mass-market as they struggle through the gauntlet of finding PMF.
On the BigCo and Web2 side, in terms of builders, we have a group of existing players trying to reach the high net worth, existing web3 niche either as an experiment or as an extension of their loyalty footprint in the overlap between their most loyal and valuable customers and the existing web3 userbase. This brings to mind examples like the Tiffany's NFT drop and even the Twitter profile picture integration, which was really meant to pull in existing web3 users who already had NFTs and simply give them a great experience in their niche on Crypto Twitter.
And finally, we have what might be the Big Kahuna, which are the existing Web2 brands and large companies trying to reach the mass market. They might be targeting their existing audience, sometimes a new audience, but fundamentally trying to reach normal people with web3 tech for one reason or another. The user experience premium here is likely to be very high, particularly as these forays into web3 from existing migrate from innovation departments to more line-of-business deployments (and therefore more rigorous ROI/P&L analysis). I’d also lump influencers and celebrities who are getting deeper into web3 into this category, being as they are “enterprises” in and of themselves.
Some of the big players like OpenSea have chosen to stitch together their own solution (afaik) and integrate lower level libraries like themselves, but a growing sector of startups and projects are emerging to solve these problems in a repeatable way for dapp developers across all segments. They are diverse in their approaches, starting points and strategies, but are convergent around the central needs of this end to end “experience stack” mentioned above. Note that many of these categories are overlapping, and this inexhaustive list of players in the industry is an attempt to categorize in order to understand, not in order to reify. Broadly, contenders are coming from the following starting points and/or initial emphases:
Key management focused (like Lit Protocol), leaning heavily into security, programmability of recovery options, trustlessness)
General web3 platforms (Like Thirdweb), who provide a wide range of dapp building blocks all under one roof
Smart account platforms (Like ZeroDev), which focus the offering around the smart account, and put things like gasless or sponsored transactions in the center. Some vendors are focused on the account piece itself, while others are focused on the paymaster/bundler piece separately
Authentication focused providers (Like Dynamic and web3auth), who have a robust feature set around managing users in relation to their wallets, including Oauth integration, session management etc.
Minting platforms (Like Hypermint), who are focused on a specific common web3 use case and providing a complete turnkey solution for brands
Wallet connector libraries (Like RainbowKit), which focus almost exclusively on the minimal connect/sign-in-with-wallet functionality required to interact onchain
These solutions are increasingly competitive with one another directly, and indirectly are all competing for the very scarce integrator attention and relationship that the “experience stack” usually confers to the vendor/solution which occupies that keystone position. They typically come in the form of SDKs with backend services attached, and vary in terms of what standards they are built upon. This type of convergence around the same problem chokepoint from a variety of solution angles highlights the category creation moment that we are in.
The fog of war here is thick, there are several moving parts, and different players taking different views of those moving parts - views as to what is core versus peripheral to value, what should be a proprietary differentiator versus an open standard. That means several instances of “friendpetition” where one or more of these players, all vying for either the same limited attention and budget from the buyer around their “web3” needs, or as a value-added infrastructure component that can extract a tax without being designed out of the equation due to some accrued long term advantage.What will the winning focus(es) turn out to be? What are the winning combinations of components in this space to in-house and differentiate on, versus outsource or integrate with? There are a few different theories of the case, which I'll illustrate here in a quick Wardley map of this emerging landscape at a very high level. For those that aren't familiar with Wardley mapping, it's a way to represent a value chain of components and dependencies in a dynamic competitive system evolving over time. I’ve mapped out here a high level view of the value chain underpinning the wallet-centric experience stack:
Players like ZeroDev are focusing on the smart account services side and integrating with arbitrary key management providers, whereas Dynamic is taking a more holistic auth-centric approach which spans embedded wallets to session management. Safe, of course the creator and maintainer of the Safe smart account core protocol, has widened its offering with basic kits for auth, making transactions and connecting to relayers. Some components in the value chain aggregate one another, and it's unclear whether the aggregation of things like MPC wallet providers or SCW implementations by the experience stack provider will be what wins over customers, or whether an integrated approach that chooses one implementation for each of these major pieces does better. For instance, Zero Dev, starting from the smart account SDK corner of the problem and working backwards, has has multiple custodial, semi-custodial and non-custodial embedded wallet offerings that it lets you use with its smart accounts. Thirdweb and Dynamic similarly offer both first party and third party options for these "instant wallets," and one can probably anticipate players experimenting with integrating one or more smart account frameworks as the “account abstraction” fueled smart contract wallet bonanza continues.
