Reliability was particularly critical in their high-stakes environment, where deployment errors carry hundreds of millions in liability exposure.
Strategic infrastructure decisions
Matt Buckley, VP of Engineering for Avalara's VAT reporting products, managed systems across multiple acquisitions. With 200+ engineering teams processing billions in tax remittance, infrastructure decisions required careful evaluation.
"We had established processes and subject matter expertise, but needed comprehensive data to guide platform consolidation decisions and quantify investment returns," Buckley explains.
The team evaluated GitLab as a unified DevSecOps platform to consolidate toolchain complexity and standardize development processes. The challenge: proving infrastructure investment value before implementation.
Establishing baselines
Buckley used Uplevel to establish baseline performance metrics before Avalara adopted GitLab, a single end-to-end DevSecOps platform that would consolidate toolchain complexity, automate security standards, and increase development lifecycle visibility.
"The value of Uplevel is identifying opportunities to improve before adopting a new technology. Once we've identified those opportunities and implemented something, I use it as a continuous feedback loop to see if we're trending better or worse."
An initial pre-migration assessment revealed optimization opportunities: merge requests were taking extended time to reach production due to toolchain complexity.
GitLab provided Avalara a unified platform for collaboration, component sharing, and code security. The adoption enabled comprehensive infrastructure modernization: containerizing applications, implementing shared CI/CD pipelines, and standardizing deployment infrastructure.
Key results reveal quantified infrastructure impact
The infrastructure changes delivered measurable improvements that put Avalara among the elite category for enterprise engineering performance according to Uplevel benchmarks:
Development velocity
-
Deployment frequency improved by 1,100%
-
Merge request cycle time was cut from 4 weeks to 3 hours, 3x faster than Uplevel benchmarks
-
Throughput is now 2.2 merge requests per developer per week, 2.75x more than Uplevel benchmarks
-
Issue velocity is now 3.8 issues per developer per week, 3x faster than Uplevel benchmarks
Quality maintenance
Quality metrics remained stable throughout the transformation. Uplevel data confirmed that bug rates stayed below benchmark during the productivity improvements.
Business impact
Director, Product Management Rachael Marshman observed direct customer benefits: "We're gaining confidence with our customers by making strong predictions and delivering features on time, planning roadmaps more effectively, and getting feedback more quickly."
Surfacing data before, during, and after the transformation, Uplevel provided Buckley and his teams a continuous feedback loop for optimizations along the way.
For example, when teams began sharing pipelines, new coordination challenges emerged — teams "treading on each other's toes."
Measurement immediately surfaced this constraint. Buckley implemented a rotating "champion" system where team members alternated responsibility for unblocking shared infrastructure.
Building momentum at scale
The continuous optimization demonstrated a broader organizational principle: technical improvements require stakeholder alignment. Data became the bridge between engineering decisions and business justification.
"I used Uplevel data to communicate transformation impact to non-technical stakeholders," Buckley confirms. "The metrics clearly demonstrated progression."
Quantified results drove organizational adoption. Other teams implemented similar approaches, and mission-critical systems handling massive financial exposure migrated to the new infrastructure based on Buckley’s demonstrated success.
The measurement advantage
Measurement turned infrastructure experiments into justified budget decisions. By providing multi-dimensional visibility into speed, quality, and sustainability metrics, Uplevel enabled Avalara to achieve dramatic productivity improvements while maintaining the reliability their customers demand.
“Over the last 50 to 60 years, people have tried measuring software engineering productivity without definitive solutions. The best approach is measuring multiple factors and correlating them. Having these views tied together in one place actually helps you gain a more 360° view of your overall process.”
Matt Buckley, VP Engineering, Avalara
For Avalara, adopting the single platform, along with taking a comprehensive measurement approach, delivered transformational results: 3x developer throughput gains, maintained customer reliability standards, and repeatable processes for continued growth in a competitive market.
Ready to Uplevel like Avalara?
Book a demo here: