Reviewing Your Own Fill History to Understand Maker and Taker Costs

Most people who trade on OKX can see a total fee number somewhere in their account, but they cannot explain why one order cost more to execute than another that looked almost identical. The answer is usually hiding in the fill history, not in the summary. Each individual execution carries its own record: which order it came from, whether it added liquidity or removed it, what size it was, and when it happened. When you read those records side by side, patterns appear. You start to notice that your quick market entries all behave one way and your patient limit orders behave another, and you can connect that difference to the way the fee schedule is structured. This guide is about building that reading habit. It walks through where to find your own fill records on OKX, how to match fills to orders, what to write down so the picture stays clear across weeks, and where the official fee documentation fits in. It does not tell you what to trade or how to trade it. It is a method for looking at what already happened in your account and understanding the cost side of it.

Start from your own fills, not from the fee table

The fee table tells you how the exchange classifies orders in general. Your fill history tells you how that classification landed on your specific trades. Reading the table first is tempting, but it usually produces the same confusion you started with, because the table describes categories and your account contains events. Begin with the events. Open your order history in the OKX account area and switch to the view that lists individual fills rather than whole orders. A single order can produce several fills if it was large enough to match against more than one counterparty, and each of those fills is a separate cost event.

Once the fill list is in front of you, resist the urge to sum anything yet. Your first pass is purely descriptive. Note which column shows the order that generated the fill, which shows the size, which shows the time, and whether there is a column or tag indicating the role the fill played in the book. Column names and layouts change, so treat whatever you see as something to confirm against the current interface rather than something fixed. If a column is ambiguous, the help centre is the place to check what it means before you build any conclusions on top of it.

The reason to start here is that maker and taker costs are not properties of you as a trader. They are properties of each individual execution, decided at the moment it happened by whether your order was resting in the book or crossing into it. Two orders you placed ten minutes apart can end up on opposite sides of that line for reasons that had nothing to do with your intent. Only the fill record shows you which side each one landed on.

Matching each fill back to the order that produced it

With the fill list open, pick a small number of recent fills and trace each one back to its parent order. The order identifier is the thread that connects them. Write down, for each fill, the order it belongs to, the side, the size, the timestamp, and the role indicator if the interface shows one. Then open the corresponding order and compare. You are looking for the moment of transition: was the order resting when the fill arrived, or was it still travelling into the book?

This comparison is where most of the useful learning happens. A limit order that sat untouched for a while and then filled is the classic resting case. A limit order that filled the instant you submitted it, because it was priced through the current book, is the classic crossing case even though it was technically a limit order. Market orders almost always cross immediately, but the fill record is what confirms it rather than your assumption. If you cannot tell from the order view alone, the help centre documentation on order types and execution is the right reference, and the fee page explains how the roles are treated once identified.

Keep the sample small at first. Five or six fills that you have fully traced are worth more than fifty you have only glanced at. The goal of this pass is not statistics, it is calibration: teaching yourself what a resting fill and a crossing fill actually look like in your own account, in your own interface, at your own account settings. Once you can classify a fill confidently by eye, you can widen the sample.

Recording what you find so the pattern survives the week

Memory is a poor record keeper for execution costs. A simple table, kept somewhere you will actually return to, does the job. One row per fill, with columns for date, order identifier, side, size, role, and any fee figure the interface shows for that fill. Add a short note column for anything unusual: a fill that arrived in several pieces, an order you amended before it filled, a moment when the book looked thin. Those notes are what turn a list of numbers into an explanation later.

Decide in advance how often you will update the table. A weekly pass over the previous seven days is usually enough to keep it current without turning it into a chore, and it keeps the fills recent enough that you still remember the context. When you update, do not just append rows. Look at the previous week's rows alongside the new ones and ask whether the mix of roles has shifted. If your fills have drifted toward one role, that is a change in how you are trading, and it will show up in your costs before it shows up anywhere else.

Record the fee figures exactly as the interface presents them, and note the date you read them. Fee schedules are revised, account tiers can change, and promotional arrangements come and go. A figure copied today may not describe a fill from three months ago. The official fee page is the reference for how fees are currently structured, and the account settings area is where your own tier and any active arrangements are described. Neither of those pages replaces your own record of what was actually charged on a specific fill.

Turning the record into specific questions

Once you have a few weeks of rows, the table starts generating questions that are worth answering. Which of your orders tend to rest and which tend to cross? Is the split stable, or does it swing with the time of day or the size you are working? Are there orders you believed were patient that actually crossed on arrival because of how they were priced? Each of these questions has an answer in your own data, and each answer tells you something concrete about where your execution costs come from.

Some questions will point back at the interface rather than at your behaviour. If a fill shows a role you did not expect, check the order type definitions in the help centre and the treatment described on the fee page before concluding anything about your own habits. If a fee figure looks inconsistent with the schedule, check whether your account tier or any active arrangement has changed since the fill, and check the date on which the schedule you are reading took effect. Interface labels and schedule wording are things to verify against the current official pages, not things to assume.

Set a stop condition for yourself. If you find that two or three fills cannot be reconciled after a reasonable check, do not keep digging indefinitely. Record what you observed, note the discrepancy, and move on. A fill history review is a tool for understanding your own patterns, not an audit that must balance to the last unit. The value comes from the recurring shape of the record, and a handful of unresolved rows will not change that shape.

Risk boundary: OKX Independent Review

Digital assets are volatile and derivatives can amplify losses. This website has no login, wallet connection, deposit form or customer-support chat. A referral link only records attribution; it does not guarantee access, pricing, rewards, approval or investment results. Availability can differ by residence, legal entity and product, so no regional access is assumed from language or branding alone.

Scenario checkpoint

• Open your OKX fill history and identify which column shows the parent order, the size, the timestamp, and any role indicator.
• Trace five recent fills back to their orders and classify each one as resting or crossing at the moment it filled.
• Build a simple table with one row per fill and a notes column for anything unusual about the execution.
• Update the table weekly and compare the new rows against the previous week's role mix.
• Record fee figures exactly as shown, with the date you read them, and check the official fee page for how fees are currently structured.
• When a fill does not reconcile after a reasonable check, note the discrepancy and move on rather than digging indefinitely.