In Strategic Portfolio Management, Visibility is the Key to Unlocking the Power of AI

In pursuing organizational strategy, many companies find a wide gap between what they wish to do and the reality of what they can do. While almost all businesses recognize the critical nature of Strategic Portfolio Management (SPM), less than half feel that their current processes are truly effective. That was the key takeaway from Broadcom’s recently released annual SPM survey.

Brian Nathanson, Head of Product Management for Clarity at Broadcom, defined SPM as an evolution of Project Portfolio Management (PPM). “Strategic portfolio management aims to expand beyond projects, to include things like operations, products, infrastructure,” he told SD Times. He said the goal is to get full visibility into every investment—not just the 20-30% that’s often considered new work—to make sure they’re driving the overall business strategy.

The biggest disconnect, according to Nathanson, is due to organizational siloing. When business units develop data for their own purposes, they create clusters of information that prevent a unified view of the portfolio. This is despite almost everyone agreeing that SPM is important.

“The challenge is that you have all the information needed to make those decisions spread across the organization and across all these different databases,” said Nathanson. This causes decision makers to be “unsatisfied with the actual data and the system or processes they have in place to make those decisions.” The difficult, expensive task of modernizing legacy systems is often delayed because “there is no cost in place for the business unit or anyone asking to continue to leave the status quo.” The Illusion of AI as a ‘Magic Button’

The survey revealed a near-unanimous belief, with 99% of respondents indicating that AI can provide benefits for improving the SPM portfolio. However, Nathanson cautioned that this enthusiasm ignores a fundamental dependency: data quality.

“If your data is not aggregated and not of high enough quality, the information you get will be skewed or at least incomplete,” he said. He suggested that many leaders are “looking for a magic button to avoid that problem you were describing, which is, I don’t really want to modernize all these things.”

The truth, he warned, is that this is “a rude awakening.” Businesses still need to give AI access to the data it needs, which will inevitably require modernization of legacy systems to some extent.

Step One: Visibility and Data List

For organizations looking to move from PPM to a holistic SPM approach, Nathanson advises starting with visibility. This is not just about throwing all the data into the data pool, he said “he thinks every detail is really important.” Instead, it requires a focused effort to identify and measure data.

“What you really want to do is spend time going through all of your data and all of your systems and determine what data is important data, and then make sure that data is accessible throughout the organization,” he explained. This starts with simple clarity: knowing where to go for what information, before moving to automation. Nathanson’s advice is to start with “a list of applications or a list of data” to understand why data is also captured if it is really important for strategic decision-making.

The shift to SPM also reflects a change in industry terminology. Broadcom’s previous research focused on Value Stream Management (VSM), a term that Nathanson said was “somewhat co-opted by downstream DevOps,” which most managers don’t focus on. The move to Strategic Portfolio Management was a deliberate attempt to address the C-suite in a sense they are already familiar with.