Prior to 2025, exchanges did one thing. They were centralized and therefore their capabilities were limited. They could only launch products that fit on top of this model and that meant maybe they could do a futures product, maybe they could do some yield products. Now the landscape has shifted, there is enough talent consolidated in major exchanges, the exchanges have defi interfaces and they can ship fast. We saw what Coinbase did with the Base wallet, the decentralized interface. They locked everyone in a room and got it shipped in weeks. They even launched Base, their own chain with a vibrant ecosystem. Fresh entrepreneurs now want to build on Base for the Coinbase brand association.
When exchanges can ship like this and they have the tier 1 brand cachet, their total addressable market expands and they attract innovators. Now you are in the wolf’s den — can you expand to be bigger than the exchange, or will you be forever a subset of their market?
The competitive landscape for crypto startups has fundamentally changed. The question is no longer what exchanges won’t do — it’s what they can’t do. That list is getting shorter by the quarter.
What Changed in the Last Eighteen Months
Coinbase launched Base in August 2023. A publicly traded, SEC-scrutinized company now operates one of the largest L2 blockchains by transaction volume. Kraken announced Ink, their own L2, launching in early 2025. Binance has been running BNB Chain since 2020. The “exchanges don’t touch L1s” thesis is dead.
The Coinbase Wallet now interfaces directly with DeFi. Their Bitcoin-backed loans product hit over $1 billion in originations by October 2025, running through Morpho on Base. They wrapped Bitcoin as cbBTC. They launched a card that lets you spend crypto at merchants. The “exchanges can’t do DeFi” thesis is also dead.
The regulatory moat — the idea that compliance complexity would keep large players out of innovative product lines — evaporated when the largest regulated exchange in America started running a blockchain and offering permissionless lending.
The Startup Graveyard Is Filling Up
The pattern is clear. If what you’re building can be replicated by a team at Coinbase, Kraken, or Binance in a quarter, you don’t have a company. You have a roadmap feature.
Privy built embedded wallets — elegant technology, 75 million accounts. Stripe acquired them in June 2025. They became a feature.
Farcaster raised $150 million from Paradigm and a16z at nearly a $1 billion valuation. Monthly active users dropped from 80,000 to under 20,000. Dan Romero put it bluntly: “We tried for 4.5 years to put social first, but it didn’t work.” SocialFi was a feature.
Lens Protocol raised $46 million total. Daily active users dropped from 42,000 at peak to around 8,000. New signups cratered 99%. The decentralized social graph was a feature waiting for a platform that never came.
What Moats Actually Exist
Network effects that compound. The key word is “compound” — if your network doesn’t get meaningfully stronger with each new participant, an exchange can bootstrap a competing one with their existing user base.
Product lines they haven’t imagined. The window between “novel concept” and “Coinbase is building this” is measured in months. You need to find the genuinely orthogonal ideas.
Technological innovation that’s hard to replicate. In an open-source ecosystem, the advantage has to come from something deeper — proprietary data, hardware integrations, or cryptographic research that takes years to replicate. When you ship a novel mechanism, you’re effectively publishing a research paper that Coinbase’s team will read, improve upon, and ship with better distribution.
Terminal velocity before they notice. If you can reach escape velocity before an exchange decides to compete, you might survive through sheer momentum. Hyperliquid pulled this off with perpetuals. Most don’t.
The Two-Part Test
For founders, the calculus has changed. The question isn’t “Is this a good idea?” It’s a two-part test:
- Can Coinbase, Kraken, or Binance build this with their existing resources, user base, and regulatory posture?
- If yes, can you reach terminal velocity before they do?
If the answer to the first question is yes and the second is “probably not” — you’re building a feature. Features get acquired at a discount to their potential, or they get copied and die.
The path forward is narrower than it was. Focus on networks the exchanges can’t bootstrap. Understand your competitive edge. Find the business opportunities that exchanges are too big to create.