Routing Node

A routing node is a Lightning Network node that forwards payments on behalf of others, earning fees for the liquidity and connectivity it provides. Because no participant opens a direct channel with everyone they transact with, routing nodes form the connective tissue of the Lightning Network: a payment from a coffee buyer to a merchant may hop across two, three, or more intermediary nodes, each one briefly committing its own funds to move the payment along.

How routing actually works

Lightning uses source routing: the sender computes the entire path. The sending node consults the public channel graph — gossiped between nodes under the BOLT 7 specification — to find a route whose channels plausibly have sufficient capacity on the correct side of each hop. It then wraps per-hop instructions in a Sphinx onion, the same layered-encryption idea Tor uses. Each routing node peels only its own layer, learning just enough to do its job: verify the fee and timelock terms, then forward a matching HTLC to the next hop. No intermediary sees the full route, the ultimate sender, or the final recipient — a node in the middle cannot even tell whether the hop before it was the origin or just another router. The HTLC construction makes the whole chain atomic: either the recipient reveals the payment preimage and every hop settles, or the timelocks expire and every hop rolls back. A routing node risks lockup time, not loss of principal, when a payment fails mid-route.

Fees and the liquidity business

A routing node advertises two fee components per channel: a base fee, a flat charge per forward, and a fee rate, proportional to the amount. Fees are set per-channel and per-direction, and competitive routing is a genuinely thin-margin business — revenue comes from volume across well-placed channels, not from any single forward. Profitability depends on three things: balanced liquidity (a channel exhausted in one direction can only route one way), good uptime (offline nodes get pruned from pathfinding), and placement (channels that bridge otherwise poorly connected regions of the graph see the most flow). Managing all this is a continuous discipline of rebalancing, fee tuning, and peer selection — the same craft covered under inbound liquidity.

Running one at home

Routing is one of the few ways a home node runner can put capital to work while strengthening the network's decentralization: every independent routing node is one more path that does not run through a large custodial hub. Be honest about expectations, though. A modest home router with a few million satoshis deployed will usually earn modest fees; the real returns are operational experience and network resilience. The hard requirements are uptime and monitoring, and the useful metrics are forwarding volume, failure rates, and how quickly your channels drain to one side — numbers every mainstream node package now surfaces. A routing node holds live channel states, and a node that goes dark for long stretches risks missing a channel breach — a counterparty publishing an outdated state. That is why prudent operators delegate breach monitoring to a watchtower, keep their node on reliable power, and treat channel backups with the same seriousness as any other part of self-custody.

For the sovereign Bitcoiner, a routing node is infrastructure in the truest sense: unglamorous, always-on, and quietly essential. You are not just using the payment network — you are a piece of it, and the fees you earn are the network paying you for keeping one more corridor open that no one can close. Start small, learn the rhythms of your channels for a few months, and scale capital only once your uptime and backups have earned it.