Budgets Reach New High

Five years ago, singer Kelly Clarkson set a record with her first single jumping up the US Billboard Hot 100 from 52 to number one. Today, it is reference data that makes a similar leap. Reference data has traditionally been seen as an area financial institutions have been reluctant to invest in, with market data attracting the bigger budgets.

But the environment is now changing and reference data projects have jumped up the chart. Close to 70% of the end user participants in the Inside Reference Data Survey 2007 (IRD Survey) said their organizations spent less on reference data last year compared to this.

Mark Husler, head of reference data at the London Stock Exchange, says the result does not surprise him considereing the ever-evolving products that the industry is trading. "All of these new instrument set-ups result in challenges for data providers and operational areas to ensure asset set-up is correct and also maintained. We certainly see this area continuing to present challenges and further focus on reference data standardization," he says.

In the next 12 months, the majority of the financial institutions surveyed are likely to spend the most money on reference data projects, as opposed to market data projects, or about the same amount in reference data and market data projects. "Maybe they have just realized that they are not good enough," says a humorous Gavin Quinn, EMEA business development manager at Sybase, adding: "How many times have the SEC had a quiet word [with financial instit­utions]? How many firms have been encouraged to become members of Markit?"

Quinn says there has been a very rapid reference data evolution in a short period of time, something panelists echoed at the North American Financial Information Summit (NAFIS) in New York in May.

"What I did two years ago is very different than what I do today. The infrastructure that we built four or five years ago is continuously being modified throughout every single year because new products get created, new business lines come out-it never changes," said James Perry, vice president, product data quality, Goldman Sachs, speaking at a reference data panel on getting sponsor buy-in.

And sorting out reference data issues can, in Quinn's words, "potentially be very expensive and painful for people". Data management professionals have a long list of problems they want solved. The IRD Survey, which asked participants to rate problems in importance, found that over 80% had reducing risk and improving data quality at the top of their agendas. Improving client services and lowering operational costs generated mixed responses, while only 17% saw complying with new regulation as "very important".

Still at NAFIS, Peter Serenita, chief data officer, JPMorgan remarked that regulation "can be your best friend" from a data management perspective. It can be a significant business driver when trying to get a data management program off the ground.

However, the importance of reference data projects will naturally always vary depending on people's responsibilities, and complying with new regulation will probably be seen as vital in the compliance department. In terms of the survey, Quinn says: "It depends on who you spoke to. The person who runs FX does not care about the quality of bonds reference data, for example."

He adds that one of the problems with many large financial institutions is the lack of having one person in charge of all data projects. "The people who set the agenda are not looking [at reference data] holistically. They want short-term benefits to their particular problems," Quinn says, adding: "Compliance is a cost center. Their voice is not that strong."

CDO Role on the Rise

But the introduction of chief data officers (CDOs) has partly addressed the problem of centralizing decisions. "This shows me that the problem is being recognized at the highest level," says Quinn. He argues that it is advantageous to have a single point of responsibility, and the institutions that currently have CDOs are all Tier 1 "cutting edge organizations" with greater budgets to solve problems. In future, he expects Tier 2 and Tier 3 institutions to follow the same thought process.

The CDO role has also functioned as an enabler for enterprise data management (EDM), with the ultimate goal of CDOs often being to centralize processes and systems firm-wide. The IRD Survey found that the main barriers to EDM are considered to be overcoming platform and technology migration difficulties and consolidating reference data across silos (see graph), followed by getting sponsor buy-in.

In general, data management professionals are struggling with getting sponsor buy-in for reference data projects despite growing budgets. Panelists at NAFIS discussing the topic highlighted that even though the problem has not changed, the nature of the actual data sponsor has evolved. "When we first centralized back in 2000/1999, the primary decision-makers were senior managers within operations and technology," said Perry, explaining that it is more his decision now.

Still, it is seen as difficult to make data sponsors understand the advantages of reference data initiatives. According to the IRD Survey, 50% of respondents said the main barrier to getting sponsor buy-in is that data sponsors do not recognize the benefits of reference data projects. The often-discussed problem with demonstrating return on investment (ROI) came out as the second biggest challenge, while lack of patience for vendor shortcomings was mentioned as another issue. "They want it all from one source," commented a participant.

At NAFIS, moderator Tom Davin, managing director FISD, raised the question: "What is the number one challenge for getting sponsor buy-in?" Reflecting the findings of the survey, Serenita and Perry both underlined the importance of ROI.

Serenita, who organized a show of hands, said: "How many of us have had to cost-justify an ROI case around some of our data programs? How many of us have struggled with putting together an ROI case for our data management projects? Have any of us put this together and said: 'Wow! This is amazing.'?" He admitted that he wished it was this easy, but the truth is: "It just doesn't work that way."

The Trouble with Intuition

The problem is that the business case is intuitive, with numbers that do not really support that intuitiveness, showing that sometime numbers don't tell the whole story, according to Serenita. Perry also argued the case for focusing on value. "If I go to the business and say: "Would you like to spend $200,000?", it's like asking people if they want back surgery. Of course they don't want that," he said, adding: "You have to say, there's a need, we have a problem, we have an issue and we need to purchase it. In some cases, I just purchase it myself and allocate it back out."

And there have been success stories among completed reference data projects. Survey respondents highlight infrastructure, globalization, centralization and data ownership projects as having proved the most productive. One respondent argued that starting with small initiatives had generated the best results.

The most common benefit experienced following reference data projects was improved data quality, the issue most of the institutions also recognized as the most urgent problem to solve. Reference data projects further resulted in increased automation, while improved client services and lower costs shared a third place.

As many institutions are starting to benefit from reference data initiatives completed some years back, the market is receiving increased recognition, driven by a need for better data to reduce risk. It is clear that reference data has jumped the charts, no longer necessarily secondary to market data.