Every week brings a new promise about artificial intelligence, digital twins, cloud platforms, or next-generation PLM systems. For engineering managers, the challenge is no longer finding new technologies to evaluate. The challenge is deciding where to focus before those technologies arrive.
I see many manufacturing companies investing time to discuss AI adoption strategies. Yet a surprising number are overlooking the single factor that will determine whether those investments succeed.
I don’t think the companies that gain the most value from future technologies will not necessarily be the first adopters.
They will be the best prepared.
The most important preparation can be summarized in three areas:
Prepare your information. Prepare your knowledge. Prepare your processes.
AI Is Only as Good as the Engineering Data Behind It
Much of the conversation around AI in engineering focuses on automation and productivity. Some predictions describe AI replacing engineering tasks, while others promise dramatic efficiency gains.
I’d like to offer a more practical view: AI is more useful as an engineering assistant than as an engineering replacement. It can help engineers find information faster, analyze historical designs, identify risks, and make more informed decisions.
But there is a crucial condition: AI cannot create value from information it cannot understand.
Engineering organizations often accumulate years of product data across CAD systems, PDM vaults, ERP platforms, spreadsheets, shared folders, and tribal knowledge. Over time, naming conventions drift, metadata becomes inconsistent, and revision practices vary between teams.
People learn to work around these imperfections, but AI systems do not.
An AI assistant may be able to search millions of files in seconds, but if product information has inconsistent structures, duplicate records, or missing context, the results quickly become unreliable.
This is why I see that engineering data quality should be viewed as a business capability rather than an administrative task.
Strong metadata, consistent product properties, controlled revision practices, and clear information structures create a foundation that future technologies can use.
Poor data quality becomes a multiplier of future problems, while good data quality becomes a multiplier of future opportunities.
The Digital Thread Starts With Ownership, Not Software
Many organizations assume digital transformation is primarily a technology challenge.
In the projects and environments I have worked with, the difficult questions are often not technical at all:
- Who owns product information?
- Who creates it?
- Who updates it?
- Who consumes it?
- Where does the information get duplicated?
- Where are manual handovers still required?
These questions sit at the heart of what is commonly called the digital thread.
The concept itself is simple. Information should flow continuously throughout the product lifecycle. Customer requirements become designs. Designs become manufacturing instructions. Manufacturing produces products. Products generate service and operational feedback. That feedback influences future product development.
The goal is not merely connecting systems, it is connecting knowledge.
I have seen ERP, PLM, and PDM implementation projects focus heavily on integration technology while underestimating information ownership. Without clear ownership, information quality gradually deteriorates, regardless of how advanced the software may be.
Engineering managers play a critical role here. Teams need clear rules regarding where information originates, which system is considered authoritative, and how updates move through the lifecycle. Without those decisions, digital-thread initiatives often become technical diagrams with limited operational impact.
Engineering Knowledge Is Often Hidden in Plain Sight
One of the most common risks in manufacturing organizations is that critical engineering expertise exists only inside the minds of experienced individuals.
Everyone knows which expert understands a specific machine family, product configuration, or manufacturing process.
But what happens when those experts are unavailable, retire, or move into different roles?
Knowledge that is undocumented cannot easily be reused and knowledge that cannot be reused cannot easily scale.
This becomes especially important as AI-assisted engineering tools mature. Based on what I’ve seen, the most valuable future systems will not simply generate designs. They will help organizations leverage decades of accumulated engineering experience.
To benefit from that future, companies must first capture and structure their existing knowledge.
- Design standards
- Product rules
- Lessons learned
- Historical decisions
- Configuration logic
- Change management practices
These assets often represent a far greater competitive advantage than any individual software platform.
Engineering managers should therefore think beyond document management and consider knowledge management as a strategic capability.
Every piece of expertise that becomes reusable creates long-term business value.
Products No Longer Have a Finish Line
Traditional engineering workflows often follow a straightforward model: design the product, release the product, move to the next project.
Modern manufacturing is changing that mindset.
Products increasingly include software, connectivity, service data, operational feedback, and digital documentation. They continue generating information long after delivery.
As a result, product lifecycle management is becoming exactly what its name suggests: lifecycle management.
Leading companies are asking new questions:
- Which product configurations perform best in the field?
- Which options do customers actually use?
- Which components generate the highest service costs?
- Which engineering changes improve reliability?
I believe that the organizations that answer these questions fastest can typically innovate faster as well.
However, this requires collaboration between engineering, manufacturing, service, and business functions. Lifecycle thinking breaks down traditional departmental boundaries and creates a continuous learning process.
Future competitiveness will depend less on isolated engineering excellence and more on organizational learning.
Why Preparation Beats Prediction
Let’s consider a typical manufacturing company.
The business has CAD systems, PDM, ERP, numerous spreadsheets, and experienced engineers. Projects are delivered successfully. Customers are satisfied. Operations seem functional.
Then new market pressures emerge. Customers demand more customization. Lead times shorten. Service expectations increase. Management becomes interested in AI and automation.
At first, the company appears to have a technology challenge.
After closer examination, the real obstacle is often different:
- Duplicate information
- Inconsistent naming conventions
- Fragmented product data
- Manual information transfer
- Critical knowledge stored only in people’s heads
In other words, the problem is not technology but readiness.
Organizations that recognize this typically follow a more sustainable path. They improve information quality, strengthen information flows, capture engineering knowledge, and refine processes before introducing advanced technologies.
That preparation creates momentum.
When AI, advanced PLM capabilities, or digital thread initiatives are introduced later, they build upon an already reliable foundation.
Future Technologies Amplify What Already Exists
One lesson I have taken from working across AI, PLM, ERP, CAD, and digital transformation initiatives is that technology tends to amplify existing conditions.
If product information is reliable, technology helps organizations move faster.
If engineering knowledge is structured, technology helps organizations reuse expertise.
If processes are consistent, technology helps organizations scale performance.
But if the foundation is weak, advanced tools often expose weaknesses more quickly rather than solving them.
That is why I do not think that the conversation about future trends should start with selecting a platform or evaluating a new AI tool.
It should start with a simpler question:
Are we ready?
The organizations that succeed over the next decade will not necessarily be those with the newest software. They will be the companies that understand their products, trust their data, and continuously transform engineering knowledge into business value.
Future technology will keep evolving.
Preparation is what turns that evolution into competitive advantage.
About the author
Tero Salonen is Product Director at ATR CustomTools who has spent his career working with engineering software, product data management, automation, and integrations between engineering and business systems.
Having worked as a software engineer, product manager, and product director, he helps manufacturing companies improve their engineering processes, data quality, and digital workflows to create long-term business value.
Watch our Webinar on Future Trends in CAD and PLM
Last updated: November 6, 2026
