Ethereum chatter has collapsed to 2020 levels while banks build on the chain and a nonprofit teaches institutions to buy it. The token trades as if neither audience exists. Three groups are pricing three different assets, and only one of them is right.
Summary
- Retail attention on Ethereum has collapsed: tweet volume is at 12-month lows near 40,000 mentions, levels last seen in 2020, NFT activity has gone quiet, and daily active addresses have fallen from above 1.5 million in January toward 544,000.
- Institutional commitment is moving the opposite way: a dedicated nonprofit launched to onboard institutions, tokenization is a headline topic in traditional finance, ETF flows turned positive again in July, and BlackRock, JPMorgan, and Robinhood all build on Ethereum rails.
- The price has ignored both signals, trading near $1,800, down roughly 42% this year and about 64% from its August peak near $5,000, while network fee revenue sits near cycle lows.
- The loudest defection came from inside: Bankless co-founder David Hoffman sold his remaining ETH in May, arguing the money thesis has run its course, and doubled down this month on the fee problem behind it.
- The divergence resolves through one question, value accrual: whether the activity institutions bring ever becomes fees the token captures. Retail priced a story that died. Institutions price rails that work. The token prices cash flows that keep falling.
Three different groups of people are currently looking at Ethereum, and they are not seeing the same asset. The first group, crypto-native retail, has mostly stopped looking: social mentions of Ethereum have fallen to roughly 40,000, a level last recorded in 2020 when Wall Street did not know the chain existed, and the loud consumer corners of the ecosystem, NFTs above all, have gone quiet enough to hear the servers hum. The second group, institutional finance, is arriving in the opposite direction, with a purpose-built nonprofit teaching banks how to hold ETH, tokenization on every conference agenda, and the largest asset managers in the world settling real products on Ethereum rails. And the third participant, the market itself, is pricing the token as if neither group matters: ETH trades near $1,800, down about 42% on the year and nearly two-thirds below its August peak, while the chain’s fee revenue scrapes along at cycle lows. In May, the divergence produced its emblematic moment, when one of Ethereum’s most committed public advocates announced he had sold every coin he owned while insisting he still believed in the network. All three groups are behaving rationally. They are simply pricing three different things, and working out which of the three the token actually is has become the most consequential question in crypto’s second-largest asset.
The retail exit, measured
The evidence that ordinary crypto users have checked out of Ethereum is not anecdotal; it shows up in every proxy for attention and grassroots usage at once.
The cleanest measure is the crudest: how much people talk about it. Tweet volume for Ethereum has fallen to fresh 12-month lows around 40,000 mentions, with Bitcoin near 130,000, and the comparison point is what makes the number land, because attention this low was last seen in 2020, before the ETFs, before the Merge, before the institutional era the industry spent a decade demanding. Social chatter is a rough instrument, but it has historically tracked retail capital and marked cycle temperature, and its collapse while institutional adoption sets records is precisely the inversion that makes this moment strange. Rising mentions once meant rising retail inflows; now the crowd that generates mentions has left the theater.
On-chain, the story repeats with better instrumentation. Daily active addresses, above 1.5 million in January, have trended down toward 544,000, a fall of nearly two-thirds that tracks the price drawdown from above $3,400 in December to under $2,000. The consumer economy that once made Ethereum a cultural object, NFT trading, consumer mints, the speculative long tail, has thinned to the point where daily NFT volumes measure in the hundreds of thousands of dollars against a $41 billion DeFi treasury sitting largely still. The capital stayed; the crowd left. Total value locked has barely budged through the attention collapse, which tells you who remains: professional and semi-professional capital that thinks in quarters, parked in lending markets and liquid staking, indifferent to vibes.
The generous reading of the exit is rotation, that retail attention went to memecoins on faster chains and to the AI trade, and rotations reverse. The harsher reading is that Ethereum’s retail base was loyal to a story, the ultrasound money, world-computer, ETH-is-money story, and stories do not survive a 64% drawdown from peak while the supply inflates and the burn sits idle. Either way, the measurable fact stands: the audience that carried Ethereum through every previous cycle is not currently in the building.
