Article

Leading remote engineering teams with Flow

Chances are that COVID-19 has altered the structure of your engineering teams, and your team has transitioned through the sudden shift. Now you're looking at how to continue to be effective for the long term, Coronavirus or not. To do this, it’s critical to manage and understand the work that your engineers undertake—including remote ones. And now that your entire engineering team is dispersed, you're probably looking for a way to understand the impact this shift has had on your business and how you can better support your team during this time. One way to do that is to compare data pre-COVID to data in the weeks and months immediately following it with Pluralsight Flow.

Three metrics to watch

Keeping your team rowing in the same direction toward organizational goals has likely become a bit more challenging. Because of this, it’s important to keep a few metrics on your watch list to ensure your team is staying strongly aligned. 

Specifically, we recommend you keep a close eye on productive throughput, the amount of code generated (very similar to impact, the quality and complexity of the delivered code); churn, the rework of recently delivered code; and pull requests, the peer reviewed contributions to a project. 

Comparing these three metrics from before your team started working from home against what they have become while in a distributed setting can offer insights into where you need to focus your energy. For example, if one of your development teams is suddenly increasing their productive throughput and you’re seeing their churn taper down, they could be well-suited for working from home. If another team’s throughput is slipping and their churn is rising, it could be a good time to check in on them to see what assistance is needed to return to normal levels. 

While there’s a lot of important granular data within each of these main KPIs that Flow surfaces, these high-level metrics are the first step to understanding the health of your team and the progress they’re making. Think of these metrics as you would a stop light. Green means go, and for your Flow data, it means you’re still on track with no major deviations in your data. Keep doing what you’re doing. At yellow, one or more of these key metrics have slipped outside a comfortable distance from your pre-remote numbers, and if the trend persists should be addressed. Finally, red means there are more significant drops that should be explored more quickly and thoroughly.

For each metric we recommend monitoring, we’ve listed industry benchmarks to use as your guide initially. However, your organization is unique and your benchmarks may differ.

Productive throughput:

Green: Throughput levels are at least 95% of average

Yellow: Throughput levels drop below 95% of average

Red: Throughput levels drop below 90% of average

Pluralsight Flow WFH metrics - productive throughput

Churn:

Green: Churned code stays under 30% of average 

Yellow: Churned code spikes above 30% of average 

Red: Churned code stays consistently above 30% of average

Pluralsight Flow WFH metrics - churn

Pull requests:

Green: Cycle time stays within 10% of average

Yellow: Cycle time increases 20% above average

Red: Cycle time stabilizes at 20% above average with unreviewed pull requests

Pluralsight Flow WFH metrics - pull requests

Match metrics with the business

he above metrics are a good starting point to look into the productivity of your team. But to get the full picture of their impact, you’ll want to match them to different business goals and objectives. To do so, there are three more metrics you’ll want to layer in: quarterly and annual trends, release level retrospectives and workflows.

Quarterly and annual trends

While it’s important to see when your team is entering code and how much value your team is driving today, it’s also  important to see the ebb and flow of how the work is being done and by whom. Within the Flow dashboard, you can see trends of any data points on metrics mentioned earlier, including productive throughput, churn and pull requests (along with more granular subsets of each). 

This trend data can be viewed per individual engineer or team with the ability to chart several weeks or months at a time. This allows you as a leader to identify the longtail trajectory of your team’s coding cycles.

Release retrospective

To tighten up your feedback loop, you can create reports in Flow to align with your release schedule and sprints. This is where those stoplight metrics start to get more detailed. How much impact is happening at each day of the sprint? What percentage of the sprint was churn and rework? 

By aligning Flow reports with sprints and release cycles, you can see which engineers are contributing what type of code (like new features and legacy refactoring) and what each team member’s workload looks like. Finally, you can see which engineers are reviewing each other's work and address issues in collaboration. You get a more insightful view of your team when you compare sprints against each other, like those before and after going remote.

Workflows

Finally, at the most detailed view, you can see what the daily workflow looks like for your engineers. As a manager, you’d do this not to micro-manage a team member but to get a better sense of how each individual’s work has been impacted by the new work environment. Some engineers may suddenly have a bunch of kids around while they work while others may find it mentally taxing to not be in the office. Flow allows you to leverage real-time data to help you navigate those needed conversations.

To see how this looks within Flow and to get more in-depth help on WFH productivity, check out our webinar Making work from home work for you.

Get started

COVID-19 continues to send ripples through companies globally, making the need to understand the big and small pictures of your team’s work more critical than ever. To get a better understanding of how your team is holding up with these new transitions (or during other disruptive times that impact your team distribution), use the above metrics to help guide your progress. And for those who don’t know where to start, Flow was created specifically to help you surface the team workloads and progress, so you can give your team the support they need most, whether in the office or from home.


Need to better understand how engineering is running at your organization, especially now that your team is distributed? Learn how you can get—and use—immediate insight on how your team works by scheduling a demo of Pluralsight Flow.