For finance
For delivery
The AI maturity model for architecture and engineering firms: What AI is doing well for AEC
- July 25, 2026
- Posted by: Febiani Marsa
- Category: AI
Key takeaways:
- AI is delivering real gains for AEC firms on proposal development, scheduling, risk flagging, and deliverable creation, but these wins stay localized to individual tasks rather than improving firm-wide margins.
- AEC firms score at the bottom of AI impact across core business applications compared to other professional services sectors, with PSA and CRM impact scores well below industry averages.
- The firms racing ahead aren’t licensing more AI tools. They’re connecting project, financial, and resource data into one operational foundation so AI can move from assisting individuals to monitoring delivery in real time.
Here’s a typical AEC scenario: The principal gets the call on a Tuesday afternoon. The client wants a fee proposal on a civic center project, similar in scale and complexity to one the firm completed two years ago, and they want it by Friday.
She pulls up the old project folder. Phase folders, discipline folders, and document folders are nested inside document folders. The timesheet data is in the practice management system. The actual hours by phase are in a spreadsheet the project manager built for his own use and never shared properly. The scope change log is a Word document she finds after twenty minutes of searching. She opens it, closes it, and opens the finance system instead.
We did this project. We have all the data. Why does building a fee estimate still feel like archaeology?
She is not wrong. The data exists; it always exists. It is just organized the way a house is organized when nobody planned the storage before they moved in: everything is technically present, but nothing is findable when you need it.
Table of contents
That fragmentation is also the most honest starting point for talking about AI in this industry: the AI readiness gap in AEC firms. AI cannot find data that a firm itself cannot find. In many ways, this is the early stage of an AI maturity model: before organizations can expect AI to generate meaningful insights or automate complex work, they need information that is structured, accessible, and trustworthy. So before looking at where AI is delivering value in architecture and engineering, it is worth being precise about which tasks actually give AI clean, self-contained data to work with, and which ones ask it to navigate the same scattered folders the principal just spent an afternoon searching.
AI use cases in architecture and engineering firms today
Two independent bodies of research give the clearest picture of where AI is delivering value in architecture and engineering right now. The 2025 SPI Research Impact of AI on Professional Services1 study surveyed 146 professional services firms globally, including architecture and engineering practices. Meanwhile, the ACEC Technology Committee published its AI adoption primer in October 20252, drawing on data from its member engineering firms across the United States.
Between them, they identify a consistent set of use cases where practices are seeing repeatable gains These examples represent the current state of the AI maturity model for AEC firms, where AI delivers measurable value in well-defined workflows but has not yet transformed firm-wide operations:
- Proposal development: Drafting fee narratives, pulling together project experience, and building RFP responses. The SPI study ranks this near the top of the impact rankings because the inputs are text-based and self-contained. AI engages reliably here because it does not need to reason across systems to produce a useful first draft.
- Project planning and scheduling: AI analyzes project schedules, simulates construction sequences, identifies potential bottlenecks, and predicts completion dates with greater accuracy than manual methods. The SPI data shows project management climbing significantly in the impact rankings, reflecting growing confidence in AI when the underlying project data is clean.
- Risk identification and proactive mitigation: AI tools compare planned versus actual progress using real-time data, flagging risks early enough to act on rather than discovering them at phase close. The shift from reactive to proactive is where the value lies. A fee overrun spotted in week six is a conversation; the same overrun discovered at month-end is a write-off.
- Project deliverable creation: Progress reports, meeting summaries, and scope change logs see consistent gains across the industry. Where templates exist and workflows are clear, AI functions seamlessly as a first-draft engine so teams review and refine rather than starting from blank pages.
These use cases share a common ceiling: they work best where inputs are structured, workflows are defined, and data is connected enough for AI to reason across it.
We are getting faster at the visible, individual tasks, but the core operational numbers are harder to move. That tension is where the AI maturity model for AEC firms becomes most useful, distinguishing between isolated productivity gains and organization-wide operational intelligence.
