Limit Order Strategies for DeFi Traders: Scaling In, Scaling Out, and Beyond

CoW Swap

@CoWSwap · Sep 25

4,173

TL;DR

Scaling in means building a position in chunks at prices you pick in advance, rather than going all-in at one price and hoping you timed it right.
Scaling out is the reverse: you take profit in stages, so you don't have to guess the exact top.
Limit orders do both jobs well because each "rung" of your ladder only fills at your price or better.
CoW Swap limit orders add some useful mechanics for ladders. You can create and cancel them without paying gas. Several orders can draw on the same balance. Orders are partially fillable by default. And if the market moves past your limit, you receive the execution-time price, not just your limit price.
The rule that matters most: plan the whole ladder as one trade before placing the first order. Decide your maximum size, your exit if you're wrong, and what you'll do if only half the rungs fill.

The problem with going all-in

You've done your research, you like ETH at around $3,000, and you hit "swap" on your whole budget. An hour later it's at $2,850. You weren't wrong about ETH. You were wrong about the timing, and the timing mistake now applies to 100% of your money.

This is the basic weakness of the single-entry trade: it's one binary bet on one moment. Traders in traditional markets worked this out a long time ago, which is why "scaling" is a standard part of position management. Investopedia defines scaling as buying multiple orders of the same asset at incrementally different prices, rather than the full amount at once. The idea is simple. Instead of trying to hit the exact price with one order, you spread your entry across a range of prices. Some orders will fill at better prices than others, and your average ends up somewhere sensible.

Limit orders are the natural tool for this. Here's why, and how to use them.

Limit orders in 60 seconds

A limit order is an instruction to buy or sell at a specified price or better:

A buy limit sets the maximum you'll pay.
A sell limit sets the minimum you'll accept.

If the market never reaches your price, nothing happens. That's the core trade-off: you get certainty of price and give up certainty of execution.

What's different on CoW Swap

On most exchanges, a limit order sits in an order book waiting to be matched. On CoW Swap, a limit order is a signed_intent_: a message saying "I'll trade X for at least Y, before this expiry date." Solvers then compete in batch auctions to settle it. That design changes some practical details, and several of them matter a lot for laddering.

Gasless order management. You can create, modify, and cancel limit orders without paying gas. The standard cancellation happens off-chain and is free. There is an on-chain cancellation option if you want one, but that one costs gas. One caveat: the first time you trade a token, you may need to approve it, and that approval is an on-chain transaction.

One balance, many orders. CoW Protocol lets you use the same balance for several open orders at once. It fills them as long as your wallet still holds the funds. We'll come back to why this is useful.

Partial fills by default. CoW Swap limit orders are partially fillable by default, so a large order can fill gradually as liquidity becomes available. You can switch to fill-or-kill in the settings if you want all or nothing.

Surplus. If the market moves beyond your limit before your order settles, you receive the execution-time price rather than being settled at exactly your limit. Price improvements that solvers find, such as a direct match with another trader (a Coincidence of Wants), also go back to the trader. A protocol fee applies to surplus on out-of-market limit orders; the fees page on docs.cow.fi has the details.

Built-in MEV protection. You sign an intent rather than broadcasting a raw swap, and orders in the same batch settle at uniform clearing prices. This structure removes the ordering games that sandwich bots depend on.

One mechanic to know before you start. CoW Protocol doesn't charge a separate fee for placing a limit order. Instead, network fees are covered out of the execution price, so the protocol waits until the market is good enough to satisfy your limit price and pay those fees. In practice, your order may not fill the instant the chart touches your number. The "Executes at" column in your open orders panel shows the price your order actually needs to reach, based on current network conditions.

Before you ladder: plan it as one trade

This is the step most people skip, and the one that separates a strategy from a series of impulse buys. A ladder is one position split into pieces, not several small independent bets. Before placing any order, write down four things:

Maximum size. If every rung fills, how much will you hold? That total must fit your risk budget.
Invalidation. At what price is your thesis wrong for the whole position?
The half-fill scenario. If only the top rung fills and price reverses, what do you do? Usually the answer is "accept a smaller position." It should never be "chase the missing size at worse prices."
Cleanup. When do unfilled rungs get cancelled? Set an expiry and stick to it.

If you can't answer those four questions, you aren't laddering yet. You're guessing more than once.

Strategy 1: Scaling in with a buy ladder

The idea: you have a budget and a price zone you like. Rather than choosing one price in that zone, you place several buy limit orders across it.

Worked example

All figures are hypothetical and for illustration only.

ETH is trading at $3,000. You have 3,000 USDC to deploy, and you'd be happy buying anywhere between $2,700 and $2,900. You place three buy limits:

Why Your CoW Swap Limit Order Hasn't Filled Yet (and What to Do)

If all three rungs fill, your average entry is about $2,798: 3,000 USDC divided by 1.0723 ETH.

That's a little below the midpoint of your range. This happens because equal dollar amounts buy more ETH at lower prices, so the cheaper rungs carry more weight in your average. It's a small effect, but it's the maths working in your favor.

