One of the fundamental aspects of PSP is the use of historical data to analyze and improve process performance. Data collection for PSP is supported by four important elements:

• - Scripts.

• - Metrics.

• - Standards.

• - Formats.

PSP scripts provide an expert-level guide to follow the steps of the process, the scripts provide a framework for applying measurements. In PSP there are four essential measurements:

• - Size – the size of a part of the product, measured in lines of code (LOC) or equivalent pieces of software (proxies) that facilitate measurement.

• - Effort – the time required to complete a task, usually measured in minutes.

• - Quality – the number of defects in the product.

• - Agenda – a measurement of project progression, comparison of what was planned against current completion dates.

Applying standards to the process can ensure that data is accurate and consistent. The data is recorded in formats, frequently recorded in applications to measure PSP.

Software developers use other measures, which are derived from the essential ones, to understand their performance. Among the derived measures are:

• - Precision estimation (size/time).

• - Interval prediction (size/time).

• - Time in the distribution phase.

• - Defect injection distribution.

• - Distribution of defect removal.

• - Productivity.

• - Rejection percentage.

• - Performance cost index.

• - Planned value.

• - Earned value.

• - Predicted earned value.

• - Density of defects.

• - Density of defects per phase.

• - Defect removal rate per phase.

• - Defect removal leverage.

• - Revision fees.

• - Process yield.

• - Performance (yield) of the phase.

• - Cost of quality (COQ) failure.

• - Evaluation (appraisal) COQ.

• - Evaluation/failure rate COQ.

Recording times, defects, and sizes is essential for planning and tracking PSP projects; historical data is used to improve estimation accuracy.

PSP uses the PROBE method to improve developers' estimation skills for more accurate planning. To track the project PSP uses the earned value (EV) method.

PSP also uses statistical techniques, such as correlation, linear regression, and standard deviation, to translate data into useful information to improve estimating, planning, and quality. Statistical formulas are calculated by the PSP tools.

Producing high-quality software is the goal of PSP, and quality is measured in terms of defects. For PSP, a quality process should produce low-defect software that meets user needs.

PSP allows developers to find defects early. By finding them early, PSP reduces the amount of time spent in later phases such as the testing phase.

According to PSP it is more economical and effective to remove defects as soon as possible. Software engineers are encouraged to conduct personal reviews for each phase of development. Therefore PSP includes two review phases:

• - Design review.

• - Code review.

To conduct an effective review, you need to follow a structured review process. PSP recommends using checklists to help developers follow an orderly procedure.

PSP follows the premise that when people make mistakes, their mistakes are usually predictable, so PSP developers can customize their checklists to review their own mistakes. Software engineers are also expected to make improvement proposals, to identify weaknesses in their current performance, focusing on correcting them. Historical project data, which exposes where time is spent and defects introduced, helps developers identify areas for improvement.

PSP is a personal process that can be adapted to the individual needs of each developer. PSP is not specific to a programming language or design methodology; therefore it can be used with different methodologies including Agile software development").

Software engineering methods can range from predictive to adaptive. PSP is a predictive methodology, Agile development is considered an adaptive methodology, but despite their differences, TSP/PSP and Agile development share several conceptual approaches – particularly regarding team organization. With both it is possible:

• - Define your goals and standards.

• - Estimate and schedule work.

• - Determine realistic and achievable agendas.

• - Carry out improvement plans and processes.

Agile Development and TSP/PSP share the idea that team members take responsibility for their own work and work together to agree on a realistic plan, creating an environment of trust and responsibility. However, TSP/PSP differs from Agile development in its emphasis on documenting the process and using data to predict and define the project schedule.

Compared to UML, PSP obtains information from dynamic and static, internal and external interaction by capturing data with formats that resemble the formats of use cases, sequence diagrams, and classes. Based on a UML diagram, the base information can be obtained to create certain PSP formats.

The Software Engineering Institute (SEI), at Carnegie Mellon University offers a certification in PSP. The steps to become a certified PSP developer are: learn PSP, take the certification exam, maintain credentials.[2]​.

• - PSP: A Self-Improvement Process for Software Engineers 2005.

• - Official page.

• - Process Dashboard, tool to use PSP.

