Session 2.2 - Setting Team Goals

Chapter 3: Launching a Team Project | Duration: 1 hr

Learning Objectives

By the end of this session, you will be able to:

  • Describe why measurable goals anchor strategy and planning
  • Apply aggressive-but-realistic goal-setting principles
  • State the three standard TSPi team goals and associated measures
  • Outline steps to adapt goals for future cycles

3

Basic TSPi team goals: Quality, Management, Schedule

Measurable

Goals must be specific and quantifiable for tracking

Aggressive

Challenging but realistic targets motivate teams

Iterative

Refine goals each cycle based on data

Introduction

Goal setting is an essential step in team formation and should usually be done at the start of every project. Establishing goals is simple in concept but often difficult in practice. The reason is that goals should be precisely measured, and few of us are in the habit of being precise about our work.

"The goals establish the framework for the strategy and the plan. These, in turn, provide the foundation for everything that the team will subsequently do. With agreed-on goals, you have the basis for evaluating strategies and settling issues. Without defined goals, there is no orderly way to settle arguments, negotiate strategies, or plan the work."

— Chapter 3, Section 3.2

Although the lack of specific goals may not be a problem in most situations, it can be a problem for software teams. This session explores how to set effective team goals that guide strategy, planning, and execution.

Why Goals Matter

Goals create the framework for strategy and planning. Without explicit goals, teams lack a way to evaluate options or resolve disagreements. Chapter 3 stresses that aggressive, measurable goals motivate higher performance and allow objective tracking. Teams with clear goals consistently outperform those without.

Goal-Setting Considerations

The primary objective of most projects is to attain superior performance, something that is most likely when people strive to meet challenging goals. Understanding the balance between aggressive and realistic goals is critical for team success.

The Goal Sweet Spot
Goal Difficulty / Challenge Level Team Motivation & Performance Too Easy Low motivation No challenge SWEET SPOT Aggressive but Realistic ✓ High motivation ✓ Team jelling ✓ Best performance Impossible No motivation Demoralization Optimal

Figure: The relationship between goal difficulty and team motivation

Aggressive vs. Realistic
  • Too easy → Little motivation to strive; team doesn't jell
  • Clearly impossible → No motivation; creates demoralization
  • Sweet spot: Stretch but believable; drives performance
  • Better approach: Aggressive goals you occasionally miss than safe goals you always make
Focus on Effort
  • Teams evaluated on willingness to set aggressive goals
  • Evaluated on efforts to meet them, not just achievement
  • Avoid "safe" goals just to guarantee success
  • Learn by setting goals, measuring results, and adjusting
Key Principle

When you set goals, make them aggressive but realistic, and then strive to meet them. It is more important to have aggressive goals that you occasionally miss than safe goals that you always make.

Important: Teams should not be measured on whether or not they actually meet their goals. Instead, they should be evaluated on their willingness to set measurable and aggressive goals and on their efforts to meet them. Too much emphasis on meeting goals is counterproductive because it motivates teams to set goals that they are sure they can meet.

Standard TSPi Team Goals

The general guideline for setting goals is to first consider what would be a superior result in the eyes of your customer. In TSPi, the instructor is your customer. For TSPi, there are three basic goals:

Three Basic TSPi Team Goals
1 Goal Produce a Quality Product Key Measures: • % defects found before first compile: 80% • Defects in system test: 0 • Requirements functions included at completion: 100% 2 Goal Run a Productive Well-Managed Project Key Measures: • Error in estimated product size: <20% • Error in estimated dev hours: <20% • % data in notebook: 100% 3 Goal Finish On Time Key Measures: • Days early or late in completing the dev cycle: <4 days • Milestone adherence: 90%+

Figure: The three standard TSPi team goals with example measures from Figure 3.1

"These goals may seem obvious, and they are very general. The principal challenge of goal setting is to make the goals measurable. So the second goal-setting step is to define measurements for the goals."

Goal Description & Purpose Example Measures (Figure 3.1)
Goal 1: Quality Product Deliver a product that meets all requirements with minimal defects. This is the primary customer-facing goal.
  • Measure 1.1: Percent of defects found before first compile: 80%
  • Measure 1.2: Number of defects found in system test: 0
  • Measure 1.3: Requirements functions included at project completion: 100%
Goal 2: Productive, Well-Managed Project Use disciplined planning and tracking to guide the team effectively and make accurate estimates.
  • Measure 2.1: Error in estimated product size: <20%
  • Measure 2.2: Error in estimated development hours: <20%
  • Measure 2.3: Percent of data recorded and entered in project notebook: 100%
Goal 3: Finish on Time Meet the planned schedule for the cycle within acceptable variance.
  • Measure 3.1: Days early or late in completing the development cycle: <4
  • Measure 3.2: Milestone hit rate: 90%+
  • Measure 3.3: Weekly variance from plan: <15%
Making Goals Measurable

The best way to make goals measurable is to measure your performance and set goals to improve. To do this, however, you need data on your prior work.

Because few teams will have historical data before the first development cycle, TSPi provides these standard measures as a starting point. After cycle 1, you will have data for setting new goals for cycles 2 and 3.

