User:Rigtermi/sandbox

Introduction History TPI was launched in 1998 by Sogeti to provide a reference model and method to enhance test efficiency and effectiveness and a permanent improvement cycle. The first edition of the model was enhanced into TPI NEXT® in 2009, giving it a more business driven angle and an appropriate fit to changing IT environment and technology.

The importance of test process improvement Over the past decades the testing of software and systems has become a profession in its own right and has become an acknowledged discipline by both business and IT. At the same time the complexity of software and systems, stand-alone or integrated over multiple platforms and organisations, urges testing to become more and more effective and efficient, providing stakeholders the right decision making information. Another concern is the growing alignment between IT strategy and business goals and drivers, requiring the test process and test process improvement to consider and adapt to this context.

The essentials of the TPI Model The TPI (NEXT) model comprises 7 elements: •	Key areas •	Maturity levels •	Checkpoints •	Maturity Matrix •	Clusters •	Improvement suggestions •	Enablers All elements contribute to establishing a current process maturity and define measures and a roadmap for improvement towards more efficiency. The following sections explain the 7 elements.

Key areas The model consists of 16 Key areas that describe a group of test activities, procedures and methods. 1.	Stakeholder commitment, describing the role and influence of the different stakeholders; 2.	Degree of involvement, referring to the starting point and the relating activities; 3.	Test strategy, is about the way test effort, activities and resources are allocated; 4.	Test organization, dealing with aspects of responsibilities and accountabilities; 5.	Communication describes the way the different entities communicate in a multidirectional way; 6.	Reporting holds for the one-directional way of communication; 7.	Test process management is about the way the process is managed, steered and controlled; 8.	Estimating and planning measures the way projects are estimated and planned; 9.	Metrics are quantified and objective observations of the characteristics of a product or process; 10.	Defect management indicates the way defects are administrated, managed and communicated; 11.	Testware management ensures coherence among test artefacts and between test artefacts and their related design documents; 12.	Methodology practice measures the way a certain test method is described and applied; 13.	Tester professionalism includes the right mix the various skills and competences, disciplines, functions and knowledge; 14.	Test case design defines the methodical approach to derive and design test cases and scripts; 15.	Test tools measures the maturity of the applied tools to automate test activities; 16.	Test environment deals with the aspects of availability, maintenance, responsibilities and representativeness.

Maturity levels Each test process and each Key area can be classified into a certain level of maturity. Starting at an Initial level a test process can develop from Controlled through Efficient towards Optimizing, each level being more mature than the preceding one: •	Initial: referring to ad-hoc activities, rather unstructured and typically depending on one or more key individuals; •	Controlled: doing the right things, using the right instruments to steer and manage the process; •	Efficient: doing things the right way, having a good balance between cost (including budget, material and people) and effort; •	Optimizing: continuously adapting to ever-changing circumstances by evaluating and redesigning the Key area or process.

Checkpoints The TPI NEXT® model consists of 157 checkpoints, divided equally among the Key areas and maturity levels (except for the Initial level). The checkpoints address what (rather than the how) needs to be achieved in order to obtain a certain level of maturity. Each Key area contains 8 – 11 checkpoints and each maturity level has 2 - 4 checkpoints per Key area and, respectively, 59 (Controlled), 57 (Efficient) and 41 (Optimizing)checkpoints for the maturity levels. The checkpoints can either be met or not: there are no options in between.

Maturity matrix The Test Maturity Matrix provides an overview of all Key areas, together with their respective maturities. The Matrix lists all 16 Key areas in the horizontal rows and the Maturity levels (Initial, Controlled, Efficient and Optimizing) in the columns. Below is an example of an assessment score: the figure shows which checkpoints have been met (green)or not (white). Stepwise improvement with Clusters A Cluster is a group of checkpoints from multiple Key areas that make up one (small) improvement step and, ordered in an alphabetical sequence, create an improvement path and roadmap with predefined priorities. The checkpoints have been clustered in such a way that interdependencies and priorities (one checkpoint being more relevant or contributing to more maturity) have been taken into account. The illustration below shows the clusters from A – M (note that each checkpoint is allocated to one specific cluster). In this example the checkpoints in cluster A are either black (checkpoints have been met) or white (checkpoints not fulfilled). Business driven clustering The clusters make it possible to adapt the model to specific circumstances. These circumstances may vary from specific business driven demands (shorten time to market, cutting the cost of testing or making the process more transparent) or to agile environments, outsourcing test activities, etc. These circumstances may require other priorities to Key areas and checkpoints, one Key area contributing more to the situation or business driver than the other. By setting these priorities (high, neutral or low) the checkpoints may shift towards a succeeding or preceding cluster: a checkpoint from cluster A moves to cluster B or vice versa. The example below shows two different situation for ‘Stakeholder commitment’. The first is the normal, generic situation with the basic cluster priorities are set to N (neutral): In the next example the Key area Stakeholder commitment has been given higher priority (indicated by the X in column H), shifting the checkpoints from cluster B to A, C to B, F to E, H to G, etc. This indicates that the first step of improvement (being cluster A) contains two extra checkpoints: Improvement suggestions In a way the checkpoints that were not fulfilled can be used as an improvement suggestion to move the process to a higher level of maturity. The checkpoints in that case need to be ‘translated’ to the specific situation. The TPI NEXT® book (see below) contains an extra number of improvement suggestions, based on best practices and applied in different situations by authors and contributors.

Enablers A test process is not a standalone activity but closely related to and dependant on other processes in the Software Development Life Cycle (SDLC). These processes can benefit from exchanging best practices and working close together. This relationship is translated into ‘enablers’: connecting Key areas from the model with aspects of the SDLC in order to keep test process improvement aligned with other processes from the SDLC. Such an enabler may be the quality of requirements and design documents or the quality of project management; they very strongly affect the test process in many different ways.

Further reading and support The TPI NEXT model and the way it can be used is described in the book TPI NEXT® Business Driven Test Process Improvement. Tuthein Nolthenius, 2009. ISBN 9789072194978, written by Gerrit de Vries, Ben Visser, Loek Wilhelmus, Bert Linker and Alexander van Ewijk. Articles about TPI NEXT®, white papers on specific issues and tools to support assessments and quick scans can be found at www.tpinext.com