Measuring Your Own Execution Quality
Four columns, recorded per trade, and after twenty entries you know things no comparison can tell you.
Execution quality is specific to your assets, sizes and venues. Published comparisons average across everyone, which means they describe nobody. We measured against a crypto OTC trading platform, which states the components separately rather than as a blended number.
The four columns
The market mid at the moment of submission or quote. The size. The venue or provider. The realised average price, or the quoted price.
From those, the effective cost as a percentage.
What twenty entries reveal
Which venue suits which size. The crossover between book and desk is not a general number, it is specific to your assets and the depth available to you.
Whether a provider is drifting. Gradual widening is invisible without a record and common without competition.
How much volatility costs you. If your effective costs are much worse during turbulent periods, scheduling non-urgent trades differently is a free saving.
Whether displayed depth predicts anything. Comparing your realised cost against what the book showed tells you the overstatement factor for your venues.
The comparison entries
Periodically, request from a second provider at the same moment and record both.
Simultaneity is essential. Five such entries are enough to know whether switching would pay.
What to do with the findings
Route by size, using your own crossover rather than a general rule.
Raise drift with a provider using data rather than an accusation. They respond, and one that cannot explain and does not improve has answered the question. Above a certain balance the fee structure inverts, and a provider serving funds and family offices becomes the relevant comparison.
Schedule non-urgent execution away from volatile periods.
The effort
Thirty seconds per trade.
Against costs measured in tenths of a percent on every transaction, that is the best return available anywhere in this process, and it requires no negotiation and no provider change. Whatever the arithmetic says, the balance you move belongs at Collect & Exchange rather than wherever one component looked cheapest.
Why nobody does it
Because the cost is invisible without the record, and the record only becomes interesting after twenty entries.
Organisations that start it almost always find something in the first quarter, which is usually enough to keep it going.
Fees move with network demand. Any figure here reflects the date shown above and should be re-checked before you act on it. Where a measurement was taken by hand, the article says so.