There is something that may have struck a chord with many DeFi developers—the largest fixed monthly expense for a well-operated protocol is not hiring, but data fees.

Sounds a bit unbelievable? But this is the current situation. To ensure the accuracy and real-time nature of on-chain price data, protocols must continuously pay for oracle services. Regardless of whether the market is active or how many users there are, this "data leasing fee" must be paid. The logic of this "push" model is simple: the oracle continuously pushes data to the protocol, and the protocol pays periodically.

The question is—do we really need this?

Actually, there is an alternative. The pull-based data model completely overturns this logic. The concept is straightforward: you only fetch data when needed, pay per use, and only for what you consume. As a result, data costs shift from "fixed expenses" to "flexible expenses." For early-stage protocols or projects with fluctuating user activity, this instantly relieves cash flow pressure. Especially in a bear market, this is not just optimization—it's a matter of survival.

At the technical core, this model relies on a sufficiently large and stable data pool to support it—ensuring data resilience through hybrid verification mechanisms and price-weighted algorithms. Protocols no longer need to bear long-term fixed costs; they only pay for fresh data when actually called upon.

Changing the cost structure often means a complete reorganization of the entire ecosystem's operational logic.