One of the fundamental aspects of PSP is the use of historical data to analyze and improve process performance. Data collection for PSP is supported by four important elements:

• - Scripts.

• - Metrics.

• - Standards.

• - Formats.

PSP scripts provide an expert-level guide to follow the steps of the process, the scripts provide a framework for applying measurements. In PSP there are four essential measurements:

• - Size – the size of a part of the product, measured in lines of code (LOC) or equivalent pieces of software (proxies) that facilitate measurement.

• - Effort – the time required to complete a task, usually measured in minutes.

• - Quality – the number of defects in the product.

• - Agenda – a measurement of project progression, comparison of what was planned against current completion dates.

Applying standards to the process can ensure that data is accurate and consistent. The data is recorded in formats, frequently recorded in applications to measure PSP.

Software developers use other measures, which are derived from the essential ones, to understand their performance. Among the derived measures are:

• - Precision estimation (size/time).

• - Interval prediction (size/time).

• - Time in the distribution phase.

• - Defect injection distribution.

• - Distribution of defect removal.

• - Productivity.

• - Rejection percentage.

• - Performance cost index.

• - Planned value.

• - Earned value.

• - Predicted earned value.

• - Density of defects.

• - Density of defects per phase.

• - Defect removal rate per phase.

• - Defect removal leverage.

• - Revision fees.

• - Process yield.

• - Performance (yield) of the phase.

• - Cost of quality (COQ) failure.

• - Evaluation (appraisal) COQ.

• - Evaluation/failure rate COQ.

Recording times, defects, and sizes is essential for planning and tracking PSP projects; historical data is used to improve estimation accuracy.

PSP uses the PROBE method to improve developers' estimation skills for more accurate planning. To track the project PSP uses the earned value (EV) method.

PSP also uses statistical techniques, such as correlation, linear regression, and standard deviation, to translate data into useful information to improve estimating, planning, and quality. Statistical formulas are calculated by the PSP tools.

Producing high-quality software is the goal of PSP, and quality is measured in terms of defects. For PSP, a quality process should produce low-defect software that meets user needs.

PSP allows developers to find defects early. By finding them early, PSP reduces the amount of time spent in later phases such as the testing phase.

According to PSP it is more economical and effective to remove defects as soon as possible. Software engineers are encouraged to conduct personal reviews for each phase of development. Therefore PSP includes two review phases:

• - Design review.

• - Code review.

To conduct an effective review, you need to follow a structured review process. PSP recommends using checklists to help developers follow an orderly procedure.

PSP follows the premise that when people make mistakes, their mistakes are usually predictable, so PSP developers can customize their checklists to review their own mistakes. Software engineers are also expected to make improvement proposals, to identify weaknesses in their current performance, focusing on correcting them. Historical project data, which exposes where time is spent and defects introduced, helps developers identify areas for improvement.

PSP is a personal process that can be adapted to the individual needs of each developer. PSP is not specific to a programming language or design methodology; therefore it can be used with different methodologies including Agile software development").

Software engineering methods can range from predictive to adaptive. PSP is a predictive methodology, Agile development is considered an adaptive methodology, but despite their differences, TSP/PSP and Agile development share several conceptual approaches – particularly regarding team organization. With both it is possible:

• - Define your goals and standards.

• - Estimate and schedule work.

• - Determine realistic and achievable agendas.

• - Carry out improvement plans and processes.

Agile Development and TSP/PSP share the idea that team members take responsibility for their own work and work together to agree on a realistic plan, creating an environment of trust and responsibility. However, TSP/PSP differs from Agile development in its emphasis on documenting the process and using data to predict and define the project schedule.

Compared to UML, PSP obtains information from dynamic and static, internal and external interaction by capturing data with formats that resemble the formats of use cases, sequence diagrams, and classes. Based on a UML diagram, the base information can be obtained to create certain PSP formats.

The Software Engineering Institute (SEI), at Carnegie Mellon University offers a certification in PSP. The steps to become a certified PSP developer are: learn PSP, take the certification exam, maintain credentials.[2]​.

• - PSP: A Self-Improvement Process for Software Engineers 2005.

• - Official page.

• - Process Dashboard, tool to use PSP.