Quelle est la meilleure façon de déterminer le nombre de scripts de test automatisés nécessaires pour votre projet ?

Jerry Weinberg knew the answer to questions like these. It was a very Zen reply. “The best number is three. “You disagree? Were you expecting a different number?”

When determining the number of automated test scripts needed for a software project, it's essential to consider the scope and complexity. Focusing on automating critical, risky, and frequently used features ensures thorough and consistent testing. We should prioritize test cases that are difficult, time-consuming, or tedious to perform manually, as automation can improve efficiency and accuracy. However, we should avoid automating unstable, unclear, or human judgment dependent test cases to minimize maintenance overhead. Additionally, we need to regularly monitor the performance of automated tests to make adjustments as needed. By following these guidelines, we can ensure that automation is used effectively and efficiently in the project.

None!! There is no 'Best' way to determine the number of test cases to automate. What should drive the decision includes the following: the regression test pack, test methodology in use, level of documentation available, the skillset available on the team, type of project, how much downtime is available to automate the tests etc. Each project differs hence there is no such thing as the best way to do anything. I can define the Best way as what I have done, that really worked for me. it might not work for the next person especially if our skillset and experience differs (and vice versa).

The answer is Test Automation ROI and maintenance. The reason is we shouldn’t automate all the regression test case, we should pick the test cases that are critical and repeatedly executing for each regression cycle

The tríade of scope, time and cost is crucial to determine the complexity and what tests its good in terms of cost to automatize.

With respect to strategy, In my experience ,it was very important on deciding what should be automated versus what can be automated & that depends on various factors such as feasibility, time-saving potential & the complexity of the task. Generally, routine, repetitive tasks with clear rules or patterns are good candidates for automation, I would recommend starting and prioritizing with covering Smoke/Sanity check first while tasks requiring creativity, critical thinking, or human interaction may be better suited for manual handling. Making sure that automated tests take lesser time than manual is what should be the goto approach CI/CD should be the part of your goal. Having said the above,always think about easier ways of accomplishing

One of the things usually Automation suffer with is the Test Data; I usually instruct the Automation team as much as possible to automate the creation of Test data whenever possible, Another thing is to reset your environment where it was before the Automation, E.g. If a user get blocked while running the script, after completion unblocking the user to where it was before Automation in preparation for next round.

While coming-up with Test automation strategy its important to consider following: 1. All kinds of tests be it functional or cross-functional like performance and security 2. Test data management 3. Tool selection 3. Completion criteria On framework level we need to ensure: 1. Its easy to write the tests 2. It should be modular 3. SOLID principles are followed 4. Reporting and logging are thought through 5. Guidelines on creating/organizing the tests/utilities etc.

During testing, an automate test strategy outlines how you'll use automation frameworks. Test cases of all kinds, automation scripts, test environments and their definitions, test data, test results, execution logs, and reporting output are all included in the process of preparing tests, both manually and automatically.

Define a clear test automation strategy aligned with project goals. Choose an appropriate automation framework that suits the project's needs. The framework's efficiency in handling test cases may impact the number of scripts needed.

Evaluate the available resources and expertise within the testing team. A skilled and experienced team may be able to develop and maintain a larger number of automated test scripts efficiently. Assess the learning curve associated with the selected automation tools and technologies.

The complexity and scale of your project, as well as other considerations like priority, core functionality, integration points, cross-browser and cross-platform compatibility, etc., will all affect how many automated test scripts you need.

Determining automated test scripts: Resource considerations: Assess: Test automation resources and skills: Hardware, software, tools, and environments. Technical, domain, and testing expertise. Evaluate based on: Availability, suitability, reliability, and scalability. Ensures: Realistic scope and achievable automation goals. Efficient utilization of resources and skills.

