"At the pre-trade point, everybody is sitting down looking at the trade, adding a different flavor," he explained. "Strategy comes from all three groups. That's a little unique in the marketplace because very often you have the trading desk highly segregated from the portfolio manager."
Mr. Lenzo said he uses mainly in-house pre-trade analytic models to understand daily volatility, the number of trades in a day, liquidity, tracking error and sector characteristics.
"The reason we do that is to try to get a feel for the landscape of the trade … where's the risky part and where's the part that's not so risky," he said. "We're starting to see a lot of external transaction cost analysis going in that direction.
"People see value in that because if you can identify risk points prior to trading, you can focus on the parts where you can add value."
Mr. Madhavan said that for him, the most valuable use of pre-trade analytics is for portfolio formation.
"We look at alpha, we look at risk and we look at transaction costs," he said. "Security-specific transaction costs should be used in the (portfolio) optimization process … generating the right trade list in the first place."
Pre-trade analytics can help a portfolio manager avoid unusually costly trades and suggest other options to capture the same exposure at a lower cost, Mr. Madhavan explained.
Still, pre-trade analytics are not without drawbacks and limitations.
Mr. Sussman said the biggest drawback is that pre-trade analytics use historical data to estimate future costs.
"What we consistently heard from money managers was that pre-trade numbers, though accurate over a large set of stocks and a large set of trades, can be wildly inaccurate on individual trades," he explained. "And that inaccuracy comes right at the same moment a trader is desperate for the data."
Typically, this failure occurs in a stock the trader is not familiar with or one that is highly volatile, reacting to news or the expectation of news, he said.
Pointing to the expected growth in the use of pre-trade analytics, Mr. Sussman said money managers are beginning to understand the limitations of pre-trade analysis. In addition, transaction cost analysis developers are working on improving the accuracy of information and the accessibility of pre-trade analytics to combat limitations and money managers' low satisfaction rates.
These limitations are also driving interest in real-time trade analysis, in which traders receive information about how their trades are performing as they are being executed.