Original title: How Pumpfun is killing the trenches

Original author: Sparsity @sparsitydata

Editor's note: In Q3 2026, 3.05 million Meme coins were issued on Pump.fun, and only 24 ever reached a $10 million market cap. According to statistics from on-chain data research team Sparsity, Q4 2024 saw a similar issuance volume of 3.16 million, yet 207 reached the same market cap threshold. Token issuance has returned to near the previous peak, but the number of tokens per 100,000 issued that can rise to a $10 million market cap has fallen from about 6.5 to 0.8.

When the Meme coin market weakens, communities tend to point fingers at snipers, mass token-launching teams, and KOLs who pump and dump. Sparsity traced the problem one level up, examining where Pump.fun's revenue actually comes from. According to its statistics, about 95% of the platform's protocol revenue comes from coins still in the bonding curve stage or with a market cap below $250,000. The bonding curve is the pricing mechanism in a new token's early issuance period, where buying and selling push the price up or down along preset rules. At this stage, every buy and sell incurs a 1.25% fee, of which 0.95% goes to the platform and 0.30% to the token creator. Even if the price stays flat, a buy-then-sell round costs about 2.5%. By token age, about three-quarters of the platform's revenue comes from tokens issued less than an hour ago.

This fee structure has diverged the interests of the platform, token creators, and buyers. For the platform and creators, a token only needs someone buying and selling right after issuance to generate fees. For buyers, profit still depends on someone later willing to pay a higher price to take over. Creators can clone popular coins or launch dozens of tokens at once, and as long as a few generate trading volume at open, there is revenue. Sparsity statistics show that the top 25 wallets by issuance volume collectively issued 761,000 tokens, fewer than 1% of which reached the threshold to "graduate" from the bonding curve stage, yet these wallets cumulatively earned $128 million.

Callout rewards also amplified this divergence. Pump.fun once paid rewards based on trading volume generated by callouts, and low-market-cap coins were more easily driven by concentrated buying. On October 10, the platform said it would factor whether copiers made money into reward criteria and reduce callout rewards for low-market-cap coins. But in the following three weeks, according to Sparsity statistics, among the top 20 callers by reward ranking, the share of bonuses going to those mainly promoting low-market-cap coins actually rose from 68% to 94%. Among 14 callers separately tracked by the author, the median price change of called coins was up 6% one minute after posting, down 48% after one hour, and down 74% after one day. For each buyer, selling early makes sense. But as more and more people rush to sell, fewer are willing to take over and keep holding, making it harder for a new coin to survive its initial hype.

The article concludes by proposing a fixed-price fundraising concept. Buyers subscribe at the same price, automatic refunds occur if the minimum fundraising amount is not met, and once fully raised, the contract establishes a liquidity pool, with the platform charging fees only upon full subscription. This can weaken the advantage of front-running at launch and reduce the platform's channel for profiting from batch failed issuances, but issues like splitting wallets to bypass limits and immediate selling after listing still persist. What this concept can change is the timing of when the platform receives money. Today, even if a coin only survives for an hour, as long as buying and selling are frequent enough, the platform has already made money. Delaying fees until fundraising completion gives the platform a more direct incentive to care about whether an issuance can attract enough buyers.

The original text follows:

"On-chain conditions are bad." "All KOLs are harvesting." "We are still worshipping those who use bundled transactions to snatch up tokens."

These problems are indeed prominent, but finding the root cause at the very bottom of this system doesn't make much sense. The holding culture at the frontline of Meme coin trading has collapsed. No one is willing to hold coins anymore. Everyone copies projects to chase the traffic of hot coins, launching dozens of coins at once, using bundled transactions to snatch up tokens at launch, and then relying on shilling to farm rewards, because these things make money. The problem lies higher up, with the platforms that control this market. The largest Meme coin launch platform is precisely paying for the behavior everyone is complaining about, and has turned Meme coins into a market with infinite supply and constantly disappearing demand.

Now, everyone wants KOLs to be held accountable for their shilling, which is a good thing. But if the platform paying the shillers never has to answer for anything, holding only the shillers accountable is not enough.

pump.fun makes money from coins that ultimately go to zero

Over the past 30 days, among Meme coin launch platforms across all chains, pump.fun accounted for 92% of issuance volume and 84% of trading volume. In this market, it decides what makes money.

