How Engineering Managers Can Have Data-Driven Conversations About Team Health

A couple of days ago I was talking with an engineering manager in a mid-sized technology software company.  He had multiple developers across a few teams and was struggling to get a good understanding on how the teams were performing against business initiatives.  The bigger challenge was being frequently asked by his senior manager for reporting on the following:

1. Where is the development team spending their time to make we’re aligned with the business?

2. What is the velocity of the teams so we understand how quickly we’re getting features to customers?

These questions are similar to what I’ve heard from other engineering leaders that I’ve spoken with over the last few months.  Additional to those above, I’ve heard questions such as:

• Did my team deliver on the initiatives we said we would?

• Can I more accurately predict when code will be released? 

• Where do I have opportunities to improve?

Answering Questions About Your Engineering Team's Health Is Hard

The challenge with trying to answer these questions is the sheer effort to scrape together the data from the different tools in the toolchain and make sense of that data.  Then taking the time to get that information into a format that provides some meaningful insights into the software delivery process and the engineering team's health.  It’s really hard.  Sure it can be done if you have a spare engineer doing nothing, or budget to hire a full time person, to build and maintain a data lake, the data model and the front end.  But what engineering manager has the time or money for this? Not many.  But the questions keep coming and they need to be answered. And counting Jira story points is not an accurate assessment either. Yet of the 30-40 engineering leaders I spoke to:

• 80% use Jira tickets, story points and standup meetings to know where development time is being spent today

• 90% state manual data collection was necessary to get some visibility into team performance

• 80% state software delivery speed and predictability are important or very important

There’s got to be a better way.  By removing the headache of collecting the correct data from across the toolchain and democratizing it in a format that makes it usable and valuable, getting access to the right information and insights about your engineering team’s health and gaining visibility into the software delivery process, engineering managers can answer the questions coming from stakeholders. It also enabels them to have data-driven conversations about development time investment, delivery velocity and process improvements that can improve software delivery speed and predictability. 

How to Measure Your Engineering Team's Health and Productivity

CloudBees has removed the headache of building and maintaining a System of Record with a common data model to aggregate, correlate and normalize data from across tools. Our CloudBees Engineering Efficiency solution leverages this System of Record to surface metrics and analytics for engineering leaders and team managers to have the visibility they need to ensure development work is aligned to the business initiatives.

The Investment Area screen provides a breakdown on development effort spent on features vs. non-features work over time.

It lets you see and quantify where engineering work is happening, and proactively manage and balance resources to maintain a level of activity between feature and non-feature initiatives.  Be able to trend development work categories over time

The Activity screen enables you to analyze what initiatives and workstreams the teams’ attention is going towards, the number of contributors assigned to each project and how much progress is made on key initiatives or other things.

In one view, you can see all epics or projects that are progressing, identify any blockers stalling those projects and ensure development resources and execution are aligned to more accurately predict delivery timelines against your plan.

Cycle Time screen presents metrics and insights on the time it takes, in days, to bring new requirements to market - from code commit to deployment - while providing a breakdown of major phases of the SDLC to show progress and identify areas for improvements.

By reviewing these analytics to understand the time it takes for new development work to be picked up and be completed and available to users, engineering leaders can assess if the teams are working in properly sized increments to innovate quickly.  Additionally, by seeing trends when cycle time increases, engineering leaders can better pinpoint areas where a blockage or bottleneck is slowing team velocity.

Using CloudBees Engineering Efficiency, one of our own engineering directors was able to identify a bottleneck in the software delivery process that reduced team velocity.  Learn how she used CloudBees Engineering Efficiency to make data driven decisions to help her own team regain their velocity: Using CloudBees Engineering Efficiency to Reduce Cycle Time

I’ll provide an overview of more capabilities in CloudBees Engineering Efficiency as we add new features and tool integrations over the next couple of months. This solution is all purpose-built to enable engineering leaders to make data-driven decisions to improve software delivery speed and predictability. And keep stakeholders off your back.

In the meantime, I invite you to sign up for a customized demo.