Such a tangled web - albeit one clearly forming into this end-to-end experience stack - is not uncommon in the early stages of category creation. While a bunch of experiments in this space are happening, we can see a few different potential winning hands but not their exact details, configuration, and timing. We’ll probably only get a real category name from Gartner in 18 months when the war is effectively over (and the spoils of the peace can begin in earnest).
This moment of category creation is a liquid one, one where various permutations of similar concepts, components and messages are tried. As the customer development process continues and early PMF begins to solidify, there are still many outstanding dimensions of the idea maze that providers in this space (and their customers) will need to wander through and eventually resolve. And that means that there is still a ton of “alpha” in the space.
For instance, there are some obvious directions that we can expect this experience stack market to lean into as web3 almost speedruns web2 history, including the development of a rich and competitive “integrations” ecosystem, a deeper emphasis on dapp<>user communication and embedded communication tools (perhaps web3 ones this time like XMTP), and features for more explicitly managing the stages of the customer journey, like retention management and marketing automation tools (again, likely web3 native ones). We can likely expect sales pitches in this space to converge more around traditional metrics like LTV or ARPU, with a blockchain transactional twist. The “Salesforce” of web3 will have a lot of web2 Salesforce in it.
But there are a ton of very thorny questions pretty specific to web3 that still need to be answered before this category shakes itself out.
The spectrum of “non-custodiality” is vast, and there are a lot of strong opinions on the right design, considerations and assumptions should hold across MPC and related approaches for handling traditional EOAs from a security and usability perspective. Without delving too deep into the specifics of the debate, there's a fundamental difference between what Lit protocol is trying to accomplish and its properties, and something much more institutional and web 2.5 like Fireblocks, and yet both (and everything in between) are being evaluated by developers looking to add the ability for users to create and maintain easy to use seedless accounts. The folks at Silence Laboratories have given a good view on the different ways of thinking about this in this blog post "**Redefining the Nomenclature of Web3 Custody." We even see lower-level experimentation from the likes of the Odsy network which takes a completely different approach in the Cosmos ecosystem by making the wallet itself decentralized and programmable with so-called "dWallets."
Relatedly, the coexistence of these embedded wallets and existing standalone wallets, as well as the portability and ejectability of keys, is another major open question. There is a so-called “graduation” problem if and when a user with an embedded wallet wants to move on into a more standalone wallet experience, discussed at length in Privy CEO Henri’s Stern’s great post on “progressive onboarding.” As their newly minted web3 users “grow up” in their web3 journey, there may be a pull to a more independent standalone wallet, and dapps will need to take a view on how they want to handle that. Some experience stack vendors like Sequence are leaning into this - making their particular wallet app offering both the solution for early onboarding and later engagement, retention and graduation into a full web3 user of multiple dapps with the Sequence wallet. Others like Coinbase WaaS are looking at a lower level of technology and the flexible approach dapps to white labeling their starter wallets and allow users to eject keys to a standalone wallet. There is a natural tension between the “app wallet” thesis that the experience stack is so focused on and the “fat wallet” thesis which sees today’s wallets as embryonic superapps which will intermediate web3 on behalf of the user. But eventually, even in a world post-traction of these embedded wallets, there will need to be some natural re-aggregation for users who either have multiple embedded wallets across the different apps they are using, potentially a ripe opportunity for the major web2 platforms (like the phone operating system or the default messaging superapp) to step in and yank the end user relationship away from experience stack providers and standalone wallets alike.
