Ten years ago, very few people tracked their steps, heart rate, or sleep. Sure, pedometers existed, as did heart rate monitors and clunky sleep monitors, but they weren’t particularly commonplace. Now, it’s not uncommon to know how many steps you’ve taken in a day, and many people sport metal and plastic on their wrists that monitor their activity, heart rate, and sleep quality.
In the past, that information was inconvenient to access, and found in disparate places. Nowadays, fitness trackers and smart watches bring together all of that data so we can tap into it to make decisions.
Imagine this: I sit at my desk all day. I glance down at my fitness tracker and it says I’ve only taken 100 steps today. I feel really tired. What do I do? Take a walk? Drink another cup of coffee? Let’s say I take a walk. Awesome! A quick stroll to the park after lunch and I’ve reached my step goal of 8,000 steps! I feel great. I sleep well and my tracker says I got 8 hours.
The next day, faced with the same afternoon drowsiness, I skip the walk and opt for a second coffee. I sleep poorly, and when I wake up, I see that I only got 4 hours of sleep.
On the third day, instead of reaching for some caffeine, I choose to take a walk. It was a data-informed choice. Without the data, I might have ruined my sleep schedule by again drinking coffee too late in the day and later wondering why I felt so tired.
So, what does this have to do with engineering data? Ten years ago, the process of gathering engineering data involved a mish mosh of spreadsheets, gut feel, and self-evaluation. Leaders faced a black hole with no easily accessible information in one place.
Code Climate Velocity changes that. Now, it’s possible to view trends from the entire software development lifecycle in one place without having to wrangle reports from Jira or comb through Github. Putting everything together not only makes the data more accessible, it makes it easier to make informed decisions.
Let’s say I want to boost code quality by adding second code reviews to my coding process. Sounds great, right? More eyes on the code is better? Not quite. The data we’ve gathered from thousands of engineering organizations shows that multiple review processes tend to negatively impact speed to market. Why? Naturally, by adding an additional step to the process, things take longer.
But what if those second reviews lead to higher code quality? Code Climate Velocity gives you insight into things like Defect and Rework Rates, which can validate whether quality increases by implementing second reviews. If the data within Velocity were to show that Cycle Time increases and second reviews have no effect on Defect Rate (or worse, if they increase Defect Rate), then maybe we shouldn’t have that second cup of coffee…er, second code review.
This is exactly the situation I ran into with a client of ours. A globally distributed engineering organization, they required two reviews as part of their development process. The second review typically depended on someone located multiple time zones ahead of or behind the author of the pull request. As a result, the team’s Cycle Time spanned multiple weeks, held up by second reviews that were often just a thumbs up. By limiting second reviews, the organization would save upwards of 72 hours per PR, cutting their Cycle Time in half. They would also be able to track their Defect Rate and Rework Rate to ensure there were no negative changes in code quality.
We don’t want to know that drinking a second cup of coffee correlates with poor sleep — that is scary. But by being alerted to that fact, we are able to make informed decisions about what to do and how to change our behaviors. Then, we can measure outcomes and assess the efficacy of our choices.
There is a common misconception that applying metrics to engineering is scary — that it will be used to penalize people who don’t meet arbitrary goals. Just as smart watches don’t force you to take steps, engineering data doesn’t force your hand. Code Climate Velocity presents you with data and insights from your version control systems, project management systems, and other tools so that you can make data-informed choices to continue or change course and then track the outcome of those choices. Like fitness and sleep data, engineering data is a tool. A tool that can have immense value when used thoughtfully and responsibly.
Now, go reward yourself with some more steps! We’ve brought the wonderful world of data into our everyday lives, why not into engineering?
To find out what kinds of decisions an Engineering Management Platform like Velocity can help inform, reach out to one of our specialists.
Actionable metrics for engineering leaders.Try Velocity Free