The institutional entry, measured
Run the same exercise on institutions and every needle points the other way, which is what makes this a divergence, not a decline.
The most explicit signal is organizational: the launch of Ethereum Institutional, a nonprofit created specifically to educate banks, asset managers, and corporates on adopting Ethereum, with contributors drawn from the ecosystem’s core. Institutions do not get dedicated onboarding bodies for networks in decline; the entity exists because inbound demand outgrew the ecosystem’s capacity to answer it. Around it sits a thickening layer of professional evangelism, Etherealize pitching Wall Street directly, with its leadership publicly arguing that institutional engagement has moved past pilots into production, and a restructured Ethereum Foundation spinning out ETH Systems as a for-profit focused on institutional privacy tooling, funded by trading firms and treasuries. The ecosystem is visibly reorganizing itself around the client it now serves.
The client, meanwhile, keeps shipping. The tokenization wave that dominates traditional-finance conferences runs disproportionately on Ethereum and its L2s: BlackRock’s tokenized fund complex, JPMorgan’s settlement infrastructure reaching public rails, Robinhood building its chain as an Ethereum L2, stablecoin issuance concentrating on the network that hosts the deepest collateral markets. For more context on the institutional product driving adoption, crypto.news has explained how tokenized money market funds are moving regulated cash instruments on-chain. Even the flow data, the weakest leg of the institutional case, has stopped arguing against it: after a heavy second quarter of net outflows, US spot ETH ETFs turned positive again in July, with inflow days in the tens of millions, uneven but real. And the treasury bid persists through the drawdown, with corporate and fund vehicles continuing to accumulate at prices the retail cycle would have considered a catastrophe.
Institutions, in short, are doing exactly what the industry spent years saying it wanted: adopting the infrastructure, at scale, without asking permission from the price. Which sharpens the puzzle instead of resolving it, because their arrival has coincided with the asset’s worst sustained underperformance of the modern era.
The defection that named the problem
The reason the price ignores both audiences was articulated most clearly by the person whose exit hurt the narrative most.
David Hoffman spent years as one of Ethereum’s most effective advocates, co-founding Bankless and popularizing the ETH-is-money thesis, the argument that Ethereum’s token would become the internet’s base money, scarce, productive, and re-rated accordingly. On May 21 he sold the last of his personal ETH, and his explanation was more damaging than the sale: the thesis, he argued, has largely run its course, with ETH unlikely to be re-rated meaningfully higher or lower from here, money to some degree, but not the maximally successful version the ecosystem set out to build. Former core developer Eric Connor’s response compounded it, noting ETH has grossly underperformed the broader crypto market for years and attributing the lag to relentless supply from early millionaires, not protocol failure, an explanation that manages to be reassuring about the technology and damning about the asset simultaneously.
Hoffman has kept pressing the underlying point since, arguing this month that Ethereum faces a false choice between maximizing fees and being money, and that while it hesitates, distribution-rich competitors, Robinhood’s chain among them, are positioned to eat the revenue base out from under it. That is the distribution rival eating the revenue base. Strip the personalities away and his case reduces to an arithmetic claim: layer-one tokens are ultimately priced on the fees their block space earns, Ethereum deliberately pushed activity to L2s that pay almost nothing back, mainnet fee revenue has fallen from roughly $40 million a day in early 2025 toward $10 million, and no amount of institutional construction on top of the network changes the token’s cash flows if the construction happens where the token does not collect rent. It is the value-accrual critique, delivered by someone who spent five years selling the opposite conclusion, which is exactly why it landed.
Three prices for three assets
Here is the resolution of the divergence, and it requires taking all three groups seriously at once, because each is pricing a real thing.
Retail priced the story, and the story died. The asset retail owned was ultrasound money: a supply that shrinks with use, a burn that turns adoption into scarcity, a meme that fit on a sticker and compounded reflexively. That asset genuinely existed for a stretch after the Merge and genuinely does not now, with the burn collapsed, supply mildly inflating, and the December blob-fee floor a patch on the leak, not a restoration. That is the monetary mechanics under this divergence. Attention followed the story out. Retail is not wrong to be gone; the thing it bought is gone.
