Solana co-founder Anatoly Yakovenko said this week that when it comes to optimizing the network, a handful of high-volume consumer apps deserve priority over validators, developers, market makers or even core engineers.
What he said
- In an Oct. 23 post on X, Yakovenko named five products he believes Solana should optimize for first: Phantom, Pump.fun, Fomo, Jupiter and Kamino.
- His reasoning: “Products with a ton of AUM and a ton of retail volume are the two most important groups to optimize for.” He emphasized that those metrics — assets under management and retail trading activity — should drive optimization decisions.
- Yakovenko explicitly ranked those apps above validators, developers, market makers and core engineers in terms of who the network should be optimized for, though he did not propose any specific technical changes, timelines, or governance shifts on their behalf.
- In a follow-up comment he summed up the priority user more generally: “The most important user is the plumber, institutional or retail plumber.”
Context and reactions
- Yakovenko’s remarks came in response to criticism that some Ethereum protocol designers do not use their own network. He offered qualified praise for the Ethereum Foundation, saying it had “done a decent job at optimizing for Aave,” noting how inclusion failures during outages or congestion would break lending platforms.
- This is part of a broader debate Yakovenko has engaged in about application economics on Layer 1s. He previously argued a company could build on Solana, subsidize transaction costs, and charge users through its own interface — a position that prompted pushback in September from Offchain Labs co-founder Steven Goldfeder over Robinhood’s use of Arbitrum.
- Goldfeder defended Robinhood’s ability to retain most chain revenue under the Arbitrum Expansion Program, saying apps could keep roughly 90% of net protocol revenue. Reported terms route 10% of net protocol revenue to the Arbitrum ecosystem (8% to the DAO, 2% to the Developer Guild), with “net” measured after relevant network expenses such as Ethereum data-posting costs.
- Yakovenko’s Oct. 23 post did not announce any changes to product access for U.S. customers or any agreements with the teams he named.
Where this fits with Solana’s upgrade work
Yakovenko’s comments arrive as Solana continues rolling out major protocol upgrades and network changes that affect application teams, validators and infrastructure operators:
- Alpenglow upgrade and testing stages:
- Alpenglow moved to devnet on Sep. 26 after a testnet transition, giving application teams early access to the new consensus behavior before mainnet deployment.
- The Foundation has said apps that only submit transactions and read balances should not need to migrate, while services that process blocks or track validator activity may need to change how they read network data.
- Devnet is intended for application teams to test with valueless tokens; testnet is aimed at validators and operators running heavier, realistic conditions. Anza announced the devnet transition on Sep. 25, one day after testnet completed its switch.
- Finality and performance targets:
- The upgrade targets a finality time of roughly 150 milliseconds, a dramatic improvement over the ~12.8 seconds under TowerBFT. Coverage has stressed the distinction between consensus finality and an initial indication that a transaction landed.
- Validator economics and admission fees:
- The Foundation plans to replace separate on-chain voting fees with a flat 0.8 SOL admission charge per epoch that will be burned in full. Validators must fund that charge above their vote account’s rent-exempt minimum, and payment is required one epoch in advance. Validators unable to pay lose permission to vote or produce blocks in the following epoch.
- Nodes outside the admitted validator set normally learn finality from certificates included in later blocks, which can add a produced-block interval and lead to longer waits if blocks are skipped or delayed.
Bottom line
Yakovenko is steering a pragmatic, product-first argument: optimize the network for high-AUM, high-retail-volume apps that move real assets and transactions. He gave concrete examples but stopped short of proposing technical changes or governance moves. Meanwhile, Solana’s upgrade path (Alpenglow, finality improvements and validator fee changes) continues to reshape how apps and operators will interact with the network.
What he said
- In an Oct. 23 post on X, Yakovenko named five products he believes Solana should optimize for first: Phantom, Pump.fun, Fomo, Jupiter and Kamino.
- His reasoning: “Products with a ton of AUM and a ton of retail volume are the two most important groups to optimize for.” He emphasized that those metrics — assets under management and retail trading activity — should drive optimization decisions.
- Yakovenko explicitly ranked those apps above validators, developers, market makers and core engineers in terms of who the network should be optimized for, though he did not propose any specific technical changes, timelines, or governance shifts on their behalf.
- In a follow-up comment he summed up the priority user more generally: “The most important user is the plumber, institutional or retail plumber.”
Context and reactions
- Yakovenko’s remarks came in response to criticism that some Ethereum protocol designers do not use their own network. He offered qualified praise for the Ethereum Foundation, saying it had “done a decent job at optimizing for Aave,” noting how inclusion failures during outages or congestion would break lending platforms.
- This is part of a broader debate Yakovenko has engaged in about application economics on Layer 1s. He previously argued a company could build on Solana, subsidize transaction costs, and charge users through its own interface — a position that prompted pushback in September from Offchain Labs co-founder Steven Goldfeder over Robinhood’s use of Arbitrum.
- Goldfeder defended Robinhood’s ability to retain most chain revenue under the Arbitrum Expansion Program, saying apps could keep roughly 90% of net protocol revenue. Reported terms route 10% of net protocol revenue to the Arbitrum ecosystem (8% to the DAO, 2% to the Developer Guild), with “net” measured after relevant network expenses such as Ethereum data-posting costs.
- Yakovenko’s Oct. 23 post did not announce any changes to product access for U.S. customers or any agreements with the teams he named.
Where this fits with Solana’s upgrade work
Yakovenko’s comments arrive as Solana continues rolling out major protocol upgrades and network changes that affect application teams, validators and infrastructure operators:
- Alpenglow upgrade and testing stages:
- Alpenglow moved to devnet on Sep. 26 after a testnet transition, giving application teams early access to the new consensus behavior before mainnet deployment.
- The Foundation has said apps that only submit transactions and read balances should not need to migrate, while services that process blocks or track validator activity may need to change how they read network data.
- Devnet is intended for application teams to test with valueless tokens; testnet is aimed at validators and operators running heavier, realistic conditions. Anza announced the devnet transition on Sep. 25, one day after testnet completed its switch.
- Finality and performance targets:
- The upgrade targets a finality time of roughly 150 milliseconds, a dramatic improvement over the ~12.8 seconds under TowerBFT. Coverage has stressed the distinction between consensus finality and an initial indication that a transaction landed.
- Validator economics and admission fees:
- The Foundation plans to replace separate on-chain voting fees with a flat 0.8 SOL admission charge per epoch that will be burned in full. Validators must fund that charge above their vote account’s rent-exempt minimum, and payment is required one epoch in advance. Validators unable to pay lose permission to vote or produce blocks in the following epoch.
- Nodes outside the admitted validator set normally learn finality from certificates included in later blocks, which can add a produced-block interval and lead to longer waits if blocks are skipped or delayed.
Bottom line
Yakovenko is steering a pragmatic, product-first argument: optimize the network for high-AUM, high-retail-volume apps that move real assets and transactions. He gave concrete examples but stopped short of proposing technical changes or governance moves. Meanwhile, Solana’s upgrade path (Alpenglow, finality improvements and validator fee changes) continues to reshape how apps and operators will interact with the network.