Estimation techniques play an important role in executing a task in software development life cycle (SDLC). Going by the definition, estimation approximates the effort that would be required to complete a task. However, the effort can be measured in terms of time and cost. In order to calculate the test estimation, you can consider the past experience, data, knowledge, documents, and your assumptions.
By estimating the task, you can save a lot of time and cost, which is otherwise wasted in non-estimated techniques. The following points prove to be helpful, if you consider them before estimating testing activities.
- Verify whether all requirements are complete or not.
- Check how frequent the requirements are going to change.
- Ensure that there are no remaining dependencies and responsibilities.
- Check the testing infrastructure
- Document all risks and assumptions before estimating a task.
There are different techniques that can be used to estimate a task. Some of them are discussed below.
Delphi Technique
It uses survey to collect information from experts. This is one of the most common ways to estimate a task. Here, each team member is assigned a task with multiple rounds of surveys. Based on the information collected in each round, feedback is provided. The surveys are conducted until a final estimation is not gathered. This technique can also be used along with other techniques.
Work Breakdown Structure (WBS)
A project is more manageable if it is broken down into multiple components. The structure caters the need to capture important task based on the requirements. If the project is complex, the individual components can be further broken down into sub-modules in a hierarchical structure.
Going further, sub-modules are further divided into functionalities. The individual functionality is again divided into sub-functionalities. Finally, you must ensure that every broken component is covered in the WBS.
By breaking down a complex project into manageable parts, you can easily estimate a given task.
Three Point Estimation
Three types of estimations are performed based on the statistical data.
Optimistic Estimate: Everything is correct (A)
Most Likely Estimate: Most of the things are correct (B)
Pessimistic Estimate: Everything is wrong (C)
Value for Estimate (E) = A + (4*M) + B/6
Standard Deviation (D) = (B-A)/6
Functional Point Method
Estimation is made based on available documents, designs, and involves user considerations as well. However, it is not related to the language, technology, or methodology used by the team. Here, the importance is given to each functional point by dividing them into three categories: Simple, Complex, and Medium.
Total Estimate = Total Functional Points * Estimate per functional point
