Genius AI
How Genius AI Uses Span to Keep a Pulse on Engineering After Doubling Output With AI
400+
Business Software
New York
When I think of span, I think of ground truth. It really provides us the ground truth on how our engineering organization is performing.
Brady Allchin
VP of Engineering
Shared source
of truth on AI performance
Genius AI is an AI-forward platform that helps more than 125,000 local businesses handle administrative work with products like GlossGenius. In just six months, it had doubled engineering output and needed a way to keep track of ongoing projects as the company dove deeper into AI-driven development.
Braden Allchin, VP of Engineering at Genius AI, turned to Span to give leaders and staff the “ground truth” behind that acceleration: what took up the most time, where work got stuck, and whether using AI translated to faster delivery. In one example, Span surfaced a code review bottleneck that ultimately led to a 25% reduction in time to first review.
Today, Genius AI uses Span to track project health, identify congestion, and take pre-emptive action if necessary.
How Does an Engineer’s Job Change When AI Writes Code?
Genius AI’s investment in AI-driven development effectively doubled the engineering team’s output in just six months.
But as his team used AI tools to write more code, Allchin needed to understand how that acceleration impacted the wider org.
“The next phase for us wasn’t understanding AI usage, it was about understanding the impact of AI practices and how they could ultimately deliver better results,” Allchin said.
Why Does an AI-Forward Engineering Team Need More Visibility?
Genius AI’s engineering team had doubled its output in six months, and nobody initially knew whether that increased speed was creating unnoticed bottlenecks. The engineering team had built its own tools to understand where it spent its time and where bottlenecks were forming, but scaling that infrastructure would mean spending engineering time on something that wasn’t its core product. The company started looking for an easier way to determine how to deliver the most value with their AI tools.
Span provided the visibility that Genius AI needed to determine how effectively the org was using AI to meet its production goals and, ultimately, its business goals.
Why Did Genius AI Choose Span for Developer Intelligence and AI ROI?
For Allchin, the accessibility of the data Span provided was key to achieving the company’s goals because it empowered every engineer with data, not just the leadership team.
“What really stood out about Span was the mindset that this isn’t data just for leadership; it’s data for everyone,” he said. That way, teams could better identify and address any slowdowns in the software development lifecycle.
A two-week pilot showed what that looked like in practice. Instead of engineering data living in a leadership dashboard, Genius AI engineers could use that data to analyze their own working day.
“Span helps engineers on a day-to-day basis better understand where their time goes. Am I spending too much time in meetings? Am I getting blocked on code reviews? It helps everyone figure out where they should be deploying their time,” Allchin said.
How Does Span Measure Whether AI is Accelerating Delivery?
Span measures AI’s impact at Genius AI by tracking velocity, or how long an idea takes to reach production. For Allchin, impact isn’t a count of how many engineers have adopted a tool, but how quickly they turn an idea into a shipped customer feature.
“Velocity is the number one thing we care about here,” he said.
Not only that, but more available data means the team can understand what’s driving or inhibiting their speed. Span combines signals from across the development lifecycle to show where time is being gained—and lost—and what needs to change.
“Span gives the team underlying data on how software moves through development, so we can see how quickly we can take an idea and bring it into production,” Allchin said.
How Did Span Help Genius AI Address its Code Review Bottleneck?
Span revealed that code moving between teams was waiting longer than expected for a first review.
With the concrete data of how and where the slowdowns occurred, the next step was to discuss what changes needed to be made. After making those adjustments, the team went back to Span and saw that the changes they implemented led to a 25% reduction in the time it took for the first review to take place.
That feedback loop of finding the problem, making a change, and measuring the result is central to how Genius AI improves its engineering organization.
“Using Span has helped us bring data into our day-to-day conversation,” Allchin said. “Span’s hard data is really helpful in being objective about where we can do better: better as an organization, better as a team, and better as an individual.”
What Were the Outcomes for Genius AI After Adopting Span?
Reported by Braden Allchin, VP of Engineering at Genius AI:
25% reduction in time to first code review: After Span identified a cross-team bottleneck, Genius AI slashed the time to code review and accelerated deployment.
Shared source of truth on AI performance: Span brings together disparate engineering data so Genius AI can understand how effectively it is using AI.
Everything you need to unlock engineering excellence