This article makes the case that with the increase in data, there is a greater need to make decisions on the data and therefore a need to better represent the information for human consumption and decision making.
“Manually analyzing data is time consuming but is often done in order to maintain core business capacity, operational continuity, competitive advantage and compliance. Reviewing stacks of numbers and text is not only error prone but also makes it difficult to analyze data in order to:
1) Develop or assess a hypothesis: Those managing regulatory compliance may need to consider and assess a hypothesis like Hyman Minsky’s financial instability hypothesis to protect their firm’s future.
2) Discover errors and outliers: From a risk and compliance standpoint, a firm may want to find a way to easily monitor risk exposure across a portfolio on a trade-by-trade basis and manage outliers or trades that are over certain limits.
3) Map trends: From an investment management perspective, a firm may want to track volatility across sectors or industries to capitalize on market opportunity.
4) Create categories: A valuation and risk group may want to know if it can readily quantify exposure to all counterparties by subsidiaries.
5) Make decisions: A structured products group may want to know if it can create “what if” stress scenarios and decide on optimal product selection.
6) Understand relationships, such as spatial hierarchy and rank: For energy traders, the need may be to determine if a company can manage pipeline operations and portfolio optimization across crude, refined, natural gas and other commodities.
The need to effectively and efficiently address these concerns, individually or in combination, is a challenge for many firms. Following a thoughtfully crafted method to hone the possible visualizations choices is a good way to identify the most appropriate one. ”
By Julie Rodriguez and Francesco Brullo