A heuristic approach to Application Portfolio Management

1. INTRODUCTION
Globalization and the dilution of trade boundaries have forced industries to evaluate various options for cost-effective operations. Industries now face the challenge of channeling resources to meet organizational objectives in continuously changing environments. Information Technology plays a major role in such a dynamic business environment. However, with the advent of new technological tools and continuous advancement, the IT environment has become very complex.

Chief Information Officers (CIOs) regularly add new assets to their IT portfolio, making the portfolio more and more complex. In addition, they face the task of aligning business needs and IT assets. Cost consideration has becoming increasingly important. The challenge is to do more for less.

In such a dynamic situation, the management of the IT Portfolio has become an increasingly important and crucial task. An increasing number of CIOs are pursuing IT portfolio management as a tool to prioritize investment decisions, decide the location of various activities, evaluate the various assets with the value it delivers, and more.

Each asset in the portfolio is evaluated against parameters like cost of procurement, cost of managing, and indeed the cost of replacing the same. “Total cost of ownership” is seen in conjunction with the value delivered by the asset. Optimal management of assets leads to minimal TCO.

A study by Gartner states that, “Approximately 30 percent of the total cost of ownership during the life of an application is for its maintenance and management.” One can achieve a significant degree of cost reduction by maneuvering the costs associated with maintenance and management.
2. WHAT IS APPLICATION PORTFOLIO MANAGEMENT (APM)?
Portfolio management is the process of managing assets and investments in order to achieve desired organizational goals. The portfolio is a combination of assets that are expected to provide certain returns. It has risks associated with it. PM includes selecting a set of assets congruent with the set goal, managing the economic lifecycle of those assets, dynamically divesting and investing in different assets to optimize gains. In the IT context, the portfolio includes application software, hardware, infrastructure, resources, processes, and so on. While financial portfolio management has been in practice for many years, IT portfolio management is relatively new and gaining ground. An integrated framework that helps in investment, divestment, modifications, and movement of application assets has become a necessity to align business and IT goals of an enterprise.


Application Portfolio Management can be best described as:

• A "living program" that allows you to assess the applications in your portfolio, evaluate potential changes, and understand the risks and impact of these changes to the portfolio.
• A discipline and a set of tool that enables a CIO to respond to the pressures of managing an application portfolio.
• A framework helping in relating the total cost of ownership to revenue, identifying redundancies and gaps in current capabilities, pinpointing trouble spots, and highlighting opportunities to pursue sourcing alternatives.

3. WHY IS IT NEEDED?
An APM framework…

• Continuously monitors the environmental changes in business and keeps it optimal.
• Aligns business and IT objectives.
• Reduces portfolio complexity and creates a portfolio roadmap.
• Reduces the total cost of ownership.
4. AN APPROACH TO APM
The introduction of APM in an organization has to be phased balancing the costs, benefits, risks, and business objectives. The APM is a continuous process like a Financial Portfolio Management where the portfolio manager continuously watches the environmental changes and fine-tunes the portfolio for optimal gain.

The Application Portfolio Management necessarily requires a life cycle for effectiveness. The phases of the lifecycle are described hereunder:

4.1 Define Goals & Strategies
IT initiatives are meant to facilitate better business goal achievement. Business goals are defined for the overall organization. A top-down approach is recommended to create IT goals congruent with business goals. One of the options could be the use of Balance Score Card (BSC) developed by Kaplan and Norton. Once the IT goals are frozen, the same is broken into functional goals. This leads to a broad-level functional and technological implementation strategy.

Some key tasks are listed below:
• Identify relevant business Goals
• Identify IT goals and initiatives
• Map IT initiatives
• Cluster the applications on the basis of
o Functional
o Technology
• Evaluate alternatives strategy for clusters
• Finalize the strategy for each cluster
• Set goal for APM initiatives



4.2 Resource the initiative

Management sponsorship is a must to take the initiative forward. It is suggested that one must identify an owner for the initiative, get a budget approval, set improvement goals, and create a plan to complete the exercise. Some of the tasks during this phase are:
• Obtain management sponsorship
• Identify a lead to carry the initiative
• Get budget allocation
• Get Plan approval

4.3 Conduct an assessment
An analytical framework must be put in place to assess each application from the Maintain, Improve, and Retire perspective. The assessment is a combination of interviews with stakeholders and relevant data collection. The collected data is sanitized and analyzed using the framework. An index representing the ease of movement and value creation is established for each application. The set of applications are classified in different bands based on ease of movement, value creation, functional grouping, and technological grouping. A final sequence is arrived at by superimposing customer comfort level on the final analysis. The task sequence is listed hereunder:
• Portfolio analysis
• Model Building / Customization
• Analysis and computation
o Application Index
o Criticality, Volatility, Complexity indices
o Value Creation Index
• Evaluate applications and decide the strategy

4.3.1 Developing application indices

A detailed analysis in terms of criticality, complexity, value, cost, etc. is necessary to get an insight into the state of the application and conclude its future. Since the analysis is aimed at examining the application portfolio and improving the cost–value performance, an index is established to throw light on ease of movement.

Application index is a measure of moveability of the application. Higher the index, the more difficult it is to move. Developing the index helps in classifying applications in different bands and gets a relative assessment of application moveability.