Assessing test automation resources and skills is crucial for determining capacity and quality. Resources encompass hardware, software, tools, and environments needed for script creation and execution. Skills include technical, domain, and testing knowledge for script design, development, and maintenance. Evaluate availability, suitability, reliability, and scalability of resources and skills to meet automation needs effectively. Address any gaps through training, resource allocation, or tool selection to ensure efficient and high-quality automation implementation. Regularly update and enhance resources and skills to adapt to evolving project requirements and technological advancements.

Assess the availability of resources, including skilled automation engineers and sufficient time for script development. Consider the expertise level of your team and their familiarity with the chosen automation tools. Foster a learning culture! Encourage continuous improvement in automation skills and explore collaborative approaches to script development and maintenance.

Creating automation scripts is half of the solution. The important part of automation scripts is to think long term, and see how they can be developed to make them simpler, cutting across optimal interfaces, and useful in the long term, for any other automation tester to pick up and work on it Developing scripts for every field of the UI, or developing 1 script for 4-5 screens at a time does not make sense Very essential to have moderation in automation script writing.

Usually I focus on having a full automated scripts for sanity testing that is repeated with every build. This can also be used as entry criteria for testing and can even be used as unit testing.

In the final step, estimate the return on investment (ROI) and maintenance costs of test automation, weighing them against benefits and risks. ROI represents the value gained from automation efforts compared to costs incurred. Value includes time, money, quality, and coverage improvements, while costs involve time, and complexity invested in automation creation and maintenance. Maintenance costs encompass ongoing expenses for updating, modifying, or fixing scripts due to project changes. Aim to maximize ROI and minimize maintenance costs by automating high-value, low-cost test cases while avoiding low-value, high-cost ones. Regularly reassess and adjust automation strategies to optimize ROI and streamline maintenance efforts over time.

In my experience ROI is the ratio of the value that you gain from your test automation efforts to the cost that you spend on them. Your value can include the time, money, quality, and coverage that you save or improve by automating your test cases. Your cost can include the time, money, effort, and complexity that you invest or incur by creating and maintaining your automated test scripts.

Consider the Return on Investment (ROI) associated with test automation. Evaluate the maintenance effort required for automated scripts over time. Strive for a balance between the number of automated scripts and the maintenance effort to maximize efficiency.

Beyond the core: Start small and scale gradually: Begin with a few high-impact test cases and expand your automation scope as you gain experience and resources. Focus on maintainability: Write modular, reusable, and well-documented test scripts that are easy to understand, update, and debug. Monitor and improve continuously: Regularly assess the effectiveness of your automated tests, identify areas for improvement, and adapt your strategy based on project changes and feedback. By embracing these considerations, you can establish a data-driven and sustainable approach to determining the number of automated test scripts needed for your project.

To create an effective test automation strategy, follow these best practices: 1. Automate repetitive, data-intensive, and frequently executed test cases. 2. Divide automated testing into smaller tests that can be reused and grouped into larger test suites. 3. Prioritize automation for a high percentage of automatable test cases. 4. Monitor test performance metrics to ensure automation meets project goals. 5. Choose the right tools and ensure your team has the necessary skills. 6. Create a test automation plan to guide your future tests.

Determining the ideal number of automated test scripts involves balancing factors like project complexity, risk areas, and resource constraints. Aim for adequate coverage while considering maintainability.

In addition , it's crucial to consider the application's architecture, technology stack, and supported platforms when determining the number of automated test scripts needed for your project. Understanding the dependencies, data flows, and integration points within the system will guide the identification of critical test scenarios. Furthermore, taking into account the impact of environmental configurations, data variations, and edge cases will help in creating robust and versatile automated test scripts. By prioritizing test coverage based on risk-based analysis, usability, and business-critical features, you can ensure that the automated test suite provides maximum value in validating the project's quality and functionality.

1)Scope And Requirements 2)Risk analysis 3)Test coverage 4) Reusability 5) Maintenance efforts 6) Tooling and frame work 7) Feedback and loop By considering these factors, you can determine the optimal number of automated test scripts needed for your project to ensure comprehensive test coverage while maximizing efficiency and effectiveness.