Post-Trade TCA Reporting
How firms turn raw execution data into the standardized reports that answer whether a trade, a trader, or a broker actually performed well — and why the choice of benchmark shapes the answer as much as the data does.
Prerequisites: Implicit Versus Explicit Trading Costs
After a trade is done, the question "how well did that execution go?" isn't answerable from the fill price alone — $50.12 is neither good nor bad in isolation, it only means something once compared against some benchmark of what a reasonable execution should have cost. Post-trade transaction cost analysis (TCA) is the discipline of systematically producing that comparison, trade by trade and rolled up across traders, brokers, algorithms, and venues, turning raw fills into a report that actually answers whether execution added or destroyed value relative to a defined standard.
Choosing what to measure against
The core of any TCA report is the benchmark. Common choices include the arrival price (the price when the order was released to the market — measures total execution quality including impact), VWAP (volume-weighted average price over the trading period — measures how well an order tracked the market's own volume pattern, but can be gamed by an order large enough to move the VWAP itself), and the decision price (the price when the investment decision was made — captures delay cost as well as execution cost, the full implementation-shortfall view). Each benchmark answers a different question, and a single execution can look good against one and mediocre against another — an order that beats VWAP might still show poor performance against arrival price if the order itself was slow to start, which is exactly why a serious TCA report presents several benchmarks side by side rather than picking one and calling it "the" answer.
A well-built report also slices results by dimension — by broker, by algorithm strategy (VWAP, arrival price, liquidity-seeking), by symbol liquidity tier, by time of day — because an average number across all trades can hide the fact that one broker is excellent in liquid large-caps and poor in illiquid small-caps, information that only surfaces once the aggregate is broken apart.
Worked example: reading a simple TCA line
A trade to buy 50,000 shares has: decision price $30.00, arrival price $30.02 (some delay already occurred), average execution price $30.09, and the day's VWAP over the execution window was $30.07. Versus arrival price: bps of cost. Versus VWAP: bps of cost — the order underperformed VWAP by a much smaller margin than it underperformed arrival price. Versus decision price (full shortfall): bps. Reading all three together tells a coherent story: roughly 7 bps of the 30-bp total shortfall came from delay before trading started, the order tracked the market's volume pattern reasonably well once trading began (only 7 bps behind VWAP), but arrival-price cost of 23 bps shows meaningful impact or adverse drift during execution itself — a picture no single number would have given.
What this means in practice
Post-trade TCA is how funds hold their own trading desks, algorithms, and broker relationships accountable with data instead of impression, and it feeds directly into decisions like which broker gets more flow, which algorithm strategy is preferred for which order type, and whether a trading process needs to change. Because the benchmark choice materially changes the answer, TCA reports that show only one number invite cherry-picking — a broker motivated to look good can select the benchmark most flattering to their own execution style rather than the one most informative to the fund.
Comparing execution quality across brokers or algorithms using different benchmarks for each (say, one measured against VWAP and another against arrival price) produces numbers that aren't actually comparable, even though they're both expressed in basis points. Always fix the benchmark before comparing.
Further reading
- Kissell, The Science of Algorithmic Trading and Portfolio Management, ch. 4