AWI AI adoption in woodworking 2025 report

Insights from 40 AWI members on AI adoption, operational challenges, and where automation can deliver the most value in woodworking operations.

Download report

Industry insights from 40 woodworking professionals

Understand how shops currently use digital tools, automation, and AI in their estimating and operational workflows.

Key operational challenges across the industry

See which tasks consume the most time — including change orders, submittals, manual takeoffs from drawings, and data entry.

Where AI can create the biggest impact

Discover which processes are best suited for AI automation and what companies want to learn before adopting AI solutions.

Real case study from a manufacturing company

Learn how one casework manufacturer reduced estimating time from hours to minutes using a custom AI solution.

Expert insights on implementing AI in manufacturing

Key takeaways from industry experts on preparing data, validating AI systems, and choosing the right problems to solve first.

Survey findings

What 40 architectural woodwork manufacturers said about where AI actually fits in their operations — and what is still stopping them.

Survey period: Q4 2025
Participants: 40 AWI (Architectural Woodwork Institute) members
Method: Self-reported survey of current workflows, operational challenges and AI learning priorities, followed by a webinar addressing the most common questions.

Key takeaways

01

The industry is digital, but AI adoption is still early

Half the shops use digital tools like Cabinet Vision or Bluebeam, but only 1 in 5 have started using AI. There is a significant gap between current technology use and AI implementation.
02

Paperwork and revisions consume the most time

Two-thirds of respondents identified change orders and revisions as their biggest time drain. Submittals and specification documents come second. These document-heavy tasks are prime candidates for automation.
03

Trust is the barrier, not interest

The top question is not "what can AI do" but "can we trust it?" The industry wants proof that AI reads drawings accurately, and wants to understand how to validate reliability before investing.
04

Results are real when approached correctly

Estimation time has dropped from hours to minutes in real implementations. But success depends on defining the problem clearly first.
05

The preparation phase determines success

Companies that calculate real costs, ensure their processes are repeatable and data-rich, and plan for proper testing see better outcomes than those who jump straight to technology.

Which best describes how you work today with automation and estimation?

Respondents could select more than one option, so the figures total more than 100%.

Use digital tools (Cabinet Vision / Microvellum / Bluebeam) 51%
Mostly manual (Excel, printouts, manual counts) 37%
Already using AI or automation 20%
Tried some automation (templates, macros, scripts) 20%
Not sure / just exploring 14%

Key finding. Half the industry uses digital tools, but most workflows still require significant manual work. Only 1 in 5 shops are actively using AI.

Which tasks consume the most time in your day-to-day operations?

Respondents could select more than one option, so the figures total more than 100%.

Handling revisions and change orders 67%
Submittals, spec books and paperwork 62%
Counting from PDFs and drawings 46%
Data entry into ERP or spreadsheets 33%
Mapping to SKUs, BOMs or cut lists 8%

Key finding. Change orders and paperwork consume the most time, affecting two-thirds of respondents.

Which topics would be most useful for you?

Respondents could select more than one option, so the figures total more than 100%.

Can AI read architectural and shop drawings and do accurate takeoffs? 60%
How reliable is AI? How do you check accuracy and QA? 54%
Where to start implementing AI — step by step, and the pitfalls 51%
Real examples of AI in manufacturing and woodworking 51%
Quoting more projects with the same team, and speeding up estimation 40%
Off-the-shelf versus custom solutions 23%

Key finding. Accuracy and reliability are the top concerns. The industry wants proof that AI works before investing.

The biggest operational challenge

Asked as an open question, then grouped into themes.

Estimating and takeoffs 35%

Key finding. Estimating and takeoffs dominate, with over a third of respondents naming it as their top issue. The remaining responses split evenly, at 12–14% each, across document processing, shop drawings, scheduling, organisation and systems, and other.

Not sure where to start with AI, but reducing errors in take-offs and estimating from architectural drawings is the thing I am most interested in learning about.

Survey respondent

Webinar: where AI can and cannot help

Following the survey, COXIT and LVL10 Consulting ran a webinar addressing the questions raised most often by respondents.

Curtis Garrard

Founder, LVL10 Consulting

On the preparation work that happens before any technology gets built. Most AI projects fail not because of bad technology, but because companies skip this step.

Calculate the real cost of your problem Before looking at AI solutions, understand what your current process actually costs in time, materials and lost opportunities.
AI needs repetitive, data-rich processes The best opportunities are tasks that happen the same way repeatedly and involve processing drawings, documents or specifications.
Start with one specific problem Do not try to automate everything at once. Pick one clear problem, prove AI works for it, then expand.

Iryna Mykytyn

Founder, CEO & CTO, COXIT

On the technical side of implementation, and what actually matters when building systems for woodworking operations.

Different problems need different types of AI Reading architectural drawings requires computer vision. Processing specification documents requires language models. Knowing which you need is critical.
Budget 30–40% for testing and validation Most vendors skip this step, but you cannot know whether AI is accurate enough without proper evaluation. This investment prevents expensive failures.
Your data quality determines success AI learns from your existing data. If your drawings are inconsistent or your processes vary wildly, clean that up first — AI amplifies what you feed it, good or bad.

These are the full findings. The designed PDF — including the Stevens Industries case study and the document-agent walkthrough — is available below.