Hook
Pavel Durov just announced Telegram is applying for the .gram top-level domain. The pitch: every username becomes a website, a Mini App portal, a piece of digital real estate. 10 billion users, one domain. The crypto community is already salivating. But here is the data point Durov conveniently omitted: ICANN's next new gTLD application window is not scheduled until April 2026. Telegram has not filed a single document. The entire narrative is a pre-emptive hype bomb, timed to catch the bull market euphoria.
Context
Domain names are the oldest form of digital identity. ENS on Ethereum, .crypto on Unstoppable, even .ton.site on TON—they all promise the same thing: a human-readable address that you control. Telegram's twist is scale. 10 billion monthly active users, each with a unique username. If every @durov becomes durov.gram, the distribution is virtually instant. The product would be a combination of DNS registration and Mini App hosting: type a command, generate a website. No hosting, no configuration.
But the infrastructure to support this is not trivial. Telegram currently resolves usernames via t.me/username. That's a simple HTTP redirect. A TLD requires a dedicated authoritative DNS server cluster, DNSSEC signing, WHOIS/RDAP compliance, and abuse handling. Telegram has never operated a DNS registry. The company's core competency is encrypted messaging, not root zone coordination.
Core: The Technical Skeleton Under the Hype
Let me break down what .gram actually requires. First, ICANN approval. The process costs approximately $185,000 in evaluation fees alone, plus annual registry fees. The application window is not guaranteed—ICANN has delayed new gTLD rounds before. If approved, Telegram must sign a Registry Agreement, commit to a set of public interest commitments, and implement a Sunrise period for trademark holders. This is a regulatory gauntlet, not a weekend hackathon.
Second, the username-to-domain mapping. Telegram holds a database of 10 billion usernames. Under ICANN rules, a registry cannot reserve domain names arbitrarily—they must be "reserved names" approved by ICANN. But reserving every username would require a massive list submission. More critically, if someone else registers the same name as a domain (e.g., durov.gram by a squatter), the conflict resolution process is external to Telegram. The company would have to rely on ICANN's Uniform Domain-Name Dispute-Resolution Policy (UDRP), which is slow, expensive, and favors deep pockets.
Third, the Mini App integration. The article claims users can create interactive websites hosted on Telegram. But the existing Mini App platform is a WebView container—it renders HTML/JS, but the actual hosting is on Telegram's servers. A .gram domain would point to a specific Mini App, but the relationship between the domain and the underlying app is not a true DNS record; it's a routing table. This is not a distributed web; it's a walled garden with a nicer URL.
Patterns hide in the noise floor. I've seen this architecture before—in 2021, when NFT projects promised "on-chain metadata" but stored everything on IPFS with a centralized gateway. The illusion of decentralization is the product. Telegram's .gram is no different. It's a centralized domain registry with a decentralized marketing narrative.
Contrarian: The Unreported Angle—This Is a Liquidity Fragmentation Play
The mainstream take is that .gram will unlock a new wave of personal websites and creator economies. The contrarian truth: this is a textbook case of slicing already scarce liquidity. The domain market is saturated. There are over 1,500 new gTLDs registered, with an average renewal rate below 30%. Users are tired of new suffixes. Google's .app and .dev have traction because they are backed by a search engine monopoly. Telegram has no such advantage.
Chasing the ghost in the liquidity pool. The real game is not domains—it's locking users into Telegram's ecosystem. By giving each user a .gram domain, Telegram creates a sticky identity that is hard to migrate. But the domain itself has no intrinsic value. Unlike ENS, which is tied to a wallet address and can be used across dApps, a .gram domain is only resolvable within Telegram's infrastructure. It's a branded link, not a portable asset. The unit economics are worse: high gross margin (low cost), but low LTV if users don't renew. The only way to sustain value is to turn the domain into a scarce resource—limited registration, premium fees, or bundling with Telegram Premium.
Moreover, the competition is not just from other TLDs. It's from existing Web3 identity protocols. ENS has over 2 million registered .eth names, and it's integrated with hundreds of wallets and dApps. Telegram's own TON ecosystem already has ton.site, which maps wallet addresses to decentralized websites. Why would a developer choose .gram over .eth or .ton? The answer: because Telegram forces them. That's not a moat; that's a cage.
Volatility is the price of admission. The bull market is masking the fundamental problem: users don't want domains. They want utility. A .gram domain that only points to a Telegram Mini App is a glorified bot link. Without the ability to point to external websites, without DNS-level flexibility, the domain is a decorative asset. The same mistake was made by .crypto and .nft—they launched with limited use cases and died when the hype faded.
Takeaway
Telegram's .gram application is a strategic signal, not a product. It tells us that the company is shifting from communication infrastructure to identity infrastructure. But the path is littered with regulatory landmines, technical debt, and user indifference. The first signal to watch: does ICANN even open a new gTLD window in 2026? If yes, does Telegram actually submit? If no, the entire narrative is vapor.
Dissecting the anatomy of a pump. The real alpha is not in buying the domain—it's in shorting the hype. Watch for Telegram's next move: if they announce a token airdrop tied to .gram registration, run. That's the same playbook as liquidity mining, but with domains instead of yields. Yields are just lies with better formatting. So are domains when the underlying infrastructure is a promise.
Speed is the only alpha left. I'll be tracking the ICANN application queue, not the Telegram channel. The truth is in the registry, not the rhetoric.