Pump.fun's revenue is the cleanest single illustration of crypto's bull-market product-market fit: hundreds of millions of dollars in annualized fees from a small team running a single core product. The headline is real and impressive. The harder question, asked too rarely, is whether the revenue is durable. My answer, with apologies to its many enthusiastic defenders, is that it almost certainly is not, and the analytical framework needed to see why is the same one that has correctly called every crypto-cycle exchange revenue chart since 2014.
The reason is the activity profile. Pump.fun's fees come almost entirely from token-launch trading, an activity that is procyclical to the point of caricature: it surges in bull markets and collapses in bears. The platform's monthly fee revenue tracked Bitcoin price almost perfectly through 2024 and 2025, with correlation coefficients above 0.85 across multi-month windows. There is nothing wrong with that — it is exactly what a well-run, bull-market-aligned product looks like. It is also, definitionally, not a recession-proof revenue stream, and projecting current fees forward as a steady-state expectation is a forecasting error in formation.
The 2018 and 2022 bear cycles produced near-zero issuance volumes for similar (less efficient) launchpad-style products. Bittrex's altcoin listing fees collapsed roughly 90% from peak to trough across the 2018 bear market. Coinlist's primary issuance volumes did the same in 2022. There is no reason to think Pump.fun is structurally insulated from that pattern. The post-bull revenue trajectory will, in my view, look more like the typical crypto-cycle exchange revenue chart — peak revenue down 80 to 90% within twelve months of the cycle top — than the steady recurring-revenue stream that some of its admirers describe.
The defense is that Pump.fun has built a more sticky, more habit-forming user funnel than past launchpads, that its product loop is more entertainment-than-investment-driven, and that the sheer cultural momentum around memecoin launches will sustain meaningful baseline volume even in adverse market conditions. There is a fair point in the entertainment framing. Pump.fun's user base does behave more like a gambling-platform cohort than an investment cohort, and gambling demand is more bear-resistant than investment demand. But it is not bear-immune, and the magnitude of the durable baseline is going to be far smaller than the bull-market peak, just as it is for any sportsbook in a recession.
The on-chain analytics community has begun to model this carefully. Recent dashboards from Dune contributors and Allium Labs estimate that Pump.fun's "true" durable revenue floor — the share of fees not driven by speculative momentum — is between 8% and 18% of peak revenue, depending on which months and cohorts you include in the baseline. That implies a $400-million annualized peak revenue stream becomes a $30 to $70 million floor. Still a real business, especially at the operating-cost structure of a small team. Not the multi-billion-dollar growth story that some valuations imply, however. The Solana DEX volume share that Pump.fun captured during the bull peaks of 2024 was historically anomalous in any market category outside of pure speculation. There is no reason to expect it to persist when the speculative impulse cools. Smart money is already modeling the cooling into next year's revenue projections; retail equity holders mostly are not, and the mismatch will be paid in someone's portfolio.
What to watch is the rollout cadence of similar mechanics on competing chains and the response of Pump.fun's user retention to the next 30%-plus Bitcoin drawdown. If the team can produce a credible secondary product line — sustainable trading fees from a non-launchpad activity, perhaps — the category survives the bear cycle in healthier shape. If they cannot, the bear-market revenue collapse will be ugly, and the equity holders who paid current valuations will learn the same lesson their 2017 and 2021 predecessors learned about cyclicality. Either way, treating the current revenue as a permanent feature of the landscape is a forecasting mistake the industry has made before, and is now in the process of making again.