Measuring Success

With no history, teams start with the standard measures from Figure 3.1. After cycle 1, use collected data to reset targets for cycles 2 and 3 based on your team's actual performance.

The Goal-Setting Cycle
Cycle 1 Use standard TSPi goals (Figure 3.1 measures) No historical data yet Measure & Collect Track actual performance Time, size, defects Quality, schedule data Analyze Results Compare plan vs. actual Identify gaps Find improvement areas Cycle 2 & 3 Set new goals based on Cycle 1 data More aggressive targets Process Changes Merely setting improvement goals won't produce better results without changing HOW you work Continuous Improvement Through Data-Driven Goal Setting Each cycle builds on the lessons learned from previous cycles

Figure: Iterative goal-setting process across development cycles

What to Measure
  • Quality: Defects found per phase; inspection yield; test defect density
  • Effort: Plan vs. actual time by task and by phase
  • Size: Estimates vs. actual (LOC, stories, or points)
  • Schedule: Milestone adherence (on-time / late)
  • Process: Data completeness; script adherence
Using the Data
  • After cycle 1, examine your performance data
  • Set improvement goals for cycles 2 and 3
  • If you didn't meet goals and they're still reasonable, use them again
  • If goals look unrealistic, set less challenging (but still improving) goals
  • If you met goals, set aggressive improvement goals
Critical Point: Process Changes Required

As you strive to improve performance for cycles 2 and 3, you will have to work differently than before. Merely setting improvement goals and then working in the same way will not produce consistently better results. With each goal, consider what you will have to do differently to achieve it. Then figure out how to change your process so that you are likely to meet your new objectives. Then make the needed changes in your process scripts, personal practices, or team standards.

Adapting Goals

Chapter 3 outlines five specific steps to modify the standard goals once data is available from your first development cycle. This systematic approach ensures goals are thoughtful, measurable, and communicated to all stakeholders.

Five Steps to Adapt Goals
1 Write down the goals you wish to use instead. Be specific and clear about what you want to achieve in the next cycle 2 Specify how to measure these goals. Define concrete metrics, target values, and measurement methods 3 Describe why you have selected these goals instead of the ones provided by TSPi. Justify based on Cycle 1 data and team's specific improvement needs 4 Give a copy of your revised goals to the team and to the instructor. Ensure buy-in and accountability from all stakeholders 5 Have the team leader put a copy of your goals in the project notebook. Document for future reference and process improvement

Figure: The five-step process for adapting TSPi goals based on Chapter 3

When to Adapt Goals
  • After Cycle 1: You now have data to inform realistic targets
  • Goals too easy: If you easily met them, make them more aggressive
  • Goals unrealistic: If clearly unachievable, adjust downward (but still improving)
  • Goals still reasonable: If you didn't meet them but they're fair, keep them
Remember: Process Changes
  • New goals require new approaches
  • Identify what must change in your process
  • Update scripts, practices, or standards
  • Doing the same work the same way won't produce better results
Example: Adapting a Quality Goal

Cycle 1 Result: Your team found 65% of defects before first compile (goal was 80%).

Analysis: Most defects came from rushing code without personal reviews.

Adapted Goal: Find 75% of defects before first compile in Cycle 2.

Process Change: Add mandatory 15-minute personal code review checklist before any code is compiled.

Justification: More realistic than 80% given our experience, but still aggressive improvement from 65%. The process change addresses the root cause.

Summary

Key Takeaways
  • Goals establish the framework for strategy and plan
  • Without defined goals, no orderly way to settle arguments or plan work
  • Goals must be specific, measurable, and tracked
  • Aggressive but realistic goals motivate higher performance
What You Should Know
  • The three basic TSPi team goals and their measures
  • Why teams are evaluated on effort, not just achievement
  • How to use Cycle 1 data to adapt goals for Cycles 2-3
  • The five steps for modifying standard goals
Goal-Setting Principles Summary
Measurable Goals must be specific and quantifiable ✓ Defects < 5/KLOC ✓ Schedule ± 4 days ✓ Estimates < 20% error ✗ "Do better" ✗ "High quality" ✗ "Work harder" Aggressive Challenging but realistic targets Better to miss aggressive goals than always hit safe ones ✓ Motivates team ✓ Drives jelling ✓ Improves performance Data-Driven Use cycle 1 data to set cycles 2-3 goals 1. Measure actual 2. Analyze gaps 3. Set new targets 4. Change process 5. Track improvement Process-Linked New goals require process changes Can't improve results by working the same way Must change: • Scripts • Practices • Standards

Figure: Four key principles for effective goal setting in TSPi

"The only way to learn how to establish goals is to set specific and measured goals and then work to meet them. Your initial goals may aim at the wrong items or may be wildly unrealistic. With a little goal-setting experience, you will begin to see how to set aggressive but realistic goals."

— Chapter 3, Section 3.2

Next Steps

After learning about team goals:

  1. Review Figure 3.1 in Chapter 3 for complete goal measures
  2. Understand that you'll use standard goals for Cycle 1 (no historical data)
  3. Prepare to track time, size, defects, and quality data throughout Cycle 1
  4. Plan to analyze your data after Cycle 1 to set improved goals for Cycles 2-3
  5. Remember: Focus on effort and learning, not perfection
Previous Session Course Home