Part 2

Data analysis
2.0 Data Analysis Techniques

Data analysis techniques are used to assess and evaluate data in order to discover or gain deeper
information about a topic. They can also be used to support decision-making. We use many differ-
ent methods of data analysis in project management. Some of them are very broad, such as alter-
natives analysis and decision trees. Others are specific to a particular knowledge area, such as a
stakeholder analysis or a make-or-buy analysis.
The techniques described in this section include:

• Alternatives analysis
• Cost benefit analysis
• Cost of quality
• Decision tree analysis
• Earned value analysis
• Influence diagrams
• Make-or-buy analysis
• Performance indexes
• Regression analysis
• Reserve analysis
• Root cause analysis
• Sensitivity analysis
• Stakeholder analysis
• SWOT analysis
• Technical performance analysis
• Variance analysis
• What-if analysis

The techniques in this group can be some of the most challenging to master. Some of them
can be quite complex, such as a sensitivity analysis or earned value analysis. Understanding what
project data is telling you is critical to making good decisions. Therefore, the effort you put into data
analysis can have a significant impact on the success of your project.
This book does not describe data analysis techniques that are very general and easy to under-
stand, such as document analysis or assumption and constraint analysis. It also does not cover
techniques that require specialized knowledge or software, such as simulations (Monte Carlo analysis).
The type of project you are working on, the availability of quality data, and in some cases specific
software, will influence the data analysis techniques you can use on your project.
2.1 Alternatives Analysis

What It Is
Alternatives analysis is a technique used to evaluate and select options or approaches to execute
and perform project work. An alternatives analysis can be used at a high level; or you can apply
a more rigorous approach that incorporates a matrix that identifies rating criteria and weighting
factors. This technique can be used in several processes and in a wide range of industries.

How to Use It
The steps below describe how to conduct an alternatives analysis using a matrix to weight evalu-
ation criteria. Use the steps below as a guideline. Tailor the steps as necessary to work within your
environment. For example, you can apply this technique without weighting the evaluation criteria, in
which case you would omit Step 3.2.

1. Identify the problem or decision you are evaluating. For more information about decision-
making, refer to Section 5.2. For more information about problem solving, refer to Section 5.4.
2. Define Solution Requirements. Often the Solution Requirements can be extracted from
Requirements documentation, the Statement of Work, the Scope Statement, or other project
documents. For information technology (IT) projects, “Use Cases” can also be used to identify
the Solution Requirements.
3. Define Evaluation Criteria and Weightings. The relevant stakeholders should provide input on
the criteria they need in a solution or outcome. They should also provide weight values
for each criterion. The project manager or business analyst often facilitates this process. The
following steps are used to define and weight the selection criteria.
3.1.  Identify the selection criteria.
3.2. Develop a “weighted value” for each criterion. It is a best practice for all weights to add
up to 100 percent, but this is not mandatory.
3.3. Define the scoring algorithms to determine how effectively each alternative meets the
selection criteria. For example, if low cost of ownership is a selection criterion, define
how you would rate alternatives on this criterion on a scale of 1 to 5, with 1 being
the highest cost option and 5 being the lowest cost option.
3.4. Create a Scoring Matrix with a space for alternatives on one axis and the criteria and
weighting on the other axis.
 20    2.1 Alternatives Analysis

4. Identify Options and Conduct Market Research. To conduct market research, you can send
out a Request for Information (RFI), hold meetings with vendors, perform Internet research, or
talk with consultants.
5. Conduct Initial Assessment, Score and Evaluate, and Eliminate Options. The initial assess-
ment can be used to narrow the field to the final one or two options, or it may provide you with
the best solution, in which case you skip the next step.
6. Conduct a Cost Benefit Analysis with Risk Adjusted Costs to Eliminate More Options. This
step is used to evaluate the highest scoring options considering the various risks that each
option contains. Depending on the size of the investment, it may be appropriate to conduct
an extensive risk analysis on the potential solutions before finalizing a recommendation.
7. Recommend Solution. The solution that scores the highest using the selection criteria and
weighting, and has an acceptable risk profile, is presented to the stakeholder who has the
authority to make the final decision. Sometimes this is the project manager, sometimes it is
the sponsor or customer. The backup documentation and a summary of the process should
be included with the recommendation.

The following examples demonstrate a relatively simple use of alternatives analysis, followed by
a more rigorous example that incorporates weighted criteria to analyze the options.

Evaluating Time, Cost, and Resource Constraints

Scenario: You are managing a project to develop a new company intranet site.
You are analyzing whether to use in-house resources or outsource for higher-skilled resources
that cost more per hour, and can accomplish the work faster. The relative importance of the
schedule constraints, cost constraints, and resource availability are the solution requirements.
Option 1 assumes you have a team member with adequate skills who can accomplish a
task in 80 hours. Her hourly rate is $45.
Option 2 assumes you can outsource the work and use a highly skilled resource that will only
need 65 hours to accomplish the work. The higher-skilled resource has an hourly rate of $60.
By multiplying the rate times the duration you determine the cost of the in-house resource is
$3,600. The cost of the outsourced, higher-skilled resource is $3,900.
If your solution requirements state that low cost is a higher priority than the schedule, or if
you prefer to keep the work in-house, you should go with Option 1. If time is of the essence and
you have room in the budget, you should go with Option 2.

Selecting a Cafeteria Management Service

Scenario: You are the project manager for a project to implement a childcare facility
for your organization’s employees.
This example focuses on selecting a vendor to provide the food service for the Childcare Center
cafeteria. Weighted solution (evaluation) criteria are used to evaluate vendors against a set of
requirements for the cafeteria services.
 2.1 Alternatives Analysis    21  

1. Define Solution Requirements. In this example, the requirements are:

• The service provider should have food options, with a preference toward fresh fruits and
vegetables that are locally sourced.
• A reliable vendor, as determined by references.
• Pricing that is consistent with the market. Should be able to provide a healthy lunch for less
than $5.
2. Define Evaluation Criteria and Weightings.

The selection criteria and weights are as follows:

• Healthy food options: 50 percent

• Locally sourced where possible: 15 percent
• Good customer feedback: 25 percent
• Competitive pricing: 10 percent

The scoring algorithms are:

Healthy Food 5 = 75% of the menu includes fruits, vegetables, grains, legumes, and lean meat.
25% or less is frozen or processed.
4 = 66–74% of the menu includes fruits, vegetables, grains, legumes, and lean
meat. 33% or less is frozen or processed.
3 = 50–65% of the menu includes fruits, vegetables, grains, legumes, and lean
meat. 50% is frozen or processed.
2 = Most of the menu incorporates fried, frozen, or processed foods.
1 = Limited fresh options. Almost all food is fried, frozen, or processed.
Locally Sourced 5 = More than 75% is locally sourced within 100 miles.
4 = 66–74% is locally sourced within 100 miles.
3 = 50–65% is locally sourced within 100 miles.
2 = Most of the food is transported from greater than 100 miles.
1 = Almost all the food is transported from greater than 100 miles.
Good Customer 5 = 5 out of 5 references report they are happy with the vendor and would contract
Feedback with them again.
4 = 4 out of 5 references report they are happy with the vendor and would contract
with them again.
3 = 3 out of 5 references report they are happy with the vendor and would contract
with them again.
2 = 2 out of 5 references report they are happy with the vendor and would contract
with them again.
1 = 1 out of 5 references reports he is happy with the vendor and would contract
with them again.
Price Lowest average price
Within 5% of lowest average price
Within 10% of lowest average price
Within 15% of lowest average price
Greater than 15% higher than lowest average price
(continued)