In this third and final blog on IT Strategies, I look at some examples and techniques of using data analytics in Information Technology Management.  In previous postings, I wrote “information technology is the interaction between people, information and technology”. When planning IT investments, it’s important that business value be the main driver for delivering solutions. When evaluating IT value, a business must look beyond a particular product or service and identify value using the following criteria:

Identification

  • Understand what value is to the business.
  • Have a process to assess and define potential value.

Conversion

  • Find opportunities for IT to build success.
  • Don’t be afraid to revisit business models and business processes.
  • Have a plan to train and hire qualified people (IT and Business).

Realization

  • Create proactive and long-term processes.
  • Create a sustainable knowledge management process.
  • Continuously measure outcomes against expected results.
  • Access value.

As a practitioner and researcher of information technology management, I am constantly looking for new approaches to bring IT value to my company. Information is mostly about making decisions.  The first blogs discussed creating value from IT assets. Data analytics can provide a way to properly quantify that value by analyzing performance, sizing and monitor data.

Data analytics provides the ability to drive the decision-making process. However, no decision should be made by data analytics alone. When deciding on how analytics can impact decisions to be made there are two specific categories: qualitative and quantitative analytics. Qualitative requires in-depth understanding of business processes and functions to determine reasons in certain conditions and events. Quantitative analysis requires statistical, mathematical and computational methods.

In information technology management, data can be generated by multiple systems as well as business workflows, the amount of which can easily be within the domain of Big Data. Analyzing large and potentially unstructured data sets “Big Data” can give crucial insight into data-intensive environments.

Business Analysis Process

I also find it helpful to form a business analysis process as part of the overall strategy of IT systems. The business analysis process includes

  • Problem recognition
  • Review previous problems and findings
  • Modeling
  • Data collection
  • Data analysis
  • Communicating and acting on results
  • Business decisions

DataAnalytics

Data Analytics Ecosystem

2017-06-12_16-56-38

When coordinating data analytics into action that involves operation and business optimization, one imperative is to develop policies and processes that adhere to data analytic standards and practices. Applying data analytics to only one a project or process, but leaving out other areas in project or steps tends to weaken the impact or create biases in the end result or deliverable.

The evolution of business is to create data governance, new business models, policy and procedures that adhere to analytical practices. This is know as the data analytics ecosystem.

In this blog, I use examples from SAS Enterprise Miner® a data mining and predictive analytics tool.  Part of the SAS Enterprise Miner paradigm for data analysis is identified by the SEMMA™ method, which includes

  1. Sample: Create a sample set of data either through random sampling or top tier sampling.  Create a test, training and validation set of data.
  2. Explore: Use exploratory methods on the data.  This includes descriptive statistics, scatter plots, histograms, etc.
  3. Modify:  Create imputation or filter data.  Perform cluster analysis, association and segmentation.
  4. Model:  Model the data using Logistic or Linear regression, Neural Networking, and Decision Trees.
  5. Assess:  Access the model by comparing it to other model types and again real data. Determine how close your model is to reality.  Test the data using hypothesis testing.

Information Technology Management

IT strategy involves aligning overall business goals and technology investment.  The first priority is for IT resources, people and functions to be planned around the overall business organization goals.  In order for such alignment to take place, IT managers need to communicate their strategy in business terms.   What makes such efforts inefficient is not making communication and transparency a top priority.

In many companies, funding for strategic initiatives is allocated in stages so their potential value can be reassessed between those stages.  When executives introduce a new business plan to increase market share by 15 percent with a new technology, IT managers must also meet those goals by assessing the quality of the IT infrastructure.

Executives must have confidence that the IT assets that they purchase are sound.  There must be mutual trust, visible business support, and IT staff who are part of the business problem-solving team.   All of these factors are needed to properly determine the business value of IT.

When creating an IT Strategy that can align to business objectives, five themes should be addressed.  These include business improvement, business enabling, business opportunities, opportunity leverage and infrastructure.  Research has shown that companies who have a framework for making targeted investments in IT infrastructure will further their overall strategic development and direction.  When companies fail to make IT infrastructure investment strategic, they struggle on how to justify or fund for it.

Communication is critical to executives and business decision makers.  IT staff typically work across many organizational units and must be effective at translating technical requirements into business requirements and vice versa.  Communication has become mission critical in the IT business value proposition.  When deciding how to apply data analytics across the organizations, IT should work with business leaders by looking at the IT function areas that produce the most data for their organization.  These areas include:

  • business analysis
  • system analysis
  • data management
  • project management
  • architecture
  • application development
  • quality assurance and testing
  • infrastructure
  • application and system support
  • data center operations

IT strategies require full business integration.  When IT managers are proposing new strategies, an executive summary should be the most important part of the proposal, prototype, roadmap, technical architecture document, etc.

