Artificial Intelligence (AI) – A Potent Partner for Program and Project Portfolio Management

Introduction

Artificial Intelligence (AI) holds immense potential for revolutionizing business planning and operations. However, the lack of standards poses a significant hurdle in fully harnessing AI's power. This is especially true for transformation programs, project portfolios, and ongoing IT support management.

AI is the ability of machines or software to perform tasks that usually require human intelligence, such as reasoning, learning, decision-making, or natural language processing. AI can help optimize their program and project portfolio management (#PPM) by providing insights, predictions, recommendations, and automation for various aspects of the PPM lifecycle, such as strategy alignment, project task estimating, resource allocation, risk assessment, performance evaluation, or stakeholder communication.

When transformation programs and their respective PPM platforms adhere to established standards, the task of leveraging AI becomes more manageable and cost-effective. Standards play a crucial role in simplifying the layers of data harmonization required for AI to leverage the information. This emphasis on standards provides a reassuring framework for the effective use of AI in PPM.

These standards, which encompass defining project types, methodologies, work products, work product estimating, measurements, resource primary skills, etc., are not just technical details. They are a fundamental requirement for building AI systems that are not only effective but also reliable and ethical.

PPM Platforms and Candidate Data Objects for Standardization

First, we need to establish what I define as a PPM Platform. A PPM Platform is a portfolio of capabilities that include the following:

• Ideation log where innovations of new capabilities and solutions are maintained, scored, and prioritized
• Scoring criteria for proposed innovations and next-generation solutions based on growth opportunities, efficiency gains, etc., and their alignment with the strategic goals of the organization
• Ability to facilitate a “Stage/Gate” funnel of what is in development and preparing for total funding (and what is placed in ‘hold’)
• Project templates by project type for ease of approved project planning inclusive of configurable program and project smart masks
• Centralized resource center with full integration into the email active domain server
• Work product execution solutions that can include JIRA, Azure DevOps, etc., with integration into the PPM
• Centralized risk, actions, issues, and decision (RAID) log with integration into the PPM
• Metadata configurable dictionary for purposes of tagging project tasks, resources, materials, etc., for analytical and trending purposes

PPM data object candidates for standardization can include the program and project types or categories, Strategic “Key Results and Outcomes” (KROs) and “Key Performance Indicators” (KPIs), project phases, resources, baseline versions, standard deliverables/work products, and benefits.

The metadata should also be standardized within each of these data object candidates. The metadata should predominantly define items that can be selected from a list. For example, work product metadata values could include requirements traceability matrix (RTM), configuration specifications, configuration and unit test, functional specifications, etc.

Work product execution solutions like JIRA or Azure DevOps should also provide standard issue types or user stories aligning with the work product. The respective issue type workflow or user story task list should have a minimum set of standard steps to assess and measure the work product's progress.

PPM Data Quality, Consistency, and Interoperability

Consistent data quality and interoperability in a project portfolio management (PPM) platform are critical for multiple reasons. Consistency of data quality makes integrating data from different sources and systems, such as work product execution solutions, financial systems, or HR systems, much more efficient and effective. Data quality and interoperability consistency can also provide a comprehensive view of the project portfolio.

Consistent data quality and interoperability also facilitate the analysis and reporting of project planning and delivery performance across multiple dimensions, such as work product, value stream, business unit, schedule, cost, quality, risk, or expected benefit rating.

Finally, consistent data quality and interoperability provide a much more efficient application of artificial intelligence (AI) techniques by reducing the layers of data harmonization before applying AI techniques and leveraging AI’s capabilities.

These AI capabilities can help program and product stakeholders make better decisions, optimize resources, identify and mitigate risks, and deliver successful outcomes. These capabilities can also identify delivery bottlenecks via the work product execution solutions where stakeholders can address and mitigate those bottleneck issues. Therefore, having consistent data quality and interoperability in a PPM platform is essential for enhancing the performance and maturity of project portfolio management.

Fairness, Trust, Transparency, and Better AI

Data quality consistency and interoperability provide better transparency, trust, fairness, and better AI by ensuring that the data used for AI analysis is accurate, complete, and comparable across projects and portfolios.

Transparency means that AI's data sources, methods, and results are transparent to the stakeholders because of established standards. Trust means stakeholders can rely on the data and the AI to support their decision-making and accountability. Fairness means that the data and the AI do not introduce or reinforce biases or discrimination against groups or individual projects.

Better AI means that the data and the AI can produce more efficient, accurate, relevant, and actionable insights and recommendations to improve project portfolio management performance and outcomes. For example, I implemented a solution for a client that integrated a program’s project portfolio with Atlassian JIRA. The data object standards included the project-planned work products mapped to JIRA issue types.

Each JIRA issue type had a standard set of workflow steps identified and mapped for quality and progress measurement for the work product task in the project plan. This standardization, automation, and integration significantly reduced the number of FTEs required to manage a project, and a basic AI application also identified bottlenecks as workflows were being executed.

This is only one example of how AI can be leveraged in program and project portfolio management if there are consistent data quality standards, consistency of those standards, and transparency.

Conclusion

Machine learning (ML), the predecessor of AI, has been available for decades. Much like AI, ML’s reliability and accuracy are only as good as the data quality and standards it is leveraging. By standardizing the data objects and enabling metadata in a PPM platform, organizations can improve the quality and reliability of their project data and enable AI to leverage it for better insights and decisions.

Many transformation programs struggle to succeed initially, much less gain acceleration, efficiency, and velocity as they mature. While organizational change management, or lack thereof, is often a significant contributor to this struggle, the lack of program, product, and project management data quality standards is also significant.

In my experience, two global transformation programs have continuously failed to meet their interim and long-term goals and objectives over the last six years. They have expended over 25% of their respective funding while only meeting less than 5% of their deployment objectives. One of the most significant contributors to their lack of progress is the lack of program, product, and project management planning and management standards.

Having standards can help these transformation programs identify improvement opportunities, drive planning and delivery management consistency, and accelerate the transformation by leveraging AI actionable intelligence for continuous improvement.

What is the state of your transformation program and project portfolio? Are there established standards and consistent use of those standards? What lessons have you learned in establishing standards? Have you considered how AI can be leveraged with those standards?