Bitcoin’s mempool is quiet – way too quiet.
In April, the largest bitcoin mining pool, Foundry, mined a block that included just 7 transactions. It was the emptiest non-empty block in more than two years. (An empty block occurs when a mining pool deliberately mines a block with no transactions).
The later half of 2024 and the first half of 2025 have been a stark departure from the go-go days of 2023 and the first half of 2024. Back then, Bitcoin’s mempool was frothy with activity thanks to metaprotocols like inscriptions, runes, and BRC-20 tokens, so much so that bitcoin miners hauled in record USD-denominated transaction fee flows on April 20, 2024.
It’s unusual for transaction fee activity to be this low in a bull market. According to CoinMetrics data, in 2021, bitcoin miners earned $2.8 million per day in fees on average. So far in 2025, that average is just $0.55 million. So what exactly is going on?
To explain Bitcoin’s fee drought, we have to look at three different trends: technical advancements, Bitcoin’s financialization, and metaprotocols.
Segwit adoption and Bitcoin’s financialization are primary culprits
To start with an obvious and easy explanation, the majority of the Bitcoin network is now using a transaction standard that is cheaper and more data efficient.
SegWit was released in 2016, but it didn’t become a ubiquitous transaction standard until relatively recently. SegWit transactions are smaller and more data efficient than those that came before it, and as of 2023, over 90% of all active wallets have adopted SegWit.
It would stand to reason that, as SegWit has become increasingly adopted and transactions have become more efficient, we’d see this reflected in transaction fee revenue. But this only tells part of the story.
For the rest, we have to look at institutional adoption.
In a June 19 newsletter, Glassnode notes that Bitcoin’s integration into legacy finance, spurred forth by the January 2024 bitcoin ETF approvals, has recently manifested in two trends.
For the first, the value of average bitcoin transactions are increasing. The post notes that in June, 89% of transactions were settling value at or greater than $100,000, versus 66% in November 2022.
“This trend reinforces the view that high-value participants are becoming increasingly dominant within on-chain activity,” the newsletter notes.
Complementing this trend, off-chain transaction volumes are also heating up, Glassnode argues. Since January 2024, off-chain bitcoin trading volume, which includes futures, spot for ETFs, and options, has been 7-16 times greater than on-chain volume. With bitcoin ETFs and the like, investors and traders have more tradfi options to trade bitcoin than ever before. As a result, a lot of trading volume is occurring off-chain, impacting demand for block space.
“The growth of spot and derivative volumes highlights a shift in activity, with an ever growing share of volume migrating from the Bitcoin base layer towards off-chain venues. When comparing the volume traded off-chain (spot, futures and options), to value settled on the network, we note off-chain volume has regularly been 7-16x larger than the on-chain volume,” Glassnode notes.
Ordinals trading, non-monetary use cases are down but not out
The wild thing is, if it weren’t for what some call “spam” – i.e., non-monetary use cases for Bitcoin – transactions would be even lower.
According to a Dune dashboard by user jinmaa that tracks transaction fees from metaprotocol transactions (i.e., ordinals, runes, etc), non-monetary transactions made up roughly 10% of all bitcoin transaction fees in June.
One out of every ten satoshi that miners earn is coming from ordinal and ordinal-adjacent degens. That’s a pretty good portion, but it’s a far cry from 2023’s numbers. According to on-chain analyst Mononaut, that year, 36% of all transaction fees came from non-monetary transactions – mostly BRC-20 tokens. In the first half of 2024, when ordinal trading activity was at its peak for the year, metaprotocols accounted for 14% of transaction fee volume, according to Galaxy Digital.
Demand for ordinals and similar ilk have died down in 2025, and the transaction fee volumes for these non monetary use cases in June are actually higher than we’ve seen in previous months thanks to a few new metaprotocols.
Metaprotocols are still popular but they aren’t driving fees
For those not fully immersed in the latest shitcoin shenanigans, there’s still a large amount of metaprotocol activity on Bitcoin. Cumulative metaprotocol activity accounts for about 40% of all transactions on Bitcoin even though the secondary market for these fungible tokens is down everywhere you look.
The current landscape of metaprotocol activity is a mix of Runes, BRC-20, and Alkanes. Alkanes is the new kid on the block – a kind of Runes 2.0 but with an enhanced smart contract component for indexing the metaprotocol.
During a given week, there will typically be a handful of modest fee events driven by these metaprotocols, but in Q3 of 2025 there have only been five weeks where net additional fee revenue exceeded $250k (four of those weeks were driven by Alkanes mints).
These metaprotocols tend to create bursts in on-chain activity and briefly elevate fees during token mints. For example, here are mints for an Alkanes token METHANE around the end of March:
For about 50 blocks during the METHANE mint, we saw 3k – 4k transactions per block (close to the maximum number of transactions that can fit in a block) and a brief ~20x increase in fee rates from 1sat/vbyte to 20-30 sats/vbyte. This resulted in an additional 0.1 BTC to 0.2 BTC in additional rewards per block (about 10x higher than average for the period).
Another example is the Alkanes clock-in game. There is an Alkanes contract that requires users to broadcast a transaction in a specific block, like they are “clocking in for work.” There’s no token or financial gain associated with the game, but it has created an interesting and highly predictable burst in on-chain activity every day for the past 30 days. This causes intra-block fees to go from 1sat/vbyte to 20sat/vbyte once every 144 blocks.
There’s little consensus or conclusions on what to do to stop the slow bleed. Some developers advocate for additional expressivity on Bitcoin, while others believe time alone can improve Bitcoin’s fee rate issue. Maybe we just need more adoption?
As for now, Bitcoin’s mempool is a desert currently with little competition for blockspace.
Enjoyed the read? Check out Jameson Lopp’s ‘Goldiblocks’ presentation from OPNEXT, the Bitcoin scaling conference.
