Smarter Liquidity: Automated Liquidity Provision on DEXs

By DeHub Team — 2023-10-03

In Q4 2023 we implemented automated liquidity provision strategies across decentralised exchanges.

Where liquidity management used to live

Inside the token contract. That is what tax-based tokenomics were — a percentage of each transaction diverted automatically, including into liquidity, executed in the transfer path on every transaction.

Then ours malfunctioned and froze the token and the pool together, and we moved to plain standard ERC-20 with nothing in the transfer path at all.

That was the correct fix and it left a gap. Automatic liquidity growth was a genuine benefit of the old design, and removing the mechanism did not remove the need.

What replaced it

Rules-based provision operating outside the contract.

The distinction is the whole point. Contract-level automation is immutable, runs on every transaction, and cannot be corrected when conditions change. External automation follows a defined strategy, can be adjusted as markets move, and — critically — cannot freeze anyone's tokens if it misbehaves. The failure mode of a bad liquidity strategy is a bad liquidity position. The failure mode of bad transfer-path logic was 25,000 holders locked out of their own assets.

Same objective. Radically different blast radius.

Why automate rather than manage by hand

Discretionary liquidity management has two problems. It is slow, because it depends on someone being awake and paying attention. And it is unaccountable, because every decision can be explained after the fact.

A defined strategy fixes both. It responds at machine speed, and because it was published as policy in the same quarter, its behaviour can be checked against the chain by anyone who wants to.

That pairing was deliberate. Automation without a stated policy is just a faster opaque process.

What it enabled

It is the reason leaving Gate.io voluntarily was possible in the same quarter without stranding holders.

Delisting only works if the on-chain market is genuinely usable — sufficient depth, tolerable slippage, reliable pricing. Those are the properties liquidity provision produces. Announcing the delisting first and hoping depth appeared would have been the reverse of responsible.

Where it went next

The same infrastructure carried DHB's expansion across chains: Ethereum mainnet in Q1 2024, sDHB on Base that summer, and eventually the decision to discontinue Ethereum mainnet support and concentrate where users actually were.

Each of those is a liquidity decision as much as a technical one, and each was possible because provision had become a process rather than a property of the token.

What automation does not fix

Depth still has to be funded, and no strategy creates liquidity that is not there.

Automation decides how capital is deployed, rebalanced and defended. It does not decide how much capital exists, and a sophisticated strategy running on insufficient depth produces an efficiently managed thin pool.

That is worth saying because "automated liquidity" is frequently presented as though the mechanism were the substance. It is not. The substance is treasury allocation, which is a deliberate decision made by people, disclosed under the DEX policy, and visible on-chain.

The automation's contribution is that the decision stops being re-made informally every week by whoever happens to be watching the market.