95% of its protocol revenue comes from trading tokens still in the bonding curve stage or with a market cap below $250,000. On the bonding curve, every buy and sell incurs a 1.25% fee, of which 0.95% goes to pump.fun and 0.30% goes to the creator. That means even if the price doesn't move at all, a round trip of buying and selling costs 2.5%.

Protocol revenue contributed by bonding curve and low market cap tokens

Money was made when the token market cap was just in the hundreds of thousands of dollars; after that, it didn't matter how the price moved.

The same holds true when looking at token launch times. Three-quarters of this revenue came from tokens launched less than an hour ago, and 87% came from tokens launched less than a day ago.

Revenue by token launch time

In the first hour after a token launch, pump.fun had already made its money.

The tokens that surged the most contributed almost negligible revenue. Among the batch of tokens that rose the most, 15 of them reached a combined market cap of $1.4 billion at their respective peaks, bringing pump.fun total revenue of about $1.5 million. In contrast, tokens still trading on the bonding curve or with a market cap below $250,000 brought it $506 million. Simply put, 1,000 tokens that fizzle out at a $20,000 market cap make it more money than one token that reaches a $100 million market cap.

Comparison of fee revenue from top surging tokens versus small tokens

The revenue contributed by these 15 surging tokens was less than 0.33% of the revenue contributed by small tokens.

So the platform's product design revolves around increasing token issuance volume. Every incentive on the platform drives up issuance numbers, because the more that are issued, the more the platform earns.

Four Incentives

Every behavior criticized on social media has a reason to exist on this platform. All four behaviors bring in more transactions, and every transaction pays a fee to pump.fun.

The relationship between the four behaviors and platform fees

Bundled sniping, batch token issuance, shilling, and PvP mode all ultimately bring the platform the same kind of fee.

Bundlers snipe in the opening block because it's cheap

Bundlers exist because a token's initial market cap and liquidity are both very small. By spreading a few SOL across dozens of wallets and buying together in the block the token launches, they can grab a large portion of the supply and then sell that stash to you, the later entrant. The lower the initial market cap, the cheaper this is—and the initial market cap is set by pump.fun.

Token creators mass-produce tokens because creator fees are paid based on trading volume

The reason someone continuously issues tokens is that creator fee rewards are tied to token issuance. Whether a token survives for an hour or a year, creators earn 0.30% from every trade. So the most cost-effective strategy is to issue as many tokens as possible, capture fees from early trading volume, and then move on to the next one. The top 25 issuers by token count collectively issued 761,000 tokens, of which fewer than 1% "graduated" from the bonding curve stage, and they earned $128 million.

Issuance volume and revenue of top token-issuing wallets

The No. 1 wallet issued 171,195 tokens, earning about $200 per token on average.

Mass token issuance occasionally produces one or two big winners. Two wallets each had a best-performing token reach market caps of $122 million and $327 million, respectively. But producing a big winner is merely a byproduct of high-volume trading — trading volume itself is the basis for getting paid.

Shillers push low-market-cap tokens because rewards are distributed based on trading volume

Shilling rewards gave everyone a new reason to push tokens. The platform turned token promotion into a revenue stream. Since August, it has distributed $29.6 million in shilling rewards to 51,115 wallets, with rewards heavily concentrated. 10 wallets took 16%, while the median amount received by reward recipients was $13.

Distribution of shilling rewards

These 10 wallets each received an average of about $485,000.

Rewards are distributed based on the trading volume generated by each shill, and the smallest tokens are the easiest to drive trading volume. In August, among the top 50 shillers by revenue, 40 shilled tokens with an average market cap below $100,000.

Average market cap of tokens promoted by top shillers

Among the top 50 shillers, 12 promoted tokens with an average market cap below $10,000.

pump.fun has heard this feedback. On October 10, the platform said future rewards will depend on whether copiers made money, and shilling rewards for low-market-cap tokens will be reduced. In the following three weeks, among the bonuses received by the top 20, the share taken by shillers whose promoted tokens had a median market cap below $100,000 rose from 68% to 94%. Weekly bonus payouts also increased by 53%.

