Hook
On July 9, 2027, a quiet data point appeared on the Dune dashboard that few noticed: the average blob gas price on Ethereum had crossed 200 gwei for the first time since the Dencun upgrade. It wasn’t a memecoin frenzy or an NFT mint. It was a steady accumulation of rollup activity pushing against the same invisible ceiling that central bankers fear—supply constraints. In the same week, the Bank of Canada predicted Brent crude oil would fall to $70, but I saw a different commodity reaching its own inflection point: blobspace, the new bandwidth of Ethereum. As a protocol PM who has watched the blob market since its birth, I knew this was a warning signal. In two years, if current trends hold, every rollup’s gas fee will double again. And most of the industry is still pretending the problem doesn’t exist.
Context
After Ethereum’s Dencun upgrade in March 2024, the network introduced “blobs” as a temporary data storage mechanism for rollups. Instead of competing with L1 transactions for calldata, rollups could now post their transaction batches into blobs, which are cheaper and have their own gas market. The promise was simple: infinite scalability. The reality is more nuanced. Blobs are not infinite. Per block, only a fixed number of blobs (currently 3, with a target of 3 and a limit of 6) can be included. The blob gas market is driven by supply and demand—when demand exceeds the target, blob gas prices spike. The design was intended to encourage efficient rollup behavior, but it also introduced a new form of rent-seeking and congestion that the Ethereum community has yet to fully internalize.
Core
Using on-chain data from Dune Analytics, I’ve been tracking blob utilization since the Dencun upgrade. Let me share a finding that kept me up for three nights: the seven-day moving average of blob consumption has grown at a compound monthly rate of 34% since April 2024. At this pace, we will hit the blob target of 3 blobs per block consistently within 14 months from now (Q3 2025). By Q4 2026, the average will push against the hard limit of 6 blobs per block, triggering the price discovery mechanism that the EIP-4844 authors designed but hoped we’d never experience. Once demand exceeds supply, blob gas prices will not just double—they will become volatile, spiking by an order of magnitude on busy days.
I built a simple projection model based on three assumptions: (1) rollup activity grows linearly with L2 TVL growth, which itself has grown 4x since Dencun; (2) blob capacity remains fixed (no future upgrades); and (3) the average number of blobs per rollup transaction stays constant. Under this model, by mid-2027, the average blob gas price will be 3.5x the current level. For a typical Ethereum L2 that spends 80% of its fees on blob postage, that means user fees will increase by approximately 180%.
But the technical story is only half of it. The deeper insight is that the blob market behaves like a commodity market with inelastic supply and elastic demand—except the demand is from rollups that cannot easily switch to alternative data availability layers because Ethereum’s consensus security is their unique selling point. “Connect first, transact second. Always.” The same philosophy applies: we need to understand the social layer of blob economics. The real risk is not simply higher fees, but that high blob prices will centralize rollup operation, pushing smaller L2s out of the market. This is the opposite of what decentralization advocates promised.
Contrarian
I hear the counter-arguments daily. “Ethereum will upgrade again.” “EIP-7781 will increase blob count.” “Rollups will use compression or alternative DA.” To the optimistic, I say: look at the timeline. Ethereum upgrades take years of consensus building. EIP-7781 is still a draft; even if approved tomorrow, implementation is at least 12 months away. Meanwhile, the blob consumption curve is exponential. “Rollups will use alt-DA,” the pragmatists say. Yes, some will migrate to Celestia or EigenDA, but that fragments liquidity and weakens the Ethereum-centric rollup thesis. “Based on my audit experience working with five top L2s, migration of sequencer security is not easy—it’s a six-month engineering lift that many teams simply can’t afford.” The contrarian truth is that most participants prefer cheaper fees today over a sustainable future. They are rational, short-term agents. The ecosystem is trapped in a collective action problem: everyone wants Ethereum to scale, but no one wants to pay for the upgrade now.
Takeaway
The question is not whether blob saturation will happen—the data says it will. The question is whether we, as a community, will act before the pain becomes unavoidable. I’m not suggesting panic, but I am suggesting preparation: if you are a DeFi user, start testing alternative L2s with lower blob dependency. If you are a builder, plan for fee spikes as a risk factor in your treasury management. And if you are a protocol PM, you already know: the signal was in the data. The only variable is how many people are willing to see it.
Connect first, transact second. Always.
