Running volume on Phoenix

An order book is a different animal

Nearly everything written about volume campaigns assumes an automated market maker: a pool with reserves, a formula that always produces a price, and impact that scales predictably with swap size. Phoenix does none of that.

On an order book, prices exist because somebody posted them. A limit order sits at a level, waiting. When you send a market order you consume those resting orders from the best price outward until your size is filled. If nothing is resting, nothing happens, regardless of how much you wanted to trade.

That single structural difference cascades into everything else. There is no liquidity figure to read off a page, no impact percentage to simulate, and no guarantee that a price exists at all. What there is instead is a book, and reading it is the whole skill.

Depth is posted, not pooled

The most useful mental adjustment is that depth on an order book is a decision somebody made, rather than a property of a contract.

In a pool, capital sits there passively and the formula quotes against it whether anyone is paying attention or not. On a book, a market maker decides how much to post and at what prices, and can pull it in a moment. A book that looked deep an hour ago can be empty now because the participant quoting it turned off.

For a campaign this means the venue's capacity is not a stable input. It has to be checked at the time you intend to trade rather than when you plan, and it can change while a campaign is running. Pools have their own version of this problem when ranges drift, but the failure is slower and more visible.

The spread is your real cost

On an AMM you pay the pool fee plus whatever impact your size causes. On a book you pay the taker fee plus the spread you cross, and the second number dominates on anything young.

A campaign is a series of round trips: buy, then sell. Each leg crosses the spread. If the spread is wide, which it is on almost every book without a committed market maker, that cost repeats on every single trade and compounds across the campaign in exactly the way pool fees do, only larger.

The practical consequence is that the fee schedule is not the number to compare. Two venues can have identical fees and completely different real costs because one has a tight book and the other does not. The fee breakdown covers how the venue cost dominates everything the network charges, and on a wide book the spread dominates the venue fee too.

What happens when nobody is quoting

This is the failure mode that has no AMM equivalent, and it surprises people the first time.

• Orders do not fill. Not a slippage failure, not an error: there is simply nothing at your price to trade against.
• Fills happen at absurd levels. If the only resting order is far from the last trade, a market order reaches it and the chart records that price.
• The chart develops gaps. Discrete jumps between levels rather than a curve, which reads as broken to anyone looking.
• Two-sided flow becomes impossible. A book with bids and no asks lets you sell and not buy, or the reverse.

None of this is fixable with settings. A campaign cannot post liquidity to trade against itself in any way that produces a meaningful figure, and trying to do so is a different and much worse idea than routing elsewhere.

There is an upside to the same property, and it is worth stating so this does not read as a dismissal. When a book is well quoted, execution is cleaner than an AMM can offer, because you trade against real posted size at real prices rather than sliding along a curve. A large order can fill at a single level if the size is there, with no impact at all in the AMM sense. That is a genuinely better outcome and it is why serious markets use books. The problem for a young token is not the structure; it is that nobody has committed to quoting it yet.

Designing a campaign that suits the venue

Assuming the book is genuinely two-sided and reasonably tight, order book execution has real advantages worth using.

Read the book before sizing. Look at how much size rests within a percent of the touch on both sides. That number, not a liquidity figure, is your capacity per trade.

Keep orders inside the top levels. An order that consumes three levels of a thin book has moved the quoted price substantially and left a visible mark. Smaller and more frequent is better here for the same reason it is on a shallow pool.

Check both sides separately. Books are frequently lopsided, and a campaign with balanced direction will execute very differently on each leg. The buy-pressure weighting should reflect what is actually postable rather than a round number.

Do not treat it as the only venue. Even a good book is one venue, and concentrating a campaign anywhere maximises both the cost and the legibility, as the liquidity guide covers for pools.

When to skip it entirely

Three situations where the honest answer is that Phoenix is not the venue for your campaign, and no amount of tuning changes that.

The book is empty or one-sided. Nothing to trade against means nothing to produce. Check before assuming a listing implies a market.

Your token's liquidity lives in pools. For most young Solana tokens it does, and routing should follow liquidity rather than a preference for a venue type. If the pools hold everything and the book holds nothing, the book is a rounding error.

You are trying to produce trade count cheaply. Crossing a wide spread hundreds of times is an expensive way to generate trades. Smaller swaps into a deep pool do the same job for less.

None of that is a criticism of order books, which are excellent market structures when there is a committed participant quoting them. It is an argument for routing by where liquidity actually is, which is why venue selection is automatic rather than a checkbox. You can see which venues hold liquidity for your mint, and what a given target costs in SOL, in the campaign console.