Web3 peaked somewhere around late 2021. It was in every VC pitch deck, every brand relaunch announcement, every keynote from a company trying to signal it was paying attention. A16z raised a $4.5B dedicated crypto fund. Legacy corporations announced "Web3 initiatives." The term felt inescapable.
Then the cycle turned. The marketing budgets dried up, token prices fell, and Web3 largely vanished from mainstream tech conversation. That arc — stratospheric hype followed by near-silence — led a lot of people to the obvious conclusion: it was always just marketing. A rebranding exercise to make crypto sound inevitable.
That's a defensible read. It's also incomplete.
Where the Term Came From
Gavin Wood coined "Web3" in a 2014 technical paper — not a marketing document. He was co-founding Ethereum at the time, and the paper described a specific architectural vision: a suite of decentralized protocols where users would interact with applications without intermediaries controlling their data or accounts.
The framing was historical. Web1 (roughly the 1990s) was read-only: mostly static pages, centrally hosted. Web2 (early 2000s onward) was read-write: user-generated content at scale, but owned and monetized by platforms — Google, Facebook, Twitter. Web3, as Wood described it, was read-write-own: digital assets that users actually held, authenticated through cryptographic keys rather than platform-issued credentials.
The ownership claim is the load-bearing distinction. It's also the most contested one.
What Web3 Actually Describes
Three components get bundled under the label, and they're worth separating.
Smart contract platforms. Ethereum (and later Solana, others) are programmable blockchains where applications run on decentralized infrastructure rather than on servers owned by a single company. Uniswap, Aave, Compound — these protocols exist as smart contracts, which means they run without a company operating them moment-to-moment. That's architecturally different from a traditional fintech app. Uniswap has processed over $2 trillion in cumulative trading volume. That's not nothing.
Wallets as identity. In Web2, your identity is a server-side database record. A platform can delete your account, ban you, or lock you out. In the Web3 model, identity is a cryptographic key pair you control. Access to an application is proved by demonstrating key ownership — no platform can revoke it because there's nothing to revoke. Whether this matters depends on what you're doing, but the mechanism is distinct.
Tokens as ownership. Fungible tokens (ERC-20) and NFTs represent something cryptographically scarce and transferable without intermediary. Whether that scarcity has economic value is a separate question — but the mechanism of non-custodial digital ownership is real, and it doesn't exist in Web2.
The problem wasn't that these mechanisms were fake. The problem was that the term got applied to things that had almost nothing to do with them.
Where the Marketing Inflation Came In
When VC capital flooded into dedicated Web3 funds, it created institutional incentives to apply the label as broadly as possible. "Web3 gaming" often meant a centralized game with a token bolted on. "Web3 social" often meant a normal social product where posts were stored on-chain but every other part of the infrastructure was centralized. "Web3 enterprise" frequently meant a blockchain-adjacent dashboard built by a consulting firm.
The decentralization spectrum is the thing the marketing obscured. Web3 isn't a binary. It's a continuum:
- Fully decentralized: users hold their own keys, smart contracts are non-upgradeable, no admin controls exist
- Partially decentralized: smart contracts are upgradeable, a team controls the upgrade keys, the frontend is hosted by a company
- Marketing-layer Web3: a token exists, but the underlying system is centralized in substance
Most deployed applications sat in the middle or bottom of that spectrum — particularly consumer-facing products. The honest version is that full decentralization is harder and slower than marketing implied, and building genuinely decentralized applications involves tradeoffs that most consumer products weren't willing to make.
The Filter That Actually Ran
The 2022-2023 market downturn worked as an involuntary filter. Marketing budgets disappeared. Projects that couldn't sustain usage without promotional incentives mostly collapsed or went quiet.
What's still running: the DeFi protocols with genuine activity. Uniswap's volumes aren't driven by VC narrative. Aave's lending markets have been live through multiple market cycles. Stablecoin volumes on Ethereum and its L2 rollups have grown — driven less by speculation than by actual cross-border payment and settlement use cases that didn't exist in earlier cycles.
The term "Web3" has actually been quietly retired in much of the technical conversation. Developers working in the space increasingly use more specific language: "onchain," "DeFi," "crypto-native." That's probably a good sign. Vague marketing umbrellas tend to recede when the underlying technology has to stand on its own.
What Would Confirm the Marketing-Only Reading
Web3 was primarily a marketing exercise if none of the infrastructure survives the promotional cycle with durable organic usage. The clearest confirmation signal: if DeFi protocol volumes return to near-zero when speculative incentives are removed, and if self-custody wallet adoption fails to grow outside speculative cohorts, then the architecture didn't produce real behavioral change. Just a casino with better branding.
What Would Confirm the Structural Reading
Conversely, Web3 represents a real architectural shift if: DeFi protocols demonstrate sustainable fee revenue at meaningful scale; self-custody wallet adoption grows for non-speculative use cases (stablecoin payments, real-world asset access, identity); and applications built on open, composable protocols demonstrate advantages — network effects, permissionless access, composability — that closed platforms can't easily replicate. Real-world asset tokenization going beyond pilot stage would be a strong signal. So would stablecoin volumes growing without correlation to crypto speculation cycles.
Timing Perspective
Now: The DeFi substrate is functional and has activity. Ignore the label, evaluate the specific protocol. Uniswap's usage is a separate empirical question from whether "Web3" as a narrative had merit.
Next: Stablecoin payments and tokenized financial assets are the 12-24 month signal worth watching. If Web3 infrastructure ends up as payment rails for the dollar-denominated global economy, that's a different story than the 2021 narrative implied — less culturally dramatic, more structurally important.
Later: Whether wallet-based identity (cryptographic keys as login primitives) actually displaces Web2 authentication patterns at consumer scale is a multi-year open question. Account abstraction (ERC-4337) has made the UX more realistic, but adoption is still early.
Boundary
This explanation covers the technical architecture that "Web3" was coined to describe, the marketing inflation that obscured it, and the filter mechanisms that separate durable infrastructure from promotional packaging. It doesn't constitute a recommendation to use, invest in, or build on any Web3 protocol or platform.
The short answer to the question: partly. The marketing layer was real and substantially inflated. So was some of the underlying architecture. They're not the same thing, and conflating them is its own kind of error.
