SegWit — short for Segregated Witness — was a soft-fork upgrade to Bitcoin that activated in August 2017 after a multi-year debate that divided the Bitcoin community. At its core, SegWit changed how Bitcoin transactions store signature data. Before SegWit, a transaction’s signature (the “witness data” that proves the sender authorised the transaction) was stored inline with the rest of the transaction and counted toward the 1 MB block size limit. SegWit separated the witness data into its own section, stopped counting it against the original 1 MB limit, and introduced a new unit — the block weight — that allowed blocks up to 4 MB of total data while maintaining backwards compatibility with old nodes.
The immediate effects were a roughly 1.7x to 2x increase in effective block capacity (depending on how many transactions used SegWit), a fix for a long-standing malleability bug that prevented certain kinds of second-layer protocols from being built, and lower fees for users who adopted SegWit addresses (which started with 3... in P2SH-wrapped form or bc1... in native bech32 form). The fix for transaction malleability was what unblocked development of the Lightning Network, which relies on chained transactions that would have been too fragile under the pre-SegWit rules.
The Political Backstory
SegWit was the result of one of the most contentious debates in Bitcoin’s history, known collectively as the “block size wars”. The core disagreement was about how to scale Bitcoin: one faction (sometimes called “big blockers”) wanted to simply increase the block size limit from 1 MB to 2 MB or 8 MB, which would be a straightforward hard fork. Another faction (roughly, the Bitcoin Core developers and the majority of node operators) was concerned that larger blocks would make running a full node more expensive, centralising the network around wealthy operators, and wanted a more conservative upgrade that preserved decentralisation.
SegWit was the Core developers’ answer: a soft-fork upgrade that increased effective capacity without changing the nominal block size limit, and that opened up space for layer-2 scaling (Lightning) rather than trying to push all throughput onto the base chain. Big blockers rejected this as insufficient and eventually forked off their own chain — Bitcoin Cash — in August 2017, a few weeks before SegWit activated on the main chain. The split was acrimonious, and the two chains have continued on their separate paths ever since, with Bitcoin (BTC) taking the SegWit path and Bitcoin Cash (BCH) taking the larger-blocks path. Bitcoin Cash later had its own internal split, producing Bitcoin SV.
The outcome, in retrospect, validated the Core approach. Bitcoin has grown enormously in value since 2017 while maintaining the ability to run full nodes on consumer hardware. Lightning Network has become a real, if still-niche, scaling layer on top of SegWit. Bitcoin Cash has remained a much smaller chain with a fraction of the activity, and Bitcoin SV less than that. The specific prediction that bigger blocks were the only way to scale turned out to be wrong — SegWit plus Lightning plus Taproot plus the L2 explorations currently underway have together produced a scaling story that big blocks alone would not have enabled.
The Adoption Curve
SegWit was available immediately after activation but took years to reach widespread adoption. Wallets had to update their code to support the new address formats. Exchanges had to add support for sending to SegWit addresses. Users had to opt in to SegWit by choosing to generate SegWit addresses (often labeled “bech32” in wallet software). The holdouts dragged on: Coinbase infamously took until early 2020 to fully support sending to SegWit addresses, which meant Bitcoin users were still paying higher-than-necessary fees for years after the upgrade because a non-trivial fraction of the network had not yet adopted it.
As of 2026, SegWit adoption is around 90 percent of transactions, and the remaining non-SegWit usage is mostly legacy addresses and older wallet software that has not been updated. Native SegWit (bech32, bc1... addresses) is the default in most modern wallets, and Taproot (a later upgrade) builds on the SegWit framework to add more capabilities.
What SegWit Did and Didn’t Fix
SegWit increased Bitcoin’s effective throughput from about 3 transactions per second to about 7 transactions per second in typical conditions. This is nowhere near enough for Bitcoin to be a global payments network on its own — Visa processes tens of thousands of transactions per second during peak hours, and even much more modest payment networks handle volumes Bitcoin cannot match at its base layer. The point of SegWit was never to solve scalability alone; it was to create the conditions under which layer-2 protocols could be built on top of a stable, secure base layer. That plan has partially played out — Lightning exists and works — but the “Bitcoin replaces payments” narrative has not materialised at any meaningful scale, and SegWit is just one component in a scaling story that is still being written.