Lightning Nodes Passed 20,000. Still Nobody Uses Them.

The public Lightning Network now has more than 20,000 nodes and roughly 60,000 open channels. Public capacity sits above 5,000 BTC. These numbers are regularly cited as evidence of Lightning's growing maturity, and on the surface they are impressive. The network is larger, more redundant, and better connected than it was three years ago.

Payment volume has not followed.

Estimating Lightning payment volume is deliberately hard because much of the traffic is private. But the available signal sources all point the same direction. River Financial's published volume data, public probes of large routing nodes, Strike's disclosed payment flow, and merchant acquirer reports from companies like OpenNode and IBEX all suggest that real economic Lightning volume in 2026 is roughly the same as it was in 2023, possibly slightly lower, and is heavily concentrated in a small number of use cases that have not grown into general retail payment demand.

The infrastructure kept building. The demand did not.

The Numbers Nobody Wants to Line Up

Consider what the advocacy case for Lightning was in 2021 and 2022. Block space would become scarce. Fees would rise. Lightning would absorb retail transactions at zero-cost, instant settlement. Merchants would onboard. A wave of Lightning-native applications would emerge. Circular payments would train a generation of users on non-custodial wallets.

Each of these predictions has been partially or fully falsified by subsequent data.

Block space did not remain scarce. The mempool cleared and has mostly stayed clear since mid-2024. The base layer can handle current retail demand at low fees. The primary economic case for retail Lightning adoption, which was base-layer fee pressure, evaporated.

Merchant Lightning adoption stalled. A small number of category-defining merchants, including a few major remittance corridors and some high-profile vendors, integrated Lightning and use it. The long tail did not follow. Retail point-of-sale Lightning is still essentially a novelty outside of a handful of Bitcoin-forward cities. Total merchant payment volume through Lightning in 2026 is smaller than the volume processed by Square's contactless payments in a single metro market.

Lightning-native applications did not emerge at scale. Podcasting 2.0 and value-for-value streaming built a devoted community in the tens of thousands, not the millions. Gaming integrations saw brief spikes tied to specific launches and then settled into low baseline activity. Content monetization through Nostr zaps grew, but Nostr itself remains a small network measured against mainstream social platforms.

The use case that did grow is not the one anyone was advertising. Exchange-to-exchange transfers, custodial wallet rebalancing, and cross-border payment rails for fintech companies account for most of the measurable Lightning volume. These are not peer-to-peer payments. They are business-to-business settlements using Lightning as rails, with end users never touching a channel.

The Custodial Reality

Most Lightning payments that happen today happen through custodial wallets. Strike, Cash App, Wallet of Satoshi, Phoenix, Muun's custodial variant, and various neo-bank integrations process the vast majority of user-facing Lightning volume. The non-custodial Lightning experience, the version that matched the original decentralization story, remains technically demanding and comparatively rare.

This is not a neutral observation. The Lightning Network was sold as a tool for permissionless, self-sovereign payments at scale. In practice, it became a backend for custodial wallets that abstract away the protocol. For the typical user who pays with Cash App and receives Lightning payments routed through Cash App's internal network, the experience is indistinguishable from a traditional remittance service with Bitcoin used as rails. The user does not hold the keys. The user does not run a node. The user does not open a channel.

If custodial wallets and business settlement traffic account for most Lightning volume, then most Lightning volume is simply a cheaper, faster Bitcoin-denominated equivalent of the existing payments stack. It is a meaningful technical achievement. It is also a smaller thing than what was promised.

Why the Infrastructure Keeps Growing

A reasonable question: if demand has flatlined, why is node count still rising?

Three reasons. First, running a Lightning node became a hobby in the same way that running a home server became a hobby. The population of enthusiasts is small but persistent, and each enthusiast contributes nodes, channels, and capacity regardless of whether their channels see paying traffic. Most channels opened by hobbyist nodes are low-volume or effectively inactive. They count in the capacity totals. They do not move money.

Second, Lightning Service Providers like Blockstream Greenlight and Breez are spinning up nodes programmatically to support non-custodial wallet users. A single LSP can operate thousands of nodes on behalf of users who never realize they have one. This inflates the node count without corresponding payment growth, because LSP nodes primarily serve their own user bases and do not route heavy external traffic.

Third, enterprise integrations open channels for capacity guarantees rather than current demand. An exchange or payment processor might open 500 BTC in channel capacity to ensure they can handle future volume spikes. That capacity shows up in the totals. Until the volume materializes, it sits idle.

The result is a network that looks larger and healthier by infrastructure metrics while producing roughly the same amount of economic activity it produced in 2023. The disconnect is real and growing.

What This Means For the Payment Thesis

The Lightning payment thesis is not dead. It is narrower than advertised. Lightning works for specific use cases, mostly involving custodial services and cross-border corridor payments, and it works well. It has not become the retail payment layer for Bitcoin, because retail payment demand in Bitcoin has not grown to the size that required a new layer. Users who want to spend Bitcoin are still a small minority of Bitcoin holders. Most holders treat it as an investment asset and a savings instrument.

If the store-of-value thesis has won the argument inside Bitcoin, the payment layer built for the medium-of-exchange thesis will be underutilized by definition. That is what 20,000 nodes and flat payment volume look like.

The honest conclusion is that Lightning is a solution looking for the demand it was supposed to create. The network is ready. The users are not coming. Whatever the next wave of Lightning activity looks like, it will come from a use case that has not yet materialized, because every use case that was supposed to drive adoption has already been tried.

Either a new one appears, or the 20,000 nodes keep listening to a network nobody is calling.