As part of our ongoing series on data-driven performance reviews, Code Climate Founder and CEO, Bryan Helmkamp, sat down with Katie Wilde, VP of Engineering at Ambassador Labs, and Smruti Patel, Head of Engineering, LEAP & Data Platform, at Stripe to discuss the intricacies of performance reviews and how to best conduct them with the help of objective data. Read on to learn how leveraging objective data in your reviews can improve retention, mitigate bias, and positively impact developer confidence. To see the conversation in full, check out the recording here.
Performance Reviews – Intended and Ancillary Benefits
Objective data can assist in conducting fair and equitable performance reviews that help to improve retention and save organizations the added expense of sourcing and hiring new talent.
“Engineering hiring right now is extremely difficult, and what would be really great is if you didn’t have to backfill roles on your team. If you manage to retain people, that’s way better than having to hire a bunch of new folks,” Katie says, adding “[Performance reviews] are extremely important for cultivating and retaining our top people and that’s a different ‘why’ to the kind of thing people think of with performance reviews… yeah, sometimes we need to fire people for not doing a good job, but it’s actually far more important to make sure that the people that are doing a fantastic job don’t leave.”
To Katie’s point — a survey conducted by LinkedIn reports that, depending on the role, it can take the average organization over 40 days to fill a technical position, while the US Bureau of Labor Statistics projects that the demand for software engineers will grow 22% over the next decade, signaling hiring challenges to come in the space.
Echoing Katie’s sentiment, Smruti adds “[Data-driven performance reviews] are key for supporting your high performers who want a formal way of [understanding], what does the next year look like? Or what does the next part of the journey look like? And so for me, this process, at the very least, provides missing data where a manager or organization might not even be aware of all the things that [a developer] is involved in, or the initiatives that they’ve led and delivered.”
Using Objective Data to Combat Biases
Biases are a noted flaw of performance reviews, and a great way to mitigate those prejudices is to check assumptions against objective data.
Katie has seen firsthand the inequities gender bias in particular can cause. In one instance, when it came time to review a female engineer, Katie noticed inconsistencies between the engineer’s self-review and her peer reviews. While the engineer had given herself a strong rating, her peers stated that her work pace was a little slow. Katie decided to dig deeper and determine the cause of this discrepancy: had the engineer overstated her own abilities, or were her peers being overly critical?
Using Code Climate Velocity, Katie dug into the engineer’s work and discovered that her PRs were in fact moving more slowly through the development pipeline, but not for the reason one might expect. Her PRs were getting held up in code review by change requests, and the requests were often for minor, stylistic changes. Further investigation revealed that requests of this kind were not common across the team, and PRs written by male engineers tended to be approved quickly, without getting stuck in a similar feedback loop. With this information, Katie was able to check her assumptions, calibrate her feedback, and uncover bias in both the review process and the team’s day-to-day work.
Reflecting on the experience, Katie says “So that review cycle for that entire team went a completely different way than if I hadn’t used data. [Without objective data] I absolutely would have done the review of, give all the men great reviews, tell the woman she needs to be more productive and get her act together, that she’s got some decent skills, but she’s got to do a better job, not promote her, and carry on. It’s just quite eye-opening. I am a woman in engineering. I know how hard it is. So yeah, that’s a case where I’m just very grateful to have had that data.”
Articulating Impact with Engineering Data
Many times during the performance review process, ICs are asked to conduct a self-evaluation. This portion of the review process surfaces an opportunity for ICs to share their perspectives on their own performance, actions, and choices. It also helps managers understand how ICs view themselves in relation to their team and the company as a whole.
Citing personal experience, Smruti observed that ICs can be so focused on what they’ve built and delivered, that they often fail to see why their work mattered and the impact it had on the organization. She noticed that this lack of self-advocacy rang especially true for underrepresented minorities. To combat this, Smruti uses data to link individual achievements to verifiable facts to help ICs break the glass ceiling of self-perception.
“So for me, when it comes down to quantifiable or qualitative data, it comes down to finding the right engineering data that best articulates that impact to the team on business goals. So if you think about it, you can use a very irrefutable factor, say, ‘I shipped X, which reduced the end-to-end API latency from 4 or 5 seconds to 2.5 seconds. And this is how it impacted the business.’”
In using objective data, ICs are able to see the impact their work has on their organization, and advocate for themselves to secure promotions and higher salaries.
The Annual Performance Review Process, Optimized
With Engineering Intelligence data at their fingertips, leaders can develop a review approach that places quantitative and qualitative data into context to deliver meaningful and actionable feedback that promotes the well-being of ICs and the growth of their teams.
Speak to a Code Climate product specialist and see for yourself how data can help you conduct more effective annual performance reviews and drive engineering excellence.
Subscribe to our newsletter for more engineering intelligence.