Once an application inventory has been created, the next step in Application Portfolio Management (APM) is collecting and analyzing data to make informed decisions about the portfolio. Effective data collection and analysis allow organizations to evaluate applications based on cost, value, usage, and risk, enabling rationalization and strategic optimization. This section provides a detailed guide to prioritizing data, building scoring models, and interpreting insights to drive portfolio decisions.
2.10.1. Importance of Data Collection and Analysis
- Foundation for Decision-Making:
- Accurate data is essential for evaluating which applications to keep, invest in, modernize, or retire.
- Uncovering Insights:
- Data analysis reveals patterns, inefficiencies, and opportunities for cost savings, risk reduction, and innovation.
- Enabling Transparency:
- Clear, evidence-based insights foster stakeholder trust and collaboration in APM efforts.
2.10.2. Prioritizing the Right Data
Focusing on the most relevant data points ensures that APM efforts remain efficient and actionable. Key data categories include:
- Financial Data:
- TCO (Total Cost of Ownership): Costs associated with licensing, maintenance, infrastructure, and support.
- ROI (Return on Investment): The value an application delivers compared to its cost.
- Licensing Costs: Annual fees for software licenses or subscriptions.
- Usage Metrics:
- Number of users, frequency of access, and criticality to daily operations.
- Identifies underutilized applications that may be candidates for retirement.
- Business Value:
- The application’s alignment with strategic goals, customer satisfaction, and revenue contribution.
- Example: An e-commerce platform directly tied to online sales has high business value.
- Technical Metrics:
- Age of the application, technical debt, and compatibility with current systems.
- Includes information about hosting environments (on-premises vs. cloud) and integrations.
- Risk and Compliance:
- Vulnerabilities, end-of-life status, and compliance with regulatory standards like GDPR or HIPAA.
- Example: An outdated application with unsupported versions poses a high-security risk.
2.10.3. Setting Up a Scoring Model
A scoring model provides a standardized framework for evaluating applications based on key criteria. Steps to build a scoring model include:
- Define Criteria:
- Choose evaluation metrics such as cost, usage, technical health, business value, and risk.
- Example: A scoring model might allocate 25% weight to cost, 20% to usage, 30% to business value, and 25% to risk.
- Assign Weights:
- Prioritize criteria based on organizational goals. For example, a cost-focused initiative might give greater weight to financial metrics.
- Create a Rating System:
- Use a scale (e.g., 1-5 or 1-10) to rate applications on each criterion.
- Example: A highly critical application may score a 5 on business value, while a rarely used application scores a 1.
- Calculate Scores:
- Multiply each score by its weight and sum the results to generate a total score for each application.
- Example: A financial application with a score of 4.2 on a 5-point scale may warrant retention, while a legacy system scoring 1.8 may be marked for retirement.
2.10.4. Analyzing the Data
Data analysis transforms raw information into actionable insights. Key techniques include:
- Clustering Applications:
- Group applications based on shared characteristics, such as cost, usage, or risk, to identify trends.
- Example: Clustering can reveal a group of redundant applications serving the same business function.
- Identifying Outliers:
- Highlight applications with unusually high costs, low usage, or significant risks.
- Example: A low-usage application with high licensing fees may be a strong candidate for retirement.
- Benchmarking:
- Compare applications against industry standards or internal best practices to assess performance.
- Example: Benchmarking cloud migration efforts can help prioritize legacy applications for modernization.
- Scenario Modeling:
- Simulate the impact of potential changes, such as retiring or migrating an application.
- Example: A scenario model might show cost savings from consolidating redundant applications into a single platform.
2.10.5. Tools for Data Analysis
Organizations can use a range of tools to analyze application portfolio data, including:
- Spreadsheets:
- Tools like Microsoft Excel or Google Sheets allow for manual scoring and data visualization.
- Best for: Small-scale initiatives or organizations just starting with APM.
- BI Tools (Business Intelligence):
- Platforms like Power BI or Tableau provide advanced analytics and visualizations.
- Best for: Organizations seeking to create dashboards and reports for stakeholders.
- APM-Specific Tools:
- Tools like ServiceNow, LeanIX, or Apptio offer built-in scoring models and analytics capabilities.
- Best for: Mature APM programs with complex portfolios.
2.10.6. Using Data to Make Portfolio Decisions
Once analyzed, data can guide decisions about each application in the portfolio:
- Keep: Retain applications that are well-aligned with business needs and performing effectively.
- Invest: Allocate additional resources to applications that have high potential for strategic impact.
- Migrate: Modernize or re-platform applications to improve performance or reduce risks.
- Retire: Decommission redundant, underutilized, or high-risk applications.
2.10.7. Example: Data-Driven Decisions in Action
- Scenario:
- A manufacturing company analyzed its application portfolio and discovered five duplicate expense management tools across different departments.
- Data Insights:
- Cost: $150,000 annually in licensing fees.
- Usage: Two of the five tools were underutilized.
- Business Value: Consolidating tools could streamline reporting and improve efficiency.
- Action Taken:
- Retired three tools, saving $100,000 annually and reducing IT complexity.
2.10.8. Common Challenges in Data Collection and Analysis
- Incomplete Data:
- Data gaps can lead to inaccurate conclusions.
- Solution: Combine multiple sources, such as interviews, discovery tools, and financial records.
- Analysis Paralysis:
- Overanalyzing data can delay decision-making.
- Solution: Focus on high-impact metrics and avoid overloading the scoring model with unnecessary criteria.
- Stakeholder Resistance:
- Business units may hesitate to share data or accept recommendations.
- Solution: Communicate how decisions based on data will benefit the organization as a whole.
2.10.9. Key Takeaways
- Data collection and analysis are central to APM, enabling organizations to evaluate applications objectively and make informed decisions.
- A well-designed scoring model ensures consistency, transparency, and alignment with business goals.
- By focusing on relevant metrics and actionable insights, organizations can maximize the value of their application portfolio.
2.10.10. Conclusion
Effective data collection and analysis provide the foundation for making strategic portfolio decisions. By prioritizing key metrics, leveraging scoring models, and using appropriate tools, organizations can optimize their applications to support business objectives. The next section will delve into governance fundamentals, outlining how to establish policies, processes, and oversight to sustain APM success.
