Measuring developer productivity with the DX Core 4 (Category Productivity)
Today I wanted to talk about DX Core framework 4 To measure productivity. This framework is interesting because the guys from DX are one of the trendsetters in the world of developer productivity: They sawed down the DevEx model, which I already disassembled.DevEx: What Actually Drives ProductivityandDevEx in Action" The platform team includes Nicole Forsgren, who drove the development of DORA metrics. Accelerate books.I put my hand to it. SPACE framework
- They claim that DX Core 4 This work continues and includes their previous work.
Funny that in the announcement of the new framework, the authors say that engineering leaders often ask which framework to choose: DevEx, SPACE and DORA? And they decided that the answer to this question will be a new framework DX Core. 4:) This framework is so called because it addresses the question of productivity in four dimensions: speed, efficiency, quality, impact. Now a little bit about each dimension.
1.Speed This dimension is about how quickly engineers ship production-ready code. The key metric here is "diffs per engineer", which is different from the traditional lead time. Interestingly, this is actually a bandwidth metric, not a speed metric:) The guys decided to make it key, as it is clearer for non-technical stakeholders - conditional lead time and its importance is more difficult to explain to a conditional CEO, but diffs per engineer is easier to explain. The authors emphasize that diffs per engineer cannot be tracked at the individual level and used to evaluate employees. Lead time, deployment frequency, perceived rate of delivery 2. Effectiveness This dimension is about how well teams feel about development processes and workflows in general. The key metric here is the Developer Experience Index. (DXI)It shows how processes help engineers do their jobs. (Create diffs from the first dimension). Secondary metrics are time to 10th PR, ease of delivery, regrettable attrition (organizationally) 3. Quality This dimension focuses on the stability and reliability of software in production. Key metrics include failed deployment recovery time, perceived software quality, operational health and security metrics. 4. Impact The most interesting dimension, which offers to assess the impact of engineers on business. Key metric: The percentage of time spent on new opportunities (feature). Secondary metrics are initiative progress & ROI, revenue per engineer (organizationally)R&D percentage of the organization’s budget
But a framework without instrumentalization is money for the wind, so the authors say that it is closely integrated with the DX platform, where there is a lot of money. Pre-built metrics and reports are ready-made reports for key areas such as code review, delivery and output. The platform supports display and DORA metrics A bunch of adapters for pulling data from rap code, tools for incident management, CI/CD systems and so on. All this is reduced in a unified form, which is suitable for building reports and making decisions. Part with surveys where you can collect feedback from engineers and conduct analysis, including sentiment analysis Recently, the guys rolled out their concept to analyze the effectiveness of AI in development.Measuring AI code assistants and agents" (I mean, told)So they have the tools to analyze that aspect.
In general, this is an interesting framework and a platform for its implementation, which allows you to evaluate developer productivity and further make decisions that will allow you to change it for the better.
P.S. I will be recording two episodes of Research Insights Made Simple.
- Discussing this framework. Discussion of the SPACE framework and its implementation in T I'm going to have some great guests in both episodes. Stay tuned
#PlatformEngineering #Software #Processes #DevEx #Devops #Metrics