Institutions price the rails, and the rails work. The asset institutions are adopting is not the token’s monetary narrative but the network’s properties: the deepest liquidity, the most battle-tested settlement, the compliance tooling, the credible neutrality that lets BlackRock and a DeFi protocol share infrastructure. That asset is thriving, and nothing in the price contradicts it, because most institutional use, tokenized funds, L2 settlement, stablecoin rails, consumes Ethereum’s security while paying trivially for it. Institutions are not wrong to build; the thing they are buying works regardless of what ETH costs.
The market prices the cash flows, and the cash flows are falling. The token, stripped of both stories, is a claim on fees plus a staking yield plus a monetary premium the market is currently revoking. Fee revenue down roughly three-quarters from early 2025, activity migrated to venues that remit almost nothing, and a persistent seller overhang from the early-holder class Connor described: the price is not ignoring the fundamentals, it is agreeing with them, and its verdict is that until institutional construction becomes token revenue, construction is not a bull case.
Which means the entire divergence compresses into one testable question: does the institutional economy on Ethereum ever start paying Ethereum? The mechanisms are known and partly shipped, the blob-fee floor reconnecting L2 growth to burn, mainnet settlement of high-value tokenized assets that does pay real fees, staking demand from treasuries and ETFs that locks supply. If tokenization scales and its settlement gravity pulls value to mainnet, the fee line inflects, and the market re-rates the token toward what institutions already believe about the network. If the activity stays where the rent is lowest, Ethereum becomes magnificent public infrastructure attached to a stagnant asset, the outcome Hoffman priced when he sold. Both futures are live. The tape, for now, is voting with him, and the burden of proof sits, for the first time in Ethereum’s history, on the bulls’ arithmetic rather than their story.
One more actor deserves a paragraph before the watchlist, because the divergence is reorganizing Ethereum’s own institutions in real time. The Ethereum Foundation, historically the ecosystem’s ambivalent center, has spent the year restructuring around exactly the split this piece describes: research and protocol work continuing in the nonprofit core, a new institutional-outreach apparatus forming at arm’s length, and ETH Systems spinning out as a for-profit, funded by trading firms and corporate treasuries, to build the privacy and compliance tooling institutional users keep requesting. Longtime contributors have scattered across the new entities, and the ecosystem’s own commentators describe the reorganization with a candor that borders on gallows humor. The institutional turn, in other words, is not something happening to Ethereum from outside; it is something Ethereum’s leadership has chosen, budgeted, and staffed, accepting the retail exit as a completed fact and reallocating toward the audience that stayed. That choice has consequences for the token question this piece turns on. An ecosystem organized around institutional settlement will prioritize exactly the upgrades, privacy, compliance hooks, high-value mainnet settlement, most likely to make institutional activity pay mainnet fees, which is the bull path. It will also, inevitably, deprioritize the consumer-facing culture that once generated the monetary meme, which forecloses the old path back. The foundation has effectively placed the ecosystem’s bet for it: that the second audience can be converted into revenue before the absence of the first audience becomes terminal for the asset’s premium. The fee line, again, will grade the wager.
What to watch
Three lines on three charts settle this faster than any debate.
The fee line. Daily network fee revenue near $10 million is the bear case in one number; a sustained inflection, driven by blob-fee floors under growing L2 volume or high-value mainnet settlement, is the single cleanest signal the value-accrual gap is closing. Watch the trend through the fall, not any single week. That is where the fee line actually comes from.
The flow composition. ETF inflows resumed in July after a negative quarter; whether they compound, and whether staking-enabled vehicles and treasuries keep locking supply through price weakness, tests whether the institutional bid extends from the network to the token. Uneven, headline-driven flows extend the stalemate; a durable streak changes the supply math. Crypto.news has also explained how the flow machinery works.
The attention floor. Retail metrics this depressed have historically marked accumulation zones as often as terminal decline, and tweet volume at 2020 levels with institutional adoption at record highs is a configuration crypto has simply never printed before. If price ever starts responding to the institutional story, the crowd’s return would be the accelerant. Its continued absence is the cheapest real-time measure of how dead the old narrative remains.
