The number of applications and tools created using AI technologies is skyrocketing. It is amazing to see how AI tools are bringing new capabilities into our lives, redefining what we did before in a different way, removing complexity, and helping us be more creative and efficient. What inspires me the most is how AI can accelerate the development of tools today by capturing specific requirements using natural language.
Earlier today, I came across an article written by Rob Ferrone, discussing a multi-view perspective on PLM development and the PLM ecosystem. First, let me start by saying thank you to Rob Ferrone for mentioning my name in your article. It is an honor to be included in such a well-structured and insightful post about the diverse perspectives in the world of PLM.
Your breakdown of the PLM ecosystem is one of the best I have seen—I love the article. I call it: How to Survive in PLM Land.
Navigating the Complexities of PLM Land
Rob’s post highlights a crucial aspect of the PLM industry—its fragmentation. The ecosystem includes vendors, system integrators, analysts, consultants, academics, and independent thought leaders, all contributing to a noisy yet valuable discussion. Understanding this landscape is key to appreciating the broader challenges of PLM implementation and adoption.
One of the most fascinating dynamics in PLM today is the ongoing tension between PLM strategy and PLM systems. Vendors push software as the solution, while implementation teams navigate the messy realities of business process alignment. Consultants, on the other hand, attempt to bridge the gap, offering frameworks that help companies structure their PLM journey. The truth is, all three perspectives are necessary, but they rarely align smoothly.
My favorite recommendation from Rob’s article is the following one:
Keep it real. We all know what perfect looks like, but what about the gritty day-to-day reality and the things that need to be fixed today? Getting the basics right provides a solid foundation on which to build.
One of the most fascinating things I’ve discovered in my work at OpenBOM is how ugly PLM implementation are in a real life. After all the diverse discussions about strategy, trends, opinions, and best practices, companies still end up creating ugly Excel spreadsheets to store product information and share it within their organization, contractors, and suppliers.
There are multiple reasons for this, but one of the key factors is a combination of the challenges posed by specialized knowledge and the desperation to find a solution for a specific problem.
Which brings me to the vision I have for PLM in 2025 – how to build custom applications using data and AI combined together. Here are some thoughts about it.
My Vision for PLM Land in 2025
As we move forward, I see three main principles shaping the future of PLM Land:
1. Separate Data from Applications and Software
The traditional PLM approach ties data to specific systems, leading to rigid, monolithic solutions. The future of PLM must be data-centric, not application-centric. Industrial companies need open, federated, and interoperable data models that allow multiple applications to work with the same dataset. This is the foundation of modern digital threads and composable architectures.
2. Build Minimum Lovable PLM Applications
The “Minimum Viable Product” (MVP) concept is well-known, but in PLM, we need something more: Minimum Lovable Applications. Engineers and manufacturers don’t just need software that functions—they need software they actually enjoy using. A lovable PLM application is:
- Simple, with an intuitive UX/UI that engineers and supply chain professionals can adopt with minimal training.
- Highly adaptable, allowing for configurations that fit the specific needs of different teams.
- Connected, ensuring seamless integration with existing tools and enterprise systems.
3. Refine PLM Using AI
Artificial Intelligence is not a silver bullet for PLM, but it is a powerful tool for refining PLM applications. AI can assist with:
- Automating repetitive tasks like data entry, validation, and part classification.
- Enhancing search and recommendations by leveraging semantic understanding of product data.
- Improving PLM adoption by offering intelligent, context-aware suggestions to users.
The combination of data-driven architecture, user-friendly applications, and AI-assisted optimization will redefine how we approach PLM solutions in the coming years.
Conclusion: The New Era of Building PLM Applications We Love
As writing code and deploying software becomes easier, the challenge for PLM isn’t just technical—it’s philosophical and strategic. The real question is: How do we design systems that truly help industrial companies thrive?
PLM’s future isn’t about a single tool solving all problems. Instead, AI-driven customization, open data models, and user-first application design will be the driving forces in shaping the next generation of PLM systems. We are entering an era where technology will allow us to define and build applications using natural language, bringing us closer to PLM tools that we love—not just ones we tolerate.
Rob, thanks again for sparking this conversation. Looking forward to more great discussions in the evolving PLM landscape!
Just my thoughts…
Best,
Oleg
