In the hunt for latency and in the race to get more data flowing more and more quickly through trading systems, we seem to have forgotten to take into account the most influential and pervasive users of that data: the traders themselves.

The race for low latency, thus far, has focused primarily on two things: reducing the latency in receiving and processing market data and reducing the latency in executing a trade. Insofar as algorithmic trading is concerned, there's also been a third focus: increasing the speed of the algorithms, i.e., once an algorithm has received the market data, reducing the time it takes to make an execution decision.

But what about the trader? Algorithms are not yet, nor are ever likely to be, the only systems responsible for making execution decisions. Often enough, the task of making trade decisions still lies with a human trader. In this instance, the race for low latency will have been for not if we don't also manage to dramatically reduce the trader's latency, i.e., the time between when a trader sees data and then is able to act on that data. This time period is what we term 'trader latency.' It has been to date sorely overlooked.

The task facing the industry is to reduce this latency while allowing traders to deal with ever growing volumes of data updating at ever more frequencies. The solution is two-fold.

First, it is important to transform this data into information before displaying it to the user. In a low latency environment, delivering data to the trader in a classic two-dimensional grid display limits the point of latency reduction in the first place. In essence, all we've done is make the cells on a trader's display blink faster; we haven't allowed the trader to make trade decisions faster.

The emerging trend is towards transforming this rapid influx of data into innovative graphical formats that allow a buy or sell-side trader to very quickly interpret what the data actually means and immediately intuit what is happening in some aspect of the market. As an example, a colored three-dimensional smile surface (a modification of the classic smile curve) can very quickly and intuitively provide a sense of implied volatility vs. strike price vs. time vs. counterparty concentration (or a hundred other parameters) that a grid of data could never provide.

Second, it is important to engage with cues from the trader to determine a trader's context and then deliver relevant information to the trader at the time the trader needs it. The classic solution to information overload has been to throw screens at the problem: if traders want to see more data, they simply add more screens to show that data. The industry has devolved to this solution because the focus has been on delivering more data rather than delivering the right data at the right time. We have to put some intelligence into actually interpreting the trader's needs and prioritizing the display of information to suit those needs.

A simple use-case of this is in the area of turret integration and sales trading. For years turrets have been relegated to the backwaters of infrastructure. However turrets can provide incredibly valuable information about a trader's current context: what client or counterparty they're talking to. This context can be used to determine and prioritize the display of what products and slices of the market are relevant to the trader's immediate needs. It's a simple example, but dramatically increases productivity.

In all the years that the industry has been developing trading systems, user experience has rarely been a significant concern. Thankfully, the industry is finally realizing that reducing latency won't help if the only result is cells blinking faster on a trader's workstation. The remaining bastion of latency that needs to be dramatically reduced is on the trader level. This can be done by arming traders with innovative solutions that enable them to quickly and intuitively deal with real-time data.

Vivake Gupta is the co-founder and managing director at Lab49. Vivake can be reached at info@lab49.com.