Why AEC data stays fragmented: The house with no storage plan
Most AEC firms have invested seriously in the systems that run their business. There is a practice management platform, a finance system, a project delivery tool, probably a CRM, and a spreadsheet that bridges the gaps between all of them. Each system was selected carefully, and each one does what it was bought to do.
The problem is the space between them. Firms in this position technically have everything they need. The hours sit in the timesheet system, the fee sits in the finance system, and the resource allocation sits in the planning tool. But determining whether a project is on track to hit its margin target requires opening three systems, pulling numbers into a spreadsheet, and building the answer manually, before there is still time to act on it.
That spreadsheet is the operational junk drawer. It exists because the storage everywhere else was not designed for how the business is actually run. And when AI is deployed on top of that environment, it inherits the same problem: it can work faster inside each room, but it cannot see the whole house.
This structural fragmentation is exactly why AEC firms lag behind other sectors in real-world AI returns. The SPI study found that AEC firms scored at the bottom of AI impact across virtually every core business application, with project profitability analysis, demand planning, and resource management scoring well below the neutral midpoint. The data shows that while AI excels at processing text quickly within a single workflow, it stalls when required to reason across connected financial, delivery, and resource data.
AEC metrics reflect this data gap, with PSA impact sitting at a mere 1.33 out of 5, and CRM at 1.47, compared to broader professional services averages of 1.96 and 2.12. This is not a lack of technology ambition. It is a reflection of how the industry organizes work by project, phase, and discipline, which fragments operational data by default.
What the AI maturity curve for AEC firms actually looks like
The sa.global AI maturity model for AEC firms describes three stages that move firms past this fragmentation:
- Assist: This is where most firms are today. AI supports individuals through faster scheduling queries, automated time entries, and template-generated project checklists. The hours reclaimed are real, but the gains remain localized to individuals rather than embedded in how the business runs.
- Automate: This stage shifts the focus to a connected operational foundation. When project, financial, and resource data flow into a single live picture, AI stops being an individual productivity tool and starts functioning as a delivery monitor for the firm. Fee burn is tracked continuously, resource conflicts are surfaced before commitments are made, and scope drift is flagged while there is still time to respond.
- Orchestrate: This stage represents a completely different operating model where AI agents take coordinated actions across the project lifecycle. This includes optimizing resource allocation across the entire portfolio, updating financial forecasts as delivery progresses, and triggering automated workflows the moment risk signals appear.
The SPI data makes the stakes of this operational maturity concrete. The highest-performing firms, meaning those with the infrastructure to run AI across the majority of their projects, report significantly stronger financial health, outperforming the rest of the market on pure profitability. That spread does not come from more sophisticated tools; it comes from the operational discipline that made widespread AI use possible in the first place.
Back to Tuesday afternoon
The principal finally finds what she needs, but it takes most of Wednesday. The fee proposal goes out Thursday evening, built on instinct and a few recovered numbers rather than a clear read of how that past project actually performed.
The civic center project gets won. It generates its own folder, organized by phase, by discipline, and by document type. In two years, someone will open it for a similar bid.
The firms that are furthest along on the AI maturity model for AEC firms are not the ones that licensed the most tools. They are the ones that stopped treating data as a byproduct of project delivery and started treating it as an asset the business is actively building. Every project that runs through a connected operational foundation makes the next estimate more accurate, the next resource plan more reliable, and the next AI use case more possible. The house gets storage built in from the start, and nothing important ends up in the junk drawer.
References
1 SPI Research, 2025 Impact of AI on Professional Services
2 ACEC Technology Committee, AI Adoption in AEC: A Primer, October 2025
Febiani Marsa
Febiani Marsa is a part of the Global Marketing team at sa.global, where she brings together her expertise in digital transformation, cloud ERP and CRM, and the Microsoft Dynamics 365 ecosystem. She works closely with product teams to communicate the value and impact of technology implementations—helping businesses understand how intelligent solutions drive efficiency, innovation, and measurable outcomes across their operations.