Call rewards and the share of low-market-cap calls before and after the rule change

After the rules changed, the total prize pool increased, and the share received by low-market-cap callers also rose.

Most of the cost is borne by copy traders. We tracked 14 callers, and the median market cap of the coins called by 10 of them was below $100,000. These copy traders lost a combined $926,000 one day after the calls. Based on the median price change of the coins called, they rose 6% 1 minute after the call, fell 48% after 1 hour, and fell 74% after 1 day. Whether the coin price rises or not, the callers get paid.

Copy-trading P&L for the 14 callers

Of the 14, only 1 had copy traders who were profitable overall, making $10,000.

Performance of each caller's called coins one day after the call

For every caller, the median price change of the coins they called one day later was negative.

Some ask why KOLs always push coins that lose their followers money. A large part of the reason is here. Someone pays them to do it.

PvP mode makes each user do a few more trades

PvP (player versus player) works because no new money comes in, and veterans make money off newcomers. These modes are designed exactly on this logic. Trade execution is faster, each user does more trades, and every trade charges a fee. On June 30, pump.fun itself delisted one of these modes, the Tokenized Agent mode. The platform said at the time that there were too many ways to issue coins, causing "harmful player-versus-player" behavior.

Infinite supply, vanishing demand

If these incentives last long enough, the market will split into two sides.

On the demand side, you rush in to buy coins and on average lose money. Only a very small number of wallets make a decent amount. On the supply side, you issue coins and get paid with almost no risk.

There is no barrier to becoming a supplier. You can issue a coin with one click, and you already know how to do it. In seconds, you can make a clone of a hot coin, wash-trade to make it look like people are playing, and then buy historical wallets by the hundreds or thousands. This playbook gets easier every month.

So we believe that people who might have been long-term holders will increasingly shift to the supply side. Issuing coins is profitable; holding coins is not. Every new player who figures this out becomes yet another coin issuer, competing for an ever-shrinking pool of buyers. That's how holding culture collapses. No one collectively decided to stop holding — the platform made holding a losing trade.

The results show up in how these new coins perform. pump.fun's current coin issuance volume is about the same as at the market's peak in late 2024. According to our count, 3.16 million coin types were issued in Q4 2024, of which 207 reached a market cap of $10 million. In Q3 2026, 3.05 million were issued, and only 24 reached that market cap.

Weekly issuance volume vs. the rate of breakout coins

Weekly issuance has again surpassed 300,000 coin types. For every 100,000 coins issued, the number of breakout coins has dropped from 6.5 to 0.8.

Some will point out that trading volume on the bonding curve has been rising since June. That's true, but combining bonding curve and PumpSwap volume, it has fallen from roughly $25 billion to $30 billion per month in Q1 2026 to about $13.8 billion in October (bonding curve, PumpSwap). And pump.fun makes money from trading volume. Buyers make money from breakout coins — and the number of those coins is now one-ninth of what it was.

The survival data tells the same story. pump.fun has issued a total of 24.4 million coin types, and only 312 still have a market cap above $100,000. That's about one in every 78,000 coins — 0.0013%.

Token survival rate comparison across launchpads

Most launchpads issue only a fraction of pump.fun's volume, but their token survival rates are higher.

This is the first market we know of where supply is growing this fast while demand is shrinking. If the platform weren't paying for this behavior, it would never have grown to this scale.

We're responsible too

pump.fun built this system, but the trading volume that keeps it running is what we contribute. Every incentive mentioned in this article requires someone on the other side of the trade to generate revenue. Most of the time, that someone is us.

We Reward the Behavior We Complain About

We say bundlers are the problem, then turn around and pay attention to them, wondering how the next round will play out. A developer forks a hot coin, takes the fees, and walks away—and someone posts a thread analyzing how clever the move was. A person calls 20 coins a day, and everyone follows him because he's always first to call it—whether the copiers make money becomes secondary. Of the 14 callers tracked above, only 1 had copiers who were profitable overall. All copiers combined lost $926,000 one day after the calls.

We pay for all of this twice. Once with trading volume—pump.fun and creators make money from it. And once with attention—the next fork and the next call are both aimed at our attention.

