Moving faster with AI only happens when it is a goal
A software factory gives you output speed. Outcome speed needs the organization to change around it, and a change that isn't a goal loses to every change that is.
Written by
-
Benedikt Stemmildt Founder & Co-CEO
Published: 19 May 2026 Updated: 2 min read
One example from our experience: we built a software factory with a client. Ticket to pull request, the line running. Three months in, the team was faster. More pull requests, more code shipped, week after week.
And still nothing the board tracked had moved. The speed was real. It just landed on the output, not the outcome.
The line was fine. The line was never going to be the problem.
Everything around the line stayed the same
What hadn’t moved was everything around it. How work gets scoped. Who reviews the output, and how closely. What “done” means now that the first draft comes from a machine. Where the real bottleneck sits once code generation isn’t it. The org kept running the way it ran in 2023, and the factory just fed that faster.
One number captures the whole thing. In a Multitudes survey of more than 700 engineering professionals (opens in a new tab), 86% want to use AI to move faster. Only 15% have OKRs that track speed and velocity.
Wanting the outcome is not the same as making the change a goal.
A change that isn’t a goal loses every week
A change that isn’t a goal doesn’t really happen. It gets out-prioritized by every change that is one. The roadmap is full, the quarter is planned, and “adapt how we work” sits on top of all of it. Unowned, written down nowhere, losing every week to the things that are.
You can watch it in the hours. Teams are saving time with AI now. Most of that time gets handed straight back. Back to the same handoffs, the same unclear ownership, the same review that now waits on a bigger diff. The tool changed. What it feeds into did not.
A factory dropped into an unchanged organization just runs the old dysfunction faster.
Put it where the goals are written
So if moving faster with AI actually matters this year, it has to show up where the goals are written. Not a line item under engineering. Its own goal, with an owner, for the quarter. That is how a team gets the factory it thinks it already bought.
What would it take for “change how we work” to be a real goal on your team this quarter, and not an assumption?
(Source: Multitudes survey of 700+ engineering professionals, reported by LeadDev, March 2026.)
Further reading
Written by
-
Benedikt Stemmildt · Founder & Co-CEO
More on Organization and roles
- Delegate, review, own: the operating model of AI-native teams A model lab's playbook for AI-native engineering teams repeats one pattern across seven SDLC phases. The hard part is the review, where the engineer decides which agent findings matter.
- Teams that rebuilt their org chart got more out of AI McKinsey surveyed about 300 enterprises on AI adoption. Top performers were 7x likelier to run AI-native workflows, while 70 percent have not changed roles at all. Pod restructuring is an upstream investment.
- AI adoption is a leadership problem Teams with every AI tool available still drop them within weeks. Three leadership responsibilities decide whether adoption sticks: unblock usage, make adoption visible, reduce fear.
Test the constraint in your system
The Factory Readiness Assessment connects this research to your value stream, your data, and the next decision.
Training on this topic: Agentic Engineering for Engineering and Product LeadersWant to check one team first? The self-check: twelve yes or no questions, three minutes, instant result, no account. Take the self-check