Along with IT system metrics, IT managers must also keep in mind business operational metrics which are metrics based more on labor and time.  IT managers need to factor both IT and operational metrics in reports to business stakeholders.  There are several ways of reporting IT strategies to the business. Key Performance Indicators (KPIs) are fundamental to business decisions and are used to correlate business performance such as the how often a transaction results in a customer satisfaction.  KPIs examples include:

  • Efficiency rates.
  • Customer satisfaction scores
  • Capacity rates
  • Incident reporting rate
  • Total penalties paid per incident

Balanced Scorecards are strategic initiatives that align business strategy to corporate vision and goals.  It’s typically not the responsibility of IT managers to build scorecards, but rather understand the corporate balanced scorecards when building IT strategies.

Dashboards are visual representations of success, risk, status and failure of business operations.  In a very high paced organization, they allow information to be quickly disseminated and assessed by stakeholders for business decision making.  Dashboards tend to have more quantitative analysis than other types of reporting styles.

System Monitoring

Maintaining system health can be an arduous and time consuming task for system administrators. System administration include areas such as databases, network, hardware and software. Aggregating the large volumes of raw data can save time and help administrators respond more quickly to issues. Creating analytical methods around such aggregated data can help determine the present and future value of such systems, predict possible failures and security risks, planning budgets for new IT, maintaining existing assets or help plan for the migration to new platforms such as cloud.  For example, data that tracks the amount of storage area network (SAN) usage over a period of time can help create sizing requirements for new systems that will grow at similar rates.

Below are examples of the type of system performance data that can be used when creating data analytics for sizing and performance analysis.

2016-03-12_22-25-03
CPU utilization based on user, system, waits and idle times.

Disks
Disk read kilobytes per second versus disk write kilobytes per second.

Data Analytics

In the past year, I’ve learned various methods to predict trends and detect anomalies of the data I’ve received through the operation of IT systems. IT systems are constantly collecting sensing and monitoring data on CPU, networking, applications, etc. that can been used to build strategies for planning IT budgets. The types of methods I used include

Data Exploration, Cleansing and Sampling

  • Scatter Plots
  • Imputation
  • Filtering
  • Classification
  • Hypothesis Testing
  • Statistics Analysis (descriptive, process control)

Predictive Analysis

  • Logistic/Linear regression
  • Neural Network
  • Probability Distribution

Segmentation Analysis

  • Clustering
  • Association

Model Assessment, Testing and Scoring

  • ROC Charts
  • Lift Charts
  • Model Comparison
  • Data Partitioning (separating data into testing, training and validation sets)

Below are visualizations of based on analytical methods I’ve deployed for information technology management.  I recommend researching these methods to get a better understanding of how they work.  Much of this work was performed in Microsoft Excel, SAS Enterprise Miner® and Python.

1-22-2014 12-53-18 AM

Above, liner regression based on input and output (I/O) waits and the number of disk reads

Image3

Segmentation analysis based on number of processes to CPU utilization rates for various UNIX systems.

2014-11-13_16-38-27

Statistical process control (SPC) Shewart analysis of process elapsed time in seconds.

30658D68

Above, a receiver operating character curve or ROC curve, plots true positive rate against a false positive rate for points in a diagnostic test.  A ROC curve can diagnose the performance of a model.  The baseline is linear  where each model curve demonstrates the trade-off between sensitivity and specificity. More accurate models have curves that follow the left side of the chart to the upper border.  As in the model assessment tool, the data is partitioned into training and validation sets and then the models for each set are assessed for predictability.

Image6

Model scoring for logistic regression

Model_comp_Cum_lift2

Model comparison using cumulative life (training and validation data).  Lift measures the effectiveness of a predictive model using using results of that model when it is applied and when it’s not.

Again, I strongly recommend researching these techniques since there are many super intelligent people out there that I consult.  Also, If there is anything I’ve mentioned that is incorrect, please comment.

Recommendations

Below are guidelines and recommendations on how IT departments and IT managers can leverage business and data analytics to drive IT value proposition.

Determine important business metrics and create a metric measurement plan.

IT managers must understand which metrics are most important for their business.  Start by having a strong understanding of business scorecards and key performance indicators.  This goes beyond just understanding an organization’s goals and objectives. IT System metrics are principally designed only for IT managers and IT Staff; The business understands operational metrics.  When deciding which metrics to collect, focus specifically on business level KPIs and balanced scorecards.  Getting an understanding of what the business wants will drive all further actions in creating IT value for the business.   Create a metric measurement plan that formalizes the process and nomenclature of measuring IT metrics including creating a process to applying them to business functions.

