Product portfolio management has changed significantly in terms of both methods and technology over the past three years. Below is a summary of the main trends in the enterprise software market and application landscape.
1. Artificial intelligence: from analysis to decision support
Until a few years ago, the role of artificial intelligence in portfolio management was limited to predictive models. Today, enterprise application providers offer embedded AI and agent-based applications that structure product data, predict quality issues, and suggest corrective actions. Concrete use cases on the portfolio side include:
Automatically classifying and connecting free-text customer requests and complaints to a feedback pool.
Detecting duplicate product and material records through similarity analysis.
Generating a demand forecast for the new product based on the past performance of similar products.
Measuring systematic optimism bias by comparing business case assumptions with actual data.
Automatically identifying rationalization candidates and explaining the reasoning behind it.
A natural language query on portfolio data: "Which products saw their margins fall by more than 5% last quarter?"
The general expectation in the industry is that the vast majority of enterprise applications in the coming period will include embedded assistant or agent capabilities. However, the main determining factor in the field remains the same: the quality of the artificial intelligence cannot surpass the quality of the underlying product data.
2. Software development of the physical product
An increasing number of products now consist of a combination of hardware, embedded software, and cloud services. This is fundamentally changing portfolio management: the roadmap for hardware is measured in years, while the roadmap for software is measured in weeks. Synchronizing components that are evolving at different speeds within the same portfolio has become one of the most challenging problems in current portfolio management. Furthermore, since a product continues to change even after it's sold, "launch" is no longer an endpoint.
3. The merging of PLM and ERP.
For many years, product data resided in engineering (PLM) and business data in enterprise resource planning (ERP); the bridge between them was built through integration projects. Today's trend is towards eliminating this separation: leading providers are turning to cloud solutions that combine product data with supply chain, sales, and finance processes on the same data model . This has significant implications for portfolio management: when engineering data and sales and cost data can meet in the same query, portfolio analysis ceases to be an integration project and becomes a standardized report.
4. Pressure to simplify
Recent supply chain disruptions, rising input costs, and trade policy fluctuations have forced many organizations to simplify their portfolios. The common observation is that portfolios don't spiral out of control overnight; they grow slowly through the accumulation of individual decisions, each seemingly reasonable in its own context, and complexity eventually becomes a cost structure in itself. Today, simplification is no longer a crisis-specific measure but a continuously implemented discipline .
5. Making sustainability data a requirement
Environmental data is evolving from voluntary reporting to a market entry requirement. The European Union's ecodesign regulation and digital product passport require product data to be machine-readable, verifiable, and traceable throughout the supply chain. However, the level of preparedness on the ground is limited: research shows that most companies in Europe do not yet have the structured lifecycle data required for passport compliance. The main reason is not the absence of data, but its fragmentation across fragile systems .
6. The prominence of screenwriting talent.
Customs duties, currency fluctuations, supply constraints, and regulatory changes shorten the lifespan of decisions. In this environment, portfolio management is not expected to produce a single correct plan; it is expected to be able to quickly revise the plan when assumptions change . Digital twin and simulation approaches bring this capability to the heart of strategic planning.
7. Product portfolio management on leading enterprise platforms.
To see the concrete application of these concepts, it is instructive to examine the approaches of the two major providers in the market. Both platforms build upon the same basic structure under different names.
Concept
Approach 1
Approach 2
Structure
The hierarchy is: Portfolio → Bucket → Portfolio Item; authorization is delegated downwards from the portfolio level.
Product line and portfolio hierarchy; innovation objects are linked to the product registry.
Idea-concept flow
It progresses through portfolio items and proposal objects.
The chain of idea → need → concept → proposal → portfolio is modeled as separate objects.
Decision-making mechanism
Decision points, status flows, and review meeting object.
Step-by-step approach and approval workflows; proposal evaluation processes.
Prioritization
Automatic derivation of critical success factors using evaluation questionnaires and scoring models.
Portfolio selection aligned with strategic and financial objectives using best practice analysis methods.
Balance
Bucket-based financial and capacity planning; comparative evaluation of items.
Balancing the mix of core, adjacent, and transformative innovations with resources and budget.
Scenario
Discussing alternative scenarios through simulation and "what if" analysis.
Comparing alternatives through portfolio scenarios and indicator analysis.
Source
Combining resource requests from different project management systems and matching them with HR data.
Linking project portfolio management with innovation objects
Product data
Integration with product lifecycle, variant management, product compatibility, and costing solutions.
Single corporate product registration and commercialization via the product master data center.
Inference
The solutions offered by market leaders are conceptually very similar. The difference lies not in the concepts themselves, but in how many of these concepts operate on a single database without requiring additional integration projects . In most portfolio management projects that fail, the problem isn't the methodology, but the fact that the data resides in four separate systems.