Data in blue italics is completely fictitious and is there for guidance.
Project Metrics   Reference: This template is closely modeled after the “PSP Project Plan Summary” form provided in Introduction to the Personal Software Process by Watts S. Humphrey (1997).
Metric[1] Plan Actual Cumulative[2]     Sanity Checks
Minutes / LOC 8 9 6.63636364  
LOC / Hour 7.5 6.666666667 9.04109589  
Defects / KLOC 80 51.36363636 59.0909091  
Pre-Compilation Yield %[3] 62.5 73.45132743 62.3076923  
Appraisal / Failure Ratio[4] 0.3 0.353658537 0.29508197   0.353658537
Program Size (LOC) Plan Actual Cumulative  
Total New & Changed 1500 2200 55000  
Maximum Size 1650  
Minimum Size 1350  
Time in Phase (minutes)[5] Plan Actual Cumulative Cumulative %
Requirements 2500 3000 75000 20.54794521  
Design - creation 1500 3000 42500 11.64383562  
Design - review / rework 500 900 17500 4.794520548  
Implementation - coding 1500 2700 50000 13.69863014  
Implementation - code review / rework 1000 2000 27500 7.534246575  
Implementation - code build / rework 1000 1500 27500 7.534246575  
Test 4000 6700 125000 34.24657534  
Total 12000 19800 365000 100   12000
Maximum Time 13200  
Minimum Time 10800  
Defects Injected[6] Plan Actual Cumulative Cumulative % Defects / Hour
Design - creation[7] 50 40 1375 42.30769231 1.375
Implementation - coding 70 73 1875 57.69230769 1.666666667
Test  
Total 120 113 3250 100  
Defects Removed Plan Actual Cumulative Cumulative % Defects / Hour
Design - review / rework 50 50 1375 42.30769231 1.375
Implementation - code review / rework 25 33 650 20 0.371428571
Implementation - code build / rework 25 20 650 20 0.371428571
Test 20 10 575 17.69230769 0.276
Total 120 113 3250 100  
[1]
The values captured in this top-most section show the real value-add for this type of data tracking.  These values provide very useful indicators of performance improvements or performance regressions as the arrow of time moves forward for your organization.
[2]
Running total (or calculations based on running totals) across all projects that have been completed including all work to-date on the current project
[3]
100*(pre-compilation defects removed) / (pre-compilation defects injected)
[4]
= appraisal_cost_of_quality / failure_cost_of_quality
appraisal_cost_of_quality = (100 * total_review_time) / total_development_time
failure_cost_of_quality = (100 * (implementation_build_time + test_time)) / total_development_time
[5]
Taking these measurements precisely and accurately will require discipline.  Software engineers must meticulously record how they are spending their time in time logs.
[6]
Detailed defect logs must be maintained throughout the software development process life cycle in order to capture the raw data required by this worksheet.
[7]
These are the defects that are found either during the design review or during coding / code review / code build that are attributed to problems with the design.

No part of this work should be produced or used without the permission of the authors: Michael Turner and Dr. Sharon A White.