Create categories for metrics.

Specify categories of metrics to communicate including operational, KPIs, dashboards, tolerances and analytical metrics.

Operational metrics include basic observations in the IT management of specific business functional areas.  It is typically revised to include operational metrics with analytical metrics.  Types of operational metrics include measurements of function area incidents, including labor and time allocation for those incidents.  These types of metrics tend to be non-technical in nature but have a definite impact on IT management.

Analytical metrics include metrics that are used for statistical analysis, forecasting, prediction and segmentation.  The data collected for these metrics are typically produced by IT systems.

Tolerance threshold metrics measure tolerances of KPIs values.  Tolerance is very similar to the control chart example in the preceding section, except it is used more for business level control limits.

Key Performance Indicators are perhaps the most important way of communicating metrics to business stakeholders.

Build a management report.

Incident management tracks specific events that deviate from business and operational efficiency of an organization.  It can be clearly stated that server values can have a huge business and operational impact.  Less empirical incidents such as server performance issues and application response time can play a role in adverse events.  Incident management can include operational metrics and KPIs.   For example, the following list describes the type of incidents reported:

  • Total number of incidents.
  • Average time to resolve severity 1 and severity 2 incidents.
  • Number of incidents with customer impact.
  • Incident management labor hours.
  • Total available hours to work on incidents
  • Total labor hours to resolve incidents.

Data analytics can provide supportive evidence of how an incident occurred. Data analytics more importantly can help reduce major incidents by lowering incident costs and time and help improve KPI values.  Typically data analytics is not appropriate in an incident report, however, it allows IT managers the ability to report mitigation and risk factors by rating the level of risks these incidents have to business. Analytics can provide more insight into risk management and mitigation.

As mentioned earlier, data analytics can provide supportive evidence of how an incident occurred, but it can also be used to build a risk management plan and scoring system.  Since analytics provides huge benefits to IT managers about the health of systems and operations, having such information can help lower risks from incidents by allowing IT personnel to respond to problems faster and even predict problems before they occur.  This in turn helps improve the KPIs in incident management reporting.  Since KPI work on a scoring system, the IT staff can produce calculations based in part on values produced from example analytics.  For example, for metrics A, B, and C, operational KPI scores can be established through the use of proportionality.  The table below demonstrates the use of IT metrics in establishing KPI scores.

Reference Number KPI Calculation
1 Number of system incidents B/A
2 Number of network incidents C/A
3 Incident resolution rate (B/A + C/A)

Example of how KPIs are critical to managing and controlling Incident Management.

Incident management just one type of management system that can be built for metric categories where communication on metrics with the business should occur. Other management systems include:

  • Event management
  • Access management
  • Service desk management
  • Change management.
  • Release management
  • Configuration management
  • Service level management
  • Availability management
  • Capacity management
  • Continuity management
  • IT financial management

Build an IT governance program for IT business communication.

Having a data and IT governance program will ensure that data is verified and accurate before being sent to the executives.  Establishing such a program will give some formal assurance that information provided by IT comes from validated sources, has been approved, and has accountability.

Communicate effectively with executives with an executive summary and report.

As mentioned earlier, effective and regular communication will help ensure that IT managers will receive proper feedback, align with the business and prevent unexpected surprises when budget time arrives.

Give executives something to be excited about.

Business executives do not respond well to complex technical details.  Contrary to popular belief, very few people, especially in executive and mid-level positions are impressed by wordy technical details about system architecture and applications.  They need high level examples that show how the business will grow and achieve a project goals using IT management for a business function.  This can include bar charts or diagrams, but they must be business related and clearly indicate how they would achieve business objectives.

Propose a well-planned budget.

A well plan budget consist of replacement costs, unplanned purchases, reoccurring costs and tracking expenses year round.  It’s important to have a complete budget that builds out the solution for current and new architecture with an evaluation of the cost differences.

Executives will always ask for more clarity and more relevance.

An IT team may have worked many hours to produce a clean, bound and lamented report delivered with precious care and a bow to business executives, and still it can be rejected, scrutinized or sent back for clarification.  This is normal and is to be expected.  It is important for IT managers to keep in mind that the goal is always to provide the most factual and relevant information to business decision-makers.

Blog includes excerpts from Analytical Properties of Data-Driven Systems and its uses in Information Technology Management. University of North Carolina at Greensboro Bryan School of Business and Economics, Department of Information System and Supply Chain Management ISM 698-01D 2016.