How to estimate the test workload
(1) conventional methods for estimating test workloadAs a manager, are you asked how long a project will take, how much human testing is required, or as a normal tester, are you asked how long it takes to complete a task or perform a regression test? I think most people in the software industry may have such questions about workload estimates. So how did you answer this question? Are you confident in your answers? Do you finally find that the actually spent time is very different from the originally estimated time? Different people use many different methods to estimate and arrange their testing workload. Different organizations use different methods based on the project type, internal risks of the project, and technologies involved. However, most of the time the test workload is combined with the Development workload, there is no separate number. First, let's take a look at some common methods to estimate the test workload:
1. Ad-Hoc
MethodThe testing workload in this method is not based on any fixed duration. The work continues until it reaches a certain schedule preset by management or marketing personnel. Or, until the budget is used up. This situation is generally found in very immature organizations, and there are often 100% errors.
2.
Development time percentage method percentage of development timeThe basic premise of this method is that the test workload depends on the development time/Development workload. First, the development workload is estimated using methods such as loc or FP, and then some exploratory methods are used to limit the test workload. This method has changed a lot and is generally based on previous experience. Usually the total time spent on the reserved Project
35
%To test. •
5-7%Testing components and integration •
18-20%Test the system •
10%Receive tests (or regression tests, etc)
3.
Analogy (empirical method or historical data method)Estimate the workload based on the accumulated experience or historical data of previous or similar projects (mainly in the project nature, field, and scale. The accuracy of the estimation result of analogy method depends on the integrity and accuracy of historical project data. Therefore, one of the prerequisites for using analogy is to establish a good post-project evaluation and analysis mechanism, data Analysis on historical projects is trustworthy. The following historical data needs to be collected: • time spent in design and implementation • Scale of testing, such as the number of user needs, number of pages, functional points • Data styles, such as entities, number of fields • number of screens or fields • Size of the test object, such as kloc
4. WBS
(Work breakdown structure
) Estimation MethodThe project or product is divided into specific work, and then the time of each work is estimated separately, and the test workload/time of the project or product is obtained.
5. Delphi
MethodDelphi method is the most popular expert evaluation technology. Without historical data, this method can reduce the estimation deviation. Participants are encouraged to discuss the issue through Delphi. This technology requires the participation of a variety of people with relevant experience to persuade each other .. Step 1 of the Delphi Method: 1. The focal point should provide the project specifications and estimation forms to experts; 2. The focal point should convene a group meeting of experts to discuss the factors related to the scale; 3. Fill in the iteration form anonymously by experts. 4. The coordinators should organize an estimation summary and return it to the experts in the form of repeated representatives. 5. The coordinators should convene a group meeting to discuss the large estimation differences; 6. Experts review the estimation summary and submit another anonymous estimate in the iteration table; 7. Repeat the 4-6 until a minimum and maximum estimation is reached.
6. Pert
Estimation MethodPERT estimates the completion time of each project activity based on three different situations: the expected scale of a product, the minimum possibility, and the maximum possibility. The three estimates are used to obtain an pert statistical estimate of the expected scale and standard deviation of a product. PERT estimates the expected values of the code line E and standard deviation SD.