Dominic Williams challenges how blockchain transaction numbers are counted

A claim about Solana’s transaction dominance has prompted a sharper intervention from Internet Computer founder Dominic Williams, shifting the discussion from raw numbers to how blockchain activity is defined and compared.
The exchange began after Brennan Watt from Anza said Solana had processed 200 billion transactions over the past two years, a total he suggested exceeded the combined lifetime transaction count of every other blockchain. The figure circulated widely, reinforcing Solana’s reputation for high throughput and fast settlement.
Williams responded by reframing the comparison rather than disputing Solana’s performance. Solana, he said, is “doing great”, but the claim only holds when counting financial transactions such as transfers and trades. When general transactions are included, Williams argued, the Internet Computer has processed vastly more activity.
That distinction goes to the core of how different blockchains position themselves. Williams has long argued that ICP should be understood as a general-purpose compute network, where transactions include web requests, smart contract execution and application-level actions that resemble traditional internet activity. Many of these operations would never appear in transaction metrics on networks designed primarily for payments and trading.
Solana’s figures, by contrast, are typically framed around financial throughput, an area where it has invested heavily and where comparisons with other layer-one blockchains are more straightforward. For markets focused on decentralised finance, NFTs and on-chain trading, those numbers carry clear relevance.
The disagreement highlights a broader issue in crypto analysis. Transaction counts are often presented without shared definitions, making headline comparisons misleading. A higher number may reflect different design choices rather than greater adoption or economic impact.
Independent analysts frequently caution that volume alone does not capture network value, noting that security, decentralisation, developer activity and real-world usage all matter. In that context, Williams’ response can be read as a reminder that blockchains optimised for different purposes will naturally produce very different metrics.
Rather than settling the debate, the exchange underlines how contested blockchain scorekeeping remains. As networks continue to branch into finance, computing and consumer applications, what qualifies as a transaction may prove as important as how many are counted.