[1/3] Developer productivity (Category Productivity)

In a large company, it is often difficult to understand how effective an organization is. Of course, you can focus on different financial indicators, including Cost to Income Ratio, but they have two features: they are very general and difficult to understand the contribution of individual parts of the organization, and they are very late in time. Therefore, high-level managers usually want a more effective tool. When I thought about it, I realized that it was more convenient for me to build a model first and narrow down my thinking. I got a run-change-disrupt model.

• Run These are the operational processes of how we do business as usual.  - Change These are the processes that are aimed at changing the situation: creating new products, developing existing products.  - Disrupt processes related to internal startups, new business models and transformations

For system improvements, we have to work in every area, but today I wanted to focus on Change. Moreover, if the company is a product, not a service, then in turn the product process is also divided into two parts.  - Do the right things This is part of the business and discovery process.  - Do the things right Part about the development and delivery process

Today, I wanted to discuss a part of it. deliveryIt seems to me that there are usually two dual approaches to improving the efficiency of the supply of value.

1. Bottom-up Improvement of team processes in the field by line managers. Further improvement of the overall processes of the division by middle managers. General drive for the best from top managers. For this approach, companies often use tools for statistical process management. This approach is similar to how microeconomic theory works for an individual enterprise in the marketplace. That is, the head of each team is responsible for ensuring that it works effectively in the conditions of supply and demand in the market, internal processes of the team, the chain of relationships with other teams and so on.

2. Top-down Improve processes by changing the global parameters of the system:  Creating PaaS and a large number of platform teams  Selection of SRE function and system work on reliability  QA strategy with a focus on automation  Data strategy and transition to DWH to Data Mesh Active use of AI to place and not For such big changes, the top manager usually wants to understand the centralized effect and ideally wants to be able to model it. This is somewhat reminiscent of macroeconomic theory and the role of the state in it, which can influence the market through refinancing rates, taxes, government spending and so on. Companies often do not have models for collecting “macroeconomic statistics” on delivery, apart from individual indicators at the microeconomic level, which cannot be directly compared and often aggregated. But often I would like to have a developer productivity indicator and be able to influence it in a positive way.

Continuation in posts: 2 and 3

#Processes #Management #Performance #Engineering #Software #SoftwareDevelopment #Leadership