Innovation is critical to startup success. To satisfy your customers, stay ahead of your competition, and keep your team engaged, you need to be shipping often — and at high quality.

But how can engineering departments measure and improve their ability to innovate?

With the right data, you can streamline processes and unblock your engineers, clearing the way for developers to work on fresh ideas and deliver new value to your customers.

In this 45-minute webinar, three startup leaders share how they use data to boost innovation on their engineering teams. You’ll hear from:

• Edith Harbaugh, CEO and Co-founder, LaunchDarkly
• Santiago García, Co-founder and CTO, La Haus
• Dalia Havens, VP of Engineering, Netlify

Here are some of the key takeaways:

Innovation requires iteration.

Edith Harbaugh: One of the things that really helped us innovate is we just had more swings at bat. If you’re in a cadence where you can release not every year, which is incredibly stale if you’re in the app store, but every month, or every week, and then do backend updates so you can update hourly – you’re going to win. ‘Cause part of the fact is that not everything you do is going to be perfect. But you just have to keep swinging and swinging and swinging, and one of those things is going to land.

Dalia Havens: One of the things I love about the way at Netlify we define innovation is that we focus on the simplicity of the solution to these complex problems…It’s about those quick iterations. It’s a learning journey. And it’s really hard to figure out which path will take you to the right solution if you’re not iterating.

Santiago García: To bring you some context, [La Haus is] in Latin America. So in this market, we have to invent actually everything. We don’t have [real estate] databases, like you in the US have…From here, we have to create everything from scratch. And that means you need to start to create everything with a good pace, with very fast experimenting because there are things that you don’t know.

Building a metrics-driven culture.

Dalia Havens: [Metrics] are operational measures. They are not punitive, they are very much a tool to help us improve at the end of the day.

Part of my responsibility is to create a quality of life for developers to do their best work. How are we measuring that? How are we removing roadblocks? How are we identifying that there was a roadblock? That’s where the data comes in. And when you bake it into the culture and not make it a taboo to talk about or hide it from this stakeholder or not communicate the incentive behind it, it creates a different shift…creating that metrics-driven culture is actually really important for the success of using metrics in your organization.

Edith Harbaugh: I think you really need to have metrics that people buy into in terms of, “How often do we ship?” because that means that I can feel pride that my code is out there. “How often are we able to roll back?” because that means that I can feel confidence.

Santiago García: What I think about it is what Peter Drucker said — “What you don’t measure, you can’t improve.” And that’s very important, that is very cultural. You can give feedback based on metrics, and people [can] feel comfortable with that. At La Haus, we have two values at the company…We have one that is, “Strive for more.” People want to improve, so it’s good to show the metrics, so they can see they are doing things [that] could have been better. Also, we have something that is called, “Extreme openness.” When you are open to feedback…you can take your data and make these data-informed decisions. For me, it’s very cultural.

Data provides objective insights.

Santiago García: When we started working remotely, some engineers started to complain — not complain, but they started to say, “Hey, we’re working more hours, we are exhausted, we have been working more.” My feeling was that we were delivering the same amount of work or less. But in that moment, I couldn’t measure [that feeling]. The first decision was to start measuring that, so I started using Code Climate.

Edith Harbaugh: When you get bigger, you just have to get a lot more deliberate about communication and about what you’re measuring and how teams work together. And then also still make sure that you are measuring, “Are we moving forward? Are we having meetings for the sake of having meetings?” So one question that the engineering side has to understand is, “How much new versus old features are you doing?” Like tech debt versus new, which is a really tricky balance. If it swings too far in either direction, you know that you’re going amiss. If 80% of my time is on maintenance, you’re not innovating. If 100% of my time is on innovation, and I’m not doing any code maintenance, stuff is going to start breaking in the field. So just keeping a close eye on metrics like that, in terms of “Does the team have the right pressure and direction?”

Dalia Havens: I fell in love with metrics because…through this organic journey to engineering management, I found that a lot of times without them, it’s a gut-feel conversation. And those are really hard to have without it seeming personal or you’re saying that someone is not putting in more hours or the right level of effort and so on. So what metrics have allowed us to do is sort of come objectively to the conversation. We are all trying to remove these roadblocks that are getting in your way and so on.

Transparency is critical.

Santiago García: [Transparency is] very important…I went with my team in a one hour meeting, I showed them the tool, how we are going to measure the process. And this was very clear, it was to improve the process. It was not going in a very punitive way or something like that. We wanted to improve our process, and everyone wants to improve. Actually, the next day, all the engineers were asking me for access to the tools so they could see their metrics and how they could improve them.

Dalia Havens: I’ve had an experience with an established team where it…the team was not talking about them being blocked to ship. And it was because we had metrics that we could see our week-to-week trend. And we’re like, “This is weird. For some reason, all of a sudden, we are not able to get things to production. Let’s talk about it.” So I found that it’s a really, really great way to be transparent and honest with everyone on the team. And also disarms sort of that tension. Because at the end of the day, just like all the the monitoring tools we have in infra, it’s to allow us to improve and iterate and create a better environment for development.

Innovation can be game-changing or destructive — make sure you’re moving in the right direction.

Edith Harbaugh: I think innovation in and of itself is a hard word. I think there’s the bright shiny answer that innovation is game-changing stuff that moves the business forward. The flip of that coin is innovation is sometimes harmful and destructive. And you don’t really know sometimes until you flip that coin, what way it’s going to land.

Dalia Havens: A product full of bells and whistles that are not really giving a seamless user experience is not going to be as effective or as useful to the potential end-user as one that is more thought out…Innovation is a big word for us. What I translate that to is iteration. Are we able to iterate? Are we able to learn? Do we have the tools to be able to learn to gradually get things out, or to decide on killing something? …having these tools allows the team to really define what is success or how are they working toward that success metric.

Santiago García: I think is very difficult to innovate in the tech team if you are not aligning with your business, with your customers…For my team, the engineering team, its mission is to deliver value to our customers as fast as we can, with the best quality.

To start gathering insights from your own data, sign up for a 14-day free trial of Velocity.