At Sabre Holdings Corp., IT does not go by the name enabler, trusted partner or any other moniker du jour conferred
on corporate IT. Information systems are the biz at the Southlake, Texas-based travel company. The Sabre reservation system is the booking system of choice for travel agencies worldwide. Travelocity.com LP, a wholly owned subsidiary of Sabre, is the second-largest online travel agency in the U.S.
"Sabre is into travel in a big way, and we do it through the software, whether it is our own or someone else's," said Greg Bronson, director of portfolio management at the $2.8 billion company. Indeed, about one-third of Sabre's 9,000 people are involved in software development.
Bronson, a 21-year veteran who started at Sabre as a software developer, said that one of the many challenges in the volatile, competitive travel industry is making sure IT does what the market needs.
"We have to be able to change on a dime because of market conditions," Bronson said, "but it's a balancing act, because we have much more to do than 3,000 developers can do." More than twice as much: as many as 1,500 proposals for 600 funded projects annually. "And we have struggled with that over the years," he said.
In addition, the company until recently was doing "planning by product," Bronson said. Project management was focused on pushing products out the door. Projects were prioritized once a year, a three-month process, minimum. Spreadsheets were the tool of choice. "The day after it was done, a market change could make it obsolete," he said.
The liability of the slow selection process was underscored three years ago, when the CEO offered an extra $3 million for new projects and nearly three months went by before Bronson could ascertain the status of the current work and who might be available. "We lost out on 25% of the money because it took us a quarter to figure out what we could do with it," Bronson said.
Bronson is not alone. In most firms, IT organizations have too much work to do, said Lewis Cardin, an analyst at Forrester Research Inc. in Cambridge, Mass., which is why project portfolio management (PPM) -- selecting, prioritizing and monitoring project results -- is so important. Indeed, he said he believes the processes used for making project investment decisions are "the essence of IT governance."
"PPM must be more than just a decision to proceed, if IT execs want to maximize the value received from these investments," Cardin said.
To maximize the value in IT investments, the PPM process should span the lifetime of the project, from inception of the business case to real-time risk assessment through the benefits realized, Cardin said.
Once a project gets going, it can take on a life of its own, barreling on and eating up resources even when costs and schedules have gone by the boards and the rationale is a distant memory. Cardin recommends "project gating," with go/no-go decisions at each gate as a way to keep a handle on projects. PPM should also include a "benefits register" that records the baseline benefits predicted by the business owners. If the benefit deteriorates over time, a trigger is set to advise IT governance that intervention is required. The business case document for a project needs to be a living artifact, not used for one-time funding decisions.
Springtime for Sabre
Software-driven Sabre put its faith in PPM software to help manage its bulging portfolio of projects. At the end of 2004 -- after a four-year search -- Sabre chose software from Primavera Systems Inc. (Italian for spring), a project and portfolio management software vendor based in Bala Cynwyd, Pa. If an extra $3 million makes an appearance, "Now I can come up with an answer in three hours," Bronson said. "It changed the perception of IT."
The software has cut down on the time to approval for projects because everyone who needs to has access to the data, Bronson said. Workforce availability is tracked in real time. The 900 or so leftover projects of any given year stay in queue and are updated regularly, as new projects are added. And, instead of prioritizing projects once a year, Sabre decision makers put on their sorting hats at the end of each quarter.
That's music to the ears of Primavera CEO Joel Koppelman.
We lost out on 25% of the money because it took us a quarter to figure out what we could do with it.
Greg Bronson, director of portfolio management, Sabre Holdings Inc.
Koppelman and Richard Faris founded the $123 million-and-growing company 25 years ago with the aim of making the PC more functional for companies moving off the mainframe. "Project management in those days was primarily one person who had responsibility for it," he recalled. A master planner alone at his desk would somehow pull together all the information and work assignments and come up with a way to get it done.
"That was not a good idea then, although it was convenient," Koppelman said. Moving from the desktop to a client/server environment forced a change. With the Web plus software, people could work together around the clock. "Projects are a collaborative effort. It takes lots of people with different kinds of expertise. People who do the documentation are not the ones who write the code. You have to bring all these people together and let them share their knowledge. That is what people care about today. "
The challenge today, he added, is not so much how to manage a project but how to figure out which project to do. "People say, we're going to spend $2 million on a project but nobody really knows at that point what that means -- where the people will come from, what needs to be contracted out, which consultants to bring in, whether to joint venture. They don't know." Primavera's portfolio space software is designed to bring together people's ideas and give them a clearer view of a project and what it will cost to do. "I don't presume that we're social software, but in a way we are."
The big push lately at Primavera has been on connecting portfolio management with ERP systems, Koppelman said. The company announced yesterday Primavera Evolve, the latest version of its resource management software.
Implementing the software is "not an overnight process," Sabre's Bronson warned. It took six months to get the software up and running, "and we still don't take advantage of all that it can do. It is a very powerful tool."
But the benefits keep accruing. To compete better in a rapidly changing industry, Sabre has lifted product planning from a division-based task to an enterprise-wide process and switched from a get-the-stuff-out-the-door approach to planning around an ever-shifting pyramid of business initiatives. Being able to quickly assess projects as market conditions change is important, Bronson said.
The software has also kept Bronson, et al, looking good for the board of directors. It used to take four months to put together a plan for the board. When the company moved its board meeting from December to November this year, Bronson's team put out a first-draft plan in five days because 2008 was already in the pipeline, he said. "So now instead of scrambling, we can spend time on how to present a better-looking plan to the board."
Let us know what you think about the story; email Linda Tucci, Senior News Writer.