Bitcoin Ordinals and Inscriptions: Impact on Fees and Blocks

How Bitcoin Ordinals Work

Bitcoin Ordinals assign a unique serial number to every individual satoshi based on the order it was mined. The numbering scheme, called ordinal theory, starts at 0 for the first satoshi in the genesis block and increments sequentially as new satoshis enter circulation through coinbase transactions. When satoshis move between addresses, their ordinal numbers transfer using a first-in-first-out (FIFO) rule: input satoshis map to output slots in order, preserving trackability across the entire transaction graph.

The protocol was created by Casey Rodarmor, a former Bitcoin Core contributor, and launched on Bitcoin mainnet on January 21, 2023. Rodarmor inscribed the very first inscription (Inscription #0, a pixel art skull) on December 14, 2022, roughly five weeks before the public launch. The protocol relies on two prior Bitcoin upgrades: SegWit (August 2017), which introduced the witness data discount, and Taproot (November 2021), which removed size caps on witness data within a single input via BIP-342 (Tapscript).

Understanding Inscriptions

Inscriptions are the mechanism for attaching arbitrary data to individual satoshis. The data is embedded in the witness field of a Taproot (P2TR) transaction, wrapped inside an OP_FALSE OP_IF ... OP_ENDIF envelope that never executes. Within the OP_IF block, OP_PUSH operations encode three things: the marker "ord," a MIME content type, and the actual payload (image, text, audio, video, HTML, or any other file type).

Because inscription data lives in witness space, it benefits from the SegWit discount: witness bytes cost 1 weight unit each, compared to 4 weight units per byte for non-witness data. This means a 400 KB image stored as an inscription consumes only 400,000 weight units rather than the 1,600,000 it would cost in standard transaction space. A single inscription can theoretically fill nearly the entire 4 million weight unit block limit, though in practice most inscriptions are much smaller.

Key Ordinals Milestones

The timeline below tracks the major events in the Ordinals ecosystem from its technical foundations through its maturation as a protocol.

Impact on Block Space and Fees

Ordinals inscriptions fundamentally changed the Bitcoin fee market. Within three weeks of launch, inscriptions consumed roughly 50% of block space. Average block size rose from 1.18 MB to 1.74 MB (a 47% increase) and peaked at 2.29 MB in March 2024. The stripped block size (excluding witness data) showed only modest changes, confirming that inscriptions primarily occupy the witness discount portion of blocks.

The fee impact has been dramatic during peak inscription activity. Use the Bitcoin fee estimator to check current fee rates, and review the data below for historical context.

Cumulatively, inscriptions generated approximately 6,901 BTC (~$405 million) in transaction fees by September 2024. For miners, this revenue is significant: it demonstrates that demand for block space beyond simple monetary transfers can supplement the declining block subsidy over time.

UTXO Set and Mempool Effects

Inscriptions themselves sit in witness data and do not directly inflate the UTXO set. However, BRC-20 tokens built on top of inscriptions create many small UTXOs because each token operation (deploy, mint, transfer) produces a new UTXO. This distinction matters: a bloated UTXO set increases the memory and storage requirements for every full node on the network, regardless of whether the node operator cares about token activity.

The mempool impact has been equally visible. During the May 2023 BRC-20 frenzy, the mempool swelled to over 500,000 unconfirmed transactions. Low-fee financial transactions were effectively priced out during peak periods, with simple transfers sometimes waiting days for confirmation. This congestion renewed interest in layer-2 solutions like the Lightning Network and Spark for everyday Bitcoin payments.

BRC-20 Tokens and the Runes Protocol

BRC-20 is a fungible token standard created in March 2023 by pseudonymous developer "Domo." It uses Ordinals inscriptions containing JSON data to deploy, mint, and transfer tokens. Unlike Ethereum's ERC-20, BRC-20 does not use smart contracts: each operation is a standalone inscription that must be separately indexed by off-chain services. The first BRC-20 token, ORDI, reached a market capitalization of roughly $1 billion by December 2023, though total BRC-20 market cap has since declined to approximately $152 million.

BRC-20's primary technical weakness is UTXO bloat. Casey Rodarmor addressed this directly by creating the Runes protocol, which launched at Bitcoin block 840,000 on April 20, 2024 (the fourth halving). Runes stores token data in OP_RETURN outputs (80 bytes) and attaches token balances directly to UTXOs. Spending a Rune consumes the UTXO rather than creating junk outputs, making it compatible with the Lightning Network and less harmful to node operators. In its first 24 hours, 4,923 runes were etched across 801,124 transactions with 68,548 holders.

The Debate: Spam vs. Legitimate Use

Ordinals have sparked one of the most contentious debates in Bitcoin since the block size wars. The core question: should Bitcoin blocks carry arbitrary data, or should block space be reserved for monetary transactions?

