Ten Basic Principles of software R & D personnel assessment

The assessment of software developers has always been a difficult point in software enterprise management. In my long-term R & D management practices and consulting practices, I have summarized some basic principles for assessing software developers, sort out and share with you:

1. Reflect the company's values

What type of personnel does the company's values reflect? Who should I retain? What do we advocate? The recognition of these personnel can be implemented through specific assessment methods. For example, enterprises should encourage them to accumulate rich experience in a certain field, and encourage them to conduct in-depth research in a certain technical direction. They should have specific incentives for these behaviors. Therefore, when defining the assessment method, you must first consider which values should be reflected in the enterprise.

2. More work and more work, both quality and quantity

Do not let those who have completed a lot of hard work suffer, otherwise they will fight against the enthusiasm of those who really work hard. The Implementation of the Principle of more work and more work is based on the computing workload. Standard management is based on dimensions, processes, and measurements. To achieve more work and more work, we need to measure the workload well. If we only pay attention to the workload rather than the quality of work, it is obviously incorrect, data can be obtained through multiple channels, such as the number of defects found and customer feedback. Of course, the premise is that the Team has achieved the goal. If the goal is not completed, more work may not be done.

3. Encourage innovation and standardized management

Management and innovation are two wheels for the development of software enterprises. Standardized management can ensure the normal development of enterprises and realize the rapid development of enterprises through innovation, management provides the foundation for innovation to be transformed into productivity, and innovation can quickly improve the competitiveness of enterprises. Therefore, the assessment measures should reflect the recognition of the two companies. Some enterprises have set up innovation funds to reward those technical innovation and management innovation. Some enterprises have added support for process improvement work in the assessment indicators of R & D personnel.

4. Encourage reuse of technologies

Successful software companies must invest more in personnel, technology, and process. Software reuse is one of the most effective ways for software companies to increase their productivity. To establish an organizational-level technical reuse system within an enterprise, first, we should encourage everyone to actively extract reusable components and actively contribute reusable components. Such actions to extract reusable components should be rewarded based on the benefits they may bring.

5. change from time to time, but maintain continuity as much as possible

The assessment methods are tailored and time-limited. With the changes in the company's internal and external environment and the gradual stability of the company's culture, the assessment methods should be adjusted gradually, when changing the assessment method, you must ensure the continuity of the assessment method. Do not change too much. Otherwise, the people to be assessed will be at a loss and have a wait-and-see mentality, or it may take a lot of time to study the assessment methods, resulting in unnecessary productivity drops.

6. Integration of quantification and non-quantification

Without quantitative assessment indicators, all rely on non-quantitative indicators. For developers, it is difficult to reflect the principle of more work and more, and it is easy to go to the "eat big pot" model, developers cannot be mobilized to be positive. It is also difficult to fully quantify the workload. in the development process, there is a lot of work that is difficult to quantify. For example, it is difficult to quantitatively calculate the workload. Therefore, some non-quantitative indicators are allowed on the basis of quantification as much as possible during the assessment. The proportion of the two users can be determined based on the current enterprise management level. For enterprises with relatively standardized management and relatively high maturity, more quantitative indicators can be used, with a larger proportion.

7. Different positions must be differentiated.

For project managers, demand analysts, designers, programmers, testers, and quality management personnel, the characteristics of work nature, capability requirements, and performance are quite different, therefore, it is necessary to distinguish between waiting and waiting. This helps to reflect the internal and external fairness of the assessment method. For example, for quality management personnel, most of them are daily transactional work, and their performance is long-term. Their focus is on the generation of pre-defense defects, it is difficult to use quantitative data. We can consider using improvement rates and other indicators for assessment. The main task of programmers is to implement design, task scale and their work efficiency and quality can be quantified. The two types of assessment methods should be different.

8. Ensure that the assessed person's right to know in a timely manner

The assessment method should be notified to the person to be assessed in advance, and the assessment result should be promptly notified to the person to be assessed. The purpose of the assessment is to find ways to improve work performance, motivate employees to work harder, and the assessment methods also represent the company's values. Therefore, the assessed persons should be clear about the assessment methods, let them know what should be done, so that they can play an incentive role. The assessment results should be promptly notified to the persons to be assessed so as to give them a prompt positive or negative stimulus signal.

9. not based on the data provided by the person being assessed

If the data provided by the assessed person is used as the basis for the assessment, data distortion will occur. When implementing developer personal logs in software enterprises, the biggest problem is log distortion. Why? Because developers are worried that the logs they enter will become the basis for their assessment and the basis for evaluating their work effort, they will instinctively prefer to fill in their own workload at full capacity.

10. The assessment indicators should be directly related to the person to be assessed. The person to be assessed can play an important role in achieving the assessment indicators.

In many software companies, employee assessments are often linked to the company's profits, Department profits, or project profits. For sales departments, business departments, or other departments directly related to the market, this kind of assessment has an incentive effect. For R & D personnel, the incentive effect of this method is not so obvious. There are many reasons for the formation of profit. Most of the reasons may not be determined by developers. It is unreasonable to hook the factors determined by developers with their assessment. Even if developers try to work harder, it cannot help the formation of profits. Why should it be linked to profits?

The ancients cloud: easy to learn. The principle is very simple. Implementation involves all aspects of the enterprise. There are historical reasons, real problems, and future uncertainties. However, none of these should be the reason for evading the question of assessment, you must try to solve this problem!

Note: I am not liable for any of the above content from the Internet.
Article transferred from: http://developer.51cto.com/art/200711/59926.htm