Ethereum’s strange summer is best understood as an estate in probate. The old asset, the retail money-meme, has died, and its heirs have left. The new asset, institutional settlement infrastructure, is thriving but pays no rent to the name on the deed. And the token is the estate itself, valued daily by a market that only counts income. The network has never been more used or less loved, and the gap between those two facts is either the buying opportunity of the cycle or the proof that usage was never the same thing as value. Three audiences have placed their bets. The fee line will grade them.
Frequently asked questions
What does the retail exit from Ethereum look like?
Tweet volume for Ethereum has fallen to roughly 40,000 mentions, a 12-month low last seen in 2020, while Bitcoin sits near 130,000. Daily active addresses have declined from above 1.5 million in January toward 544,000, NFT activity has thinned to daily volumes in the hundreds of thousands of dollars, and the consumer-speculative corners of the ecosystem have gone broadly quiet, even as DeFi’s roughly $41 billion in locked value stays put.
What is the evidence institutions are moving in?
A dedicated nonprofit, Ethereum Institutional, launched to onboard banks and asset managers, alongside Etherealize’s direct Wall Street outreach and the Ethereum Foundation spinning out a for-profit institutional tooling arm. BlackRock’s tokenized funds, JPMorgan’s settlement rails, and Robinhood’s L2 all build on Ethereum, tokenization dominates traditional-finance agendas, ETH ETF flows turned positive again in July, and treasury vehicles kept accumulating through the drawdown.
Why did David Hoffman sell his ETH?
The Bankless co-founder sold his remaining ETH on May 21, arguing the ETH-is-money thesis has largely run its course and that he does not expect the market to re-rate the asset meaningfully in either direction. He has since pressed the structural point: layer-one tokens are priced on fees, Ethereum’s activity moved to L2s that pay almost nothing back, and competitors with distribution are positioned to erode the remaining revenue base.
Why is the ETH price ignoring institutional adoption?
Because most institutional use pays the token almost nothing. Tokenized funds, L2 settlement, and stablecoin rails consume Ethereum’s security while generating minimal mainnet fees, and daily fee revenue has fallen from roughly $40 million in early 2025 toward $10 million. The market prices the token on cash flows plus monetary premium, and with the premium fading and fees falling, the price tracks the arithmetic, not the adoption headlines.
Is this different from the ultrasound money problem?
It is the same root with a different face. The ultrasound story broke because cheap L2 data ended the fee burn that made ETH deflationary, which is monetary mechanics. This divergence is about audiences: retail owned the monetary story and left when it died, institutions own the infrastructure story and keep building, and the token’s price follows fees rather than either narrative. The December blob-fee floor addresses both by reconnecting L2 growth to mainnet revenue, at a baseline level.
What would make the price start responding?
A durable inflection in fee revenue is the cleanest trigger: growing L2 volume paying meaningful blob fees under the December floor, high-value tokenized-asset settlement on mainnet, and staking demand locking supply through ETFs and treasuries. If institutional activity starts converting into token cash flows, the market has something to re-rate. Without that conversion, adoption and price can stay decoupled indefinitely.
Could retail attention at 2020 levels be a buy signal?
Historically, deeply depressed attention has coincided with accumulation zones as often as with terminal decline, and the current configuration, record institutional adoption against 2020-level retail interest, has no precedent to price from. Low attention removes a reflexive bid but also exhausts sellers. It is a condition, not a signal, and its resolution depends on the fee and flow lines rather than on sentiment itself. This is not investment advice.
What are the key numbers to track from here?
Daily network fee revenue against the roughly $10 million cycle low, the persistence of ETH ETF inflows after July’s turn positive, staking and treasury accumulation as a share of supply, active addresses against the 544,000 area, and the growth of tokenized-asset settlement that pays mainnet fees. Together they answer the only question that closes the divergence: whether use of Ethereum ever becomes revenue for ETH.
Disclaimer: This article is for information and educational purposes only and does not constitute financial or investment advice. It describes market conditions and network metrics that change quickly, and past patterns do not guarantee future outcomes. Nothing here is a recommendation to buy, sell, or hold any asset. Always do your own research. Information is accurate as of July 21, 2026.