Those who view inscriptions as legitimate use argue:

• Bitcoin is permissionless; censoring transaction types violates its foundational design
• Arbitrary data has existed in Bitcoin since the genesis block (Satoshi's embedded newspaper headline)
• Fee revenue from inscriptions strengthens the long-term mining security budget as the block subsidy halves every four years
• Data can always be encoded in alternative ways; blocking one method pushes activity to less efficient encoding techniques

Those who view inscriptions as harmful argue:

• Network congestion prices out financial transactions, especially for users in developing economies
• Blockchain growth threatens decentralization by making full nodes more expensive to run
• Inscriptions exploit the SegWit witness discount for a purpose it was not designed for
• Storing arbitrary content on an immutable ledger creates potential legal liability for node operators

The debate reached the Bitcoin Core codebase directly. In late 2023, developer Luke Dashjr registered CVE-2023-50428, calling inscriptions an exploit that "disguises data as code" to bypass the datacarriersize limit. He submitted a Bitcoin Core pull request to apply stricter data-size limits. In January 2024, Bitcoin Core maintainer Ava Chow closed the PR, noting it had "no hope of reaching a conclusion acceptable to everyone." Dashjr's alternative client, Bitcoin Knots, does implement inscription filtering.

The policy question resurfaced in mid-2025 when Bitcoin Core v30 raised the OP_RETURN data limit. Developer Gloria Zhao argued that OP_RETURN is the "least harmful" method for data storage and that restricting it only pushes data into witness space. The change was merged, but the decision widened the rift: Bitcoin Knots's node share rose above 20% as some operators rejected the new policy.

Practical Implications for Bitcoin Users

For developers and users who primarily use Bitcoin for payments and financial transactions, Ordinals activity has a few concrete implications:

• Fee volatility increases during inscription surges; monitoring the fee estimator before broadcasting transactions becomes more important
• Replace-by-fee (RBF) and child-pays-for-parent (CPFP) fee-bumping strategies are essential when the mempool is congested
• Layer-2 solutions like Lightning and Spark become more valuable during high-fee periods since they settle transactions off-chain
• Full node operators should plan for faster disk usage growth (approximately 6-7 GB per month during periods of high inscription activity)
• UTXO consolidation during low-fee periods is more valuable than ever, given the unpredictability of fee spikes

For a broader analysis of how the fee market functions, see our research on Bitcoin fee market dynamics.

Frequently Asked Questions

What are Bitcoin Ordinals?

Bitcoin Ordinals is a numbering scheme that assigns a unique serial number to every satoshi based on mining order. Combined with inscriptions (arbitrary data embedded in Taproot witness fields), it enables NFT-like assets and tokens to exist directly on the Bitcoin blockchain without requiring a sidechain or separate token layer. The protocol was created by Casey Rodarmor and launched in January 2023.

How do Bitcoin inscriptions affect transaction fees?

Inscriptions compete for the same block space as financial transactions, which drives up transaction fees during periods of high inscription activity. In May 2023, average fees rose to ~$19 due to BRC-20 minting. In December 2023, average fees hit $38. At the April 2024 Runes launch, average fees reached a record $128. During quiet periods, inscription impact on fees is minimal.

What is the difference between BRC-20 tokens and Runes?

BRC-20 tokens use JSON inscriptions in witness data and create a new UTXO for every operation, bloating the UTXO set. Runes store token data in OP_RETURN outputs and attach balances directly to UTXOs, so spending a token consumes (rather than creates) UTXOs. Runes are also compatible with Lightning Network payment channels.

Can Ordinals inscriptions be removed from Bitcoin?

No. Once an inscription is confirmed in a block, it is permanently part of the blockchain. At the protocol level, filtering inscriptions from new blocks would require a consensus change (a soft fork) that the majority of miners and nodes would need to adopt. Attempts to introduce such filtering via Bitcoin Core PRs have been rejected. Individual node operators can run alternative clients like Bitcoin Knots that filter inscription transactions from their mempool, but this does not remove existing inscriptions or prevent miners from including them.

How many Bitcoin inscriptions exist?

Over 90 million inscriptions had been created as of late 2024, with BRC-20 operations accounting for the majority. The exact current count depends on counting methodology: some trackers include all BRC-20 events (deploy, mint, transfer) as separate inscriptions, while others use narrower definitions. Peak daily activity was ~505,000 inscriptions on November 12, 2023. Daily volumes have since declined significantly from those peaks.

Do Ordinals threaten Bitcoin decentralization?

This is debated. Inscription data increases blockchain growth rate (roughly 6-7 GB per month during active periods), which raises storage requirements for full nodes. However, because inscriptions sit in witness data, they are eligible for pruning by nodes that do not need full historical data. The more pressing centralization concern is the UTXO set bloat caused specifically by BRC-20 tokens, which the Runes protocol was designed to address.

Why do inscriptions use the SegWit witness discount?

SegWit charges witness data at 1 weight unit per byte versus 4 for non-witness data, making witness storage 75% cheaper. Taproot (BIP-342) removed the per-input witness size cap, allowing a single input to carry up to nearly 4 MB of witness data. Inscriptions exploit both features: SegWit's cost discount and Taproot's size allowance. Critics argue this was not the intended purpose of either upgrade; supporters counter that the protocol is being used exactly as designed.

This guide is for informational purposes only and does not constitute financial advice. Data is approximate and based on publicly available information. Inscription counts, fee figures, and market data change frequently. Always verify current figures before making decisions.