Enabling Service Maturity & Operational Excellence at Convoy

The Challenge

Scale the company and service maturity at the same time

As a startup that had been through years of hypergrowth, Convoy found itself with an inconsistent tech stack. In an effort to achieve product market fit, teams had (rightly so) prioritized speed and agility over standardization.

By 2019, in order to support a business that was at significant scale and still growing, engineering modernization projects were a focus. These projects proved challenging–teams were operating from different baselines and building with a variety of libraries and components.

So, to drive more consistency and unlock modernization projects, Convoy set out to build their own tooling for tracking and measuring adherence to standards and best practices. They rolled up their findings for each service into a service maturity score.

“When we were going to go all-in and build a much more detailed version of that [service maturity scoring], that’s when we came to know about OpsLevel. We said, ‘That’s exactly what we want to build, and you folks are building it already.’”

Waj Sampathkumar
Infra Engineering Manager at Convoy

Although foundational pieces of the project were already in use when they discovered OpsLevel, the Infrastructure Engineering team was aiming higher. Successfully driving consistency and service maturity across more than 40 microservices and 150 engineers was going to require a more robust approach.

Faced with the build vs buy decision, they recognized OpsLevel’s SaaS solution would free up their engineers for other valuable projects and immediately provide deeper functionality and greater leverage than their homegrown tool.

The Solution

A central hub for operational excellence

Convoy has used a regular cadence of Operational Review meetings to ensure the long list of service maturity, modernization, and operational excellence topics weren’t overlooked by engineering teams. With the introduction of Rubrics from OpsLevel, Convoy is looking to decentralize the process and make it more targeted and efficient, thanks to the levels and categories built into Rubrics.

“It’s Rubrics that’s actually changing the game for us,” said Waj. “It’s the siren call for our company: ‘This is the definition of level one, now everyone get to level one.’ We need that layer of prioritization so we can drive campaigns more effectively.”

Instead of being confronted with a daunting checklist of best practices to catch-up on, development teams can now work their way up the service maturity ladder, step by step. In addition to creating better focus and increasing autonomy for teams, Rubrics also removes guesswork. Developers can now easily follow engineering standards and know which service maturity tasks to prioritize and work on first.

In addition to formalizing service ownership and driving service maturity, Convoy found value in OpsLevel in a variety of other ways, thanks to the extensibility of the platform. OpsLevel is now a repository for mapping metadata like tech lead or application feature to the appropriate service. Machine learning and data science teams are also using OpsLevel to drive best practices around their machine learning models.

The Outcomes

A partnership to unlock reliability at scale

Since adopting OpsLevel, Convoy’s microservice architecture has continued to grow and they now have approximately 75 services.

“The implementation that OpsLevel supports, the magic behind that, is the ability to provide leverage,” said Waj. “It’s our primary method of driving best practices to components. Every check we add, every component gets it automatically. That’s easier than us doing these checks in 100 different places and not really knowing the impact. So that’s where the ROI is really high.”

Growth has only made OpsLevel more valuable to Convoy. As they continue to scale, their products and services are increasingly mission-critical to their customers. Thanks to the leverage they’ve found in OpsLevel, new services and components introduced to their architecture aren’t playing catch-up or scrambling to identify the appropriate best practices for service maturity.

Instead, Convoy can grow with consistency and reliability built-in, based on the guidelines they’ve defined in OpsLevel.

“You’ve been fantastic partners. That’s the thing that I enjoy the most. The tool is invaluable to us, but it’s the partnership that’s valued even more,” said Waj.

Going forward, Convoy intends to keep up the close partnership with OpsLevel. One of their corporate values is Raise the bar–so operational excellence is never a box that’s checked. Instead, it’s an on-going effort, but one that they’re well prepared for with OpsLevel as partners.