We also, of course, have the question of MPC vs smart account emphasis. Are these technologies substitutes or complements? Some experience stack providers are focused on one or the other, whereas some are seeing value in offering hybrid solutions (like we see with Lit + Stackup or Safe and web3auth). One observation is that key management, while there are still a host of problems to solve, is likely commoditizing in terms of the benefits that integrators in this space can absorb, so I've represented it in the map above as being on the cusp of commodity/utility status. It seems likely that pure key management providers/networks will compete to power many of these higher-level all-in-one SDKs and will not usually have a direct integrating customer relationship, unless some aspect of key management ends up becoming enough of a differentiator to move the needle. And the way that key management plays out will have a lot to do with the speed of adoption of smart accounts which are paired with key management solutions versus something like a pure MPC-based solution. Smart accounts also have an emerging concept of "modules" which extend the functionality and programmability of the wallet logic to external developers who can specialize in things like different recovery schemes. It’s unclear whether this becomes the important point of extensibility for the experience stack, or whether another higher-order abstraction like a full-stack “workflow” captures developer mindshare. Both MPC and Smart Accounts seem likely to play a big role in how the experience stack develops.
The question of depth versus breadth of chains, related to the smart account / account abstraction question, is another that will be important to track. There is one school of thought which says that vendors in this space should be on as many chains as possible, catering for the long tail of any requests clients may have in terms of chain - which speaks to one of MPC's best advantages, its fundamental blockchain agnosticity. There are forces which will pull experience stack providers to more or less custodial implementations, as well as between more or less decentralized chains, but part of the fundamental value proposition of using web3 technologies like NFTs in your product to begin with is the “return on composability” that they offer, giving the ability for your product to interoperate with native web3 standards, liquidity and existing primitives, permissionlessly. It’s unclear how this shakes out exactly, as different integrators may have value interoperability and composability more or less than other factors.
Whether experience stack providers should be generalized or focused on a particular sector is also an open question. For example Sequence is focusing on game developers holistically, so in addition to offering wallet and auth infrastructure, are offering data indexing and NFT merchandising solutions, as well as fiat on-ramps and Unity/Unreal SDKs. We will likely see other vendors leaning into the "noncustodial neobank" segment and selling to legacy and neobank fintech customers as they look to bring onchain benefits to users without the onchain hassle.Other players like Dynamic are making a much more generalized pitch. There will also be players who focus on bottoms-up distribution and web3 native developer community traction as a bet on the future innovation coming directly from the heart of web3 (i.e. the web3 native maximalist bet). Likely there will be a few big winners in a few different focus areas, and a race for underlying general infrastructure and standards which run right through the smart contract wallet and key management ecosystem questions.
There is also a fundamental fuzziness in the “account” model that is emerging here. It’s unclear what exact identity and login model will prevail in which segment. Are we talking about the Waystar Royco dapp account, or experience stack vendor AcmeCo’s embedded wallet account? Some solutions are more focused on one or the other approaches, and many have differing levels of white-labelability inherent to their design and the exactly part of the stack that they occupy. The different segments of potential experience stack customers might have different views on “owning the customer” through the wallet relationship or not, depending how how deep down the web3 rabbit hole their offering actually goes.
And finally - what is the endgame organizational structure or business model? Are we likely to see a few gigantic, centralized players left standing at the end? Probably. But there may also be a lot of open solutions that, over time, erode the margin of more centralized service providers, and constantly force them to innovate higher and higher up the stack to remain a value-add. On the other hand, there’s possibility is that the winners become so big that the only way to relieve the pressure that the contradiction of their existence (a centralized and hegemonic vendor proxying access for people to the "trustless" web3) is to decentralize or open pieces of their solution in some way.
Web3’s fundamental and most important risk remains value risk - the risk that we are not able to apply those properties which only blockchains afford us to bear upon problems outside of a narrow band of speculative activities of early adopters. In that sense, improving the UX of web3 alone will not save it from its current use case winter. But that doesn’t mean that improved UX for dapp users - web2 experiences with web3 guarantees - isn’t a necessary condition for that value risk to be reduced over time. The wallet-centric customer experience stack is the most important and biggest “picks and shovels” b2b infrastructure opportunity in crypto right now, and as far as I’m concerned, it’s wide open.
*Thanks for reading. You can follow me me on Twitter @basche42 or Farcaster *