4.3.2 Portfolio Analysis
Application portfolio analysis is the first step towards application outsourcing planning. It details:

• Application
• Technical Environment
• Functional Group
• Documentation Group
• Associated costs

The following table provides a high-level overview of the Portfolio Analysis phase:

Portfolio Analysis

Entry Criteria

1. Go ahead
Exit Criteria

1. Identification of application portfolio.
2. Determination of technical and functional grouping.

Methods and Tools

1. Data Collection—Forms/Questionnaire
2. Interviews
3. Data Summarization
Key Tasks

1. Establish a high-level plan.
2. Discuss the detailed approach.
3. Conduct interviews with application managers, key users, and IT directors.
4. Collect data from available records.
5. Identify functional group.
6. Identify technical grouping.

4.3.3 Model Building and customization
This phase helps in customizing the framework to suit specific requirements of the customer. The diagram depicts the base model, which is customized during this phase.

The following table provides a brief overview of Model Building and customization:

Model Building and customization

Entry Criteria

1. Attribute-wise application inventory.
2. Technological and functional grouping.
Exit Criteria

1. Definition of application attributes and its factors.
2. Model for application index.

Methods and Tools

1. Sum of position digit method.
Key Tasks

1. Identify significant factors affecting application outsourcing.
2. Define each factor as an index.
3. Assign weight to each application factor.

4.3.4 Analysis and computation
During this phase, significant factors are grouped together to develop various application indices. The data is checked/examined for range behavior, indices are computed, and normalization of indices is done to bring parity amongst the indices.

Analysis and computation

Entry Criteria

1. Application indices definition
Exit Criteria

1. Application index
2. Functional group
3. Technological group

Methods and Tools

1. Statistical analysis (distribution, range analysis)
2. Delphi techniques
Key Tasks

1. Establish the indices.
2. Normalize the application indices.
3. Compute composite application index.
4. Map Technological and functional grouping to each application.
5. Compute Documentation index.

Factors contributing to each index are identified along with the value range. Data cleansing and analysis is done to arrive at normalized indices. After developing various indices, the Delphi method is used to arrive at the Application Index and the Value Creation Index.
 

4.3.5 Studying Indices
The analysis of the indices gives real power to the framework. It helps to view the application portfolio from different angles and to arrive at conclusions. The following sections provide an overview of the different perspectives:

(i) Cost Vs Criticality
The criticality index for each application is a measure of the application’s criticality from the business standpoint. The quadrants of the Ansoff Matrix depict four possible situations.

Quadrant I: Low criticality and low cost: These applications will require further analysis before any recommendations are made.

Quadrant II: High on criticality and low on cost: These applications can be evaluated for retention.

Quadrant III: Low on criticality and high on cost: These applications are potential candidates for replacement.

Quadrant IV: High criticality and high cost: These applications should be examined for cost performance improvement.

(ii) Volatility Vs Criticality
Critical applications are the lifeline of an organization. The role of the CIO is to provide stable applications to the organization. The Ansoff Matrix of Volatility-Criticality helps to identify the application/application group to improve stability. This study leads to further analysis in terms of design improvement, platform rationalization, re-engineering, and so on.

Quadrant I: Low on criticality and low on volatility: These applications require more analysis for action.

Quadrant II: High on criticality and low on volatility: If there are no other compelling factors, these applications can be continued on an as-is basis. For example, if some applications from this group have the potential for savings by replacing or off-shoring them, they can be moved, or else they can be continued in their present state.

Quadrant III: Low on criticality and high on volatility: We do not have any application falling in this quadrant. Normally, these applications can be evaluated for retiring.

Quadrant IV: High on criticality and high on volatility: These applications need to be improved. The improvement should aim to reduce volatility. This could be possible by replacing or fixing the cause of volatility.

(iii) Analyzing Functional Group Complexity
The various functional group systems must be aligned with business objectives. The Complexity Index helps in identifying the most complex functional group using the Pareto analysis. The Complexity Index of each application is used to derive the Group Complexity Index through the Pareto analysis. This analysis provides an insight for group-wise BPR and re-engineering to reduce system complexity.


(iv) Gas Guzzler
Organizations have limited dollars to support application portfolios. Interestingly, all applications in the portfolio do not consume equal amount of dollars and follow the 20-80 rule. This analysis sheds light on what needs attention, what needs better control, and/or what applications need to be moved reduce overall maintenance cost.

4.4 Implement
Implementation is a crucial phase of APA frame work. A plan is developed and concurrence is obtained from stakeholders. It is recommended to have an implementation team, proper management oversight for implementation. Some of the tasks are:
• Finalize recommendations
• Discuss with stakeholders
• Present to management and obtain their concurrence
• Implement

4.5 Monitor & Control
Portfolio analysis is an ongoing work. Once the recommendations are implemented, there is a need to have periodic evaluation of the portfolio assets. It is recommended to align the periodic portfolio evaluation in the CIO’s KRA for success. Some select tasks shall be:
• Define appropriate Metrics
• Develop a score card
• Collect info, evaluate & monitor
• Realign the overall APM process
5. CONCLUSION

Application Portfolio Management is an important concept for a CIO. It is very helpful in integrating business goals with IT initiatives. There are tools available to practice in different space. However, without waiting for application of tools a beginning can be made to introduce the process and get a better understanding before one invests in the tool.

The heuristic model discussed in this paper is a great tool which can be practiced in an organization. A beginning can be made without any additional investment in APM tools.
There are always applications that consume a significant portion of support resources. An APM study revealed that the top 20 applications in an organization out of 120 accounted for 88% hours spent on maintenance. A differential strategy was suggested for these applications, which included better supervision, reassignment of key resources, and aligning some of the applications with the long-term road map of the enterprise. This helped in reducing the overall maintenance time.

The analysis can also help in identifying applications in different buckets like Maintain, Continue, Improve and Retire.