If only rung 1 fills and ETH rallies away, you hold about 0.34 ETH, bought at a price you chose in advance. That's still a good outcome. You didn't miss the move entirely, and you didn't overpay out of fear of missing out.

Choosing the shape of your ladder

Equal rungs (the example above) are simple and a good default.
Heavier at the bottom (for example 20% / 30% / 50%) puts more capital to work at the lowest prices. The trade-off is that you may never get the size you want if price doesn't dip that far.
Heavier at the top (for example 50% / 30% / 20%) makes sure most of your position fills if price only dips a little. The trade-off is a higher average entry.

Most experienced traders stick to three or four rungs. More than that tends to turn into micromanagement.

Scaling in vs. averaging down

These two can look the same on a chart, but they are opposites. Scaling in is a plan made before the trade: fixed levels, a fixed maximum size, and a fixed point where you admit you were wrong. Averaging down is usually a reaction made during the trade: buying more because you're losing, with no ceiling. The orders may look identical. The difference is whether the plan existed first.

Strategy 2: Scaling out with a sell ladder

The idea: you hold a position that's in profit. Rather than guessing the top, you sell in stages.

Scaling out solves a very human problem. Sell everything too early and you watch the rest of the rally from the sidelines. Hold everything too long and you watch your profits disappear. A sell ladder removes the pressure to be right about one exact moment.

Worked example

Again, hypothetical. You hold 3 ETH, bought at around $2,800. You place:

Sell 1 ETH at $3,200
Sell 1 ETH at $3,500
Keep 1 ETH as a "runner," with no set target

If both sell orders fill, you've taken $6,700 off the table. Your entire original cost was about $8,400, so that's a large share of your capital recovered. You still have exposure if the rally continues.

Why set the orders in advance

The best time to decide your exit prices is when you're calm, not when the chart is moving fast and your portfolio value is changing every few seconds. With CoW Swap, the ladder can sit there quietly: placing it costs no gas, and neither does changing your mind.

The runner question

The part of the position you keep still needs a plan. A profit on the first two rungs doesn't make the remaining piece risk-free. Decide in advance: will you sell the runner if price falls back to a certain level? At a date? Never? Any of those can be reasonable, as long as you've chosen one deliberately.

Strategy 3: The range bracket

This combines the first two strategies. If an asset has been moving within a range, you can place a buy ladder under the lower end and a sell ladder over the upper end. That lets you accumulate on dips and sell into strength without watching the chart all day.

It works best in sideways markets. It works poorly in strong trends, because in a downtrend your buy ladder keeps filling as price falls. That's why the invalidation level from your planning step matters.

Beyond: more advanced plays

Two orders drawing on one balance

Because several CoW Swap orders can draw on the same balance, you can place two different exits for the same tokens. For example: sell my ETH for USDC at $3,500, or swap it into another token if that pair reaches my target first. Whichever order fills first uses up the balance, and the other can no longer fill in full.

There's an important catch. This is not a true "one-cancels-the-other" order. The second order stays open. If it's partially fillable and some balance is left over, or if you add more funds later, it can still execute. When one order fills, cancel the other yourself. Cancelling is free.

TWAP: a time-based ladder with a floor

A price ladder asks "at what prices?" A Time-Weighted Average Price (TWAP) order asks "over what period?" It splits your trade into equal parts and executes them at fixed intervals.

TWAP orders on CoW Protocol also have a price protection setting. If the market falls below your chosen threshold, that part of the order waits until price recovers. So a TWAP can work like a ladder that runs over time with a minimum price. It's useful for large trades where price impact matters, or for dollar-cost averaging into a position over time.

TWAP has its own wallet and minimum-size requirements, so check the current details on docs.cow.fi before you start.

Conditional and programmatic orders

For builders and smart-contract wallet users, CoW Protocol's programmatic order framework can encode more complex logic: orders that only become active when certain conditions are met. The two articles linked below explain how this evolved and how to build with it.

CoW Swap tips for ladder builders

Watch the "Executes at" column, not just your limit price. It shows the price the market actually needs to reach for your order to fill, once network costs are included.
Keep rungs a sensible size on mainnet. A rung's fill has to cover network fees. Very small rungs may therefore need a better market price before they execute. For the same reason, a partially filled order can leave a small leftover amount.
Set expiries. A stale rung from last month's plan can fill at a price that no longer fits this month's view.
Clean up after partial fills. Re-check your remaining rungs so they match what you actually hold.

Common mistakes

Sizing each rung comfortably but never adding up the total if everything fills.
Moving rungs upward to "catch" a rally. That's chasing, not a plan.
Leaving old sell orders open after you've already closed the position by hand.
Treating a lower average price as proof you're right. A better average on a thesis that's wrong is still a losing trade.
Laddering into a strong downtrend with no point at which you stop.

The bottom line

Laddering doesn't make markets predictable. It organizes your uncertainty into something you can manage. You pick your prices when you're calm, the orders sit there without costing gas, and solvers compete to settle them at your price or better.

Ready to build your first ladder? Open CoW Swap, switch to the Limit tab, and start with three rungs.