When production is automated, your firm’s value is engineering judgment. The question is whether you’re selling it.
Engineering judgment becomes revenue when the firm sells it by subscription, as Arup does with Neuron. Neuron is a building platform for portfolio-wide performance monitoring and predictive maintenance. Arup’s engineers interpret the data from it. A project fee ends when the drawings are done. With a subscription, the client keeps paying for the judgment as long as the building runs.
What’s happening
When AI handles drafting, calculations, and initial design, what remains is engineering judgment, the ability to interpret results, assess risk, ensure code compliance, and make decisions that require decades of experience. This is the most valuable thing an architecture or engineering firm produces, and it’s always been embedded invisibly inside production work. Clients paid for drawings and got judgment included. Now they can get the drawings from AI.
Michael Rustell of Inframatic.ai frames the challenge directly: ‘Engineering judgment is built through years of doing the work AI is beginning to handle.’ The Bluebeam survey found that only 18% of architecture firms and 23% of engineering firms have AI oversight policies, compared to 41% of construction firms. The firms that haven’t formalized how judgment integrates with AI-generated output are the ones most exposed to the value shift.
The firms that recognize this shift but don’t act on it are in the most precarious position. They know the production work is being commoditized. They know their experienced professionals are the real asset. But they continue to price and sell as if the drawings are the product. The gap between what the firm knows and what the firm does is where revenue is lost.
Why the obvious responses don’t work
“Train the team on AI tools”
Adoption without new offerings. Training the team to use AI makes them faster at producing the same work. It doesn’t change what the firm sells or how it prices. Efficiency gains without new revenue streams are a cost reduction, not a growth strategy.
“Create an AI innovation lab”
Labs don’t change the business model. They produce demonstrations, internal tools, and white papers. What the firm needs is not innovation for its own sake. It’s new offerings that clients will pay for.
“Add AI features to proposals”
Features are not offerings. Mentioning AI in a proposal is marketing. Building a new service that uses AI to deliver something the firm couldn’t offer before is commercialization. Clients can tell the difference.
What’s working instead
Arup offers Neuron, an intelligent building platform for portfolio-wide performance monitoring and predictive maintenance, sold as a subscription that turns ongoing engineering judgment into recurring revenue. That’s more than a traditional engineering project. It’s an ongoing advisory relationship where AI handles the data and experienced engineers provide the interpretation. That’s the business model shift: from project fees to recurring revenue built on judgment.
The way out is to stop optimizing the old model and sell new services built on what AI can’t do. We find out what your clients will pay for, then build the AI that delivers it. Start with a pilot: in two weeks, a working first version tried on your own client work, then a decision on production. Or start with a plan for the whole firm.
Other pressures on Engineering Firms
The same pressure in other industries
Related reading
The Advice Your AI Vendor Can’t Give You
Buy the platform. It’s the floor, not the strategy. And the company selling it to you can’t give you the one piece of advice that matters: what to build that needs less of it.
A Firm Changed the Fee Before the Tools
Platinum River set fixed fees before it rolled out AI tools. People spend fewer hours on each project. The CEO says he needs more people.
Start as small as one idea, or as big as the whole firm.
Bring what put AI back on your agenda. In 20 minutes we’ll tell you whether it’s worth acting on, and where to start.
Book 20 minutes