Holding Is No Longer a Strategy

A year ago, the play was to find a coin early, endure the noise, and hold. Now, one day after a call, the median price change of the called coins is down 74%, so the only survivors are those who sell first. For each of us individually, this is rational—but everyone doing it is precisely what prevents anyone from catching a big runner. If no one is willing to hold a coin past the first hour, it can't reach a $10 million market cap. Yet three-quarters of pump.fun's revenue comes from that one hour.

What Happens If We Change Our Ways

Doing these things doesn't require pump.fun's consent. Hold the coins you genuinely believe in; stop buying in and flipping out the moment you see a chart. Judge callers by whether their copiers made money, not by how often they post. Stop spreading forks and bundling as clever plays. Direct your trading volume to launchpads that can make money from coins that survive long-term. Any one of these alone won't change the incentive system. But every trade you don't contribute to this system is one less dollar it earns.

How to Change the Incentives

We're offering these ideas directionally; the specific designs still need engineers to work out. Each approach tries to direct money toward the phase after a token's first hour of issuance.

Raise the initial market cap, deepen initial liquidity, and let liquidity accumulate continuously so the pool grows with trading volume.

Share fees with holders and liquidity providers (LPs) as well, and have creator fees unlock gradually based on how long the token survives.

Make call rewards depend on whether copiers hold. Pay callers based on whether copiers still hold the coin days later and are still profitable.

A similar approach could keep this market alive, and pump.fun has the ability to try it. It holds approximately $2.84 billion in on-chain assets, and the entire process is in its hands. In the long run, a market that continuously produces big winners will make it more money than a market that continuously exhausts buyers. It chose to harvest because harvesting earns more today, but this approach cannot last.

Even so, it may not be enough. Such changes would raise costs for token issuers. But the supply side will always have stronger tools than the demand side, and this will not change because of that.

What a token launch without a bonding curve could look like

All of the above approaches are still built on a bonding curve. One step further would be to try removing the bonding curve.

In this model, everyone pays the same price. Without a bonding curve, there is no advantage to being the first transaction in a block, and sniper front-running loses its edge. Creators would set a minimum and maximum fundraising amount. If the minimum is not reached within a specified time, such as 24 or 48 hours, the smart contract automatically refunds everyone, so a failed launch does not cost buyers their funds. Once the fundraising cap is reached, the contract creates a liquidity pool on its own and burns or locks the LP tokens, so creators never touch the raised funds. Each wallet's contribution has a hard cap, such as 1% of the total raise, so a single whale or bot cannot take half the token supply.

The platform only earns revenue when the raise is filled. The platform makes money from successfully launched tokens; even if 1,000 launches fail to fill, it earns nothing.

The platform shifts from charging a fee on every transaction to charging after a raise is filled

The fee trigger for launchpads shifts from every transaction to every filled raise.

This would eliminate behaviors such as bundling and sniping at the start of the bonding curve, sniper front-running, and draining the raised liquidity, and it would also remove the incentive to launch 50 tokens in a row and still profit even if none of them fill.

This model has loopholes worth spelling out. Per-wallet limits only work when wallets are hard to fake, and wallets with history are cheap now. Without some kind of anti-Sybil mechanism to stop the same person from using multiple wallets, bundlers only need to spread buy orders across 100 wallets. A fixed price also does not stop someone from selling on day one. All it can do is make the starting line fairer.

What happens next

Meme coins in their current form are finished. We believe they can only come back once the token issuance model itself changes—for example, by adopting the approach above, where the platform profits when a coin successfully launches.

As long as even one launchpad can profit from every coin that ultimately goes to zero, the meme coin trading frontline will remain dormant, and holding these coins will still be a losing trade. New things will emerge, but they won't be meme coins, and they won't come soon.

Nothing in this article constitutes investment advice. We do not hold $PUMP or any other tokens mentioned in the text. Data is as of November 9, 2026, and all figures are approximate. Please retrieve the underlying data to verify before use.

Welcome to join the official BlockBeats community:

Telegram Subscription Group: https://t.me/theblockbeats

Telegram Discussion Group: https://t.me/BlockBeats_App

Official Twitter Account: https://twitter.com/BlockBeatsAsia