We spoke with two engineering leaders about how they’re feeling productivity pressure more than ever and how just letting developers code is the goal we’re all striving for.
Who we spoke with:
In the intensely competitive tech industry, the need to increase efficiency and productivity has been top of mind for CTOs and startup leaders long before COVID-19 drastically threw a wrench in engineering budgets and headcount plans.
And for good reason, too. According to a 2018 report from Stripe, improving the efficiency of engineers may result in a $3 trillion increase in global GDP over ten years. That’s a GDP pie that every company—from startup to enterprise—wants a piece of.
Today, we are seeing more of a focus on engineering productivity than ever before, so we talked with two CTOs about why now is a watershed moment for engineering teams and how leaders are adapting.
From Cloud to COVID: Compounding Events Increase Productivity Pressure
Ashish Agrawal, the CTO of field service software company ServiceMax who manages a 175 person engineering team, says widespread cloud adoption is what initiated a large shift in expectations around engineering efficiency.
What's evolved is customers who are now opting for cloud-based software and expect to see new features more frequently. Your release cadence is high, and that has created an immense transition in how we develop software and how teams are organized. It's not about just writing code.
How do you test the code? How do you deploy the code? How do you integrate the code? How do you push the code into production? How do you maintain it once it's pushed? So there is a need to think in a completely different mindset as you develop software.”
As a result of the pandemic, budgets and resources have been cut (some permanently), further accelerating the need to boost efficiency. Neeraj Gupta, the SVP of Engineering and Cloud Operations at Sift who manages a team of more than 50 engineers, says the combination of increased startup competition and post-COVID trends poses a real threat to existing companies. “Even before COVID-19, many more startups are competing with established businesses. So there's been a lot of pressure on existing companies to do more with less.”
The Diminishing Returns of Productivity Tools
Historically, engineers have built products to solve problems for other business units throughout their organization, but recently there has been a surge of technology tools built specifically for engineering teams. “You used to have software companies creating software to tailor to specific business requirements. Like, to solve a supply chain problem, or a sales problem, or, in our case, a field service problem.”
But an influx of new tools has also increased the number of systems engineers work in, which comes with its own issues. “Engineering teams have become more productive and more efficient because of some of these tools,” said Gupta.
But it's a double-edged sword. Teams can communicate much more efficiently and they're more productive. Having said that, managing multiple tools can also be distracting.
Gupta also says that tool preference is very personal for engineers. “Engineers by nature like to try different things. So everyone may not be happy with any one particular tool.”
More Code. Less of Everything Else.
Fragmented developer tools in an organization can also make it hard for leaders to get an accurate assessment of productivity. “I actually had different teams using different tools, which creates another set of problems,” says Gupta.
How do you provide a holistic view on top of these tools that can normalize efficiency across different teams within the same organization?
Agrawal and Gupta both see tool convergence and aggregation as the answer. “I see that as the next evolution,” says Agrawal. “For example, we have separate tools for monitoring and logging. So, can a logging tool also have awesome visualizations so that I could use just one tool for both logging as well as monitoring at the same time?”
The ultimate goal is for engineering leaders like Agrawal and Gupta to automate as much of the development process as possible and strip away all of the mundane tasks that take up so much of their teams’ time. “The core thing engineers should have to think about is just writing software,” says Agrawal.
And that, Gupta adds, can be the difference-maker for startups. “Startups by nature have limited resources available to them so it becomes very important to figure out where they want to allocate their money.
If engineering leaders can make better data-driven decisions that improve the productivity of engineering they will have a better chance of being successful.