PDM Evolution: From “CAD Feature” to Product Data Backbone (and Back Again?)

PDM businesses were built on the back of file-based CAD systems. Managing those files was a challenge, but after the successful period of the 1990s and 2000s, I’ve seen PDM evolve into a convenient CAD add-in, with each vendor offering its own. I shared some of my thoughts about 10 years ago in the article – How CAD vendors “murdered” PDM business.

10-15 years ago I was convinced Product Data Management (PDM) would fade into CAD – just a built-in checkbox for versioning and vaulting. Today, after decades of evolution, the reality is more nuanced: some PDM did become invisible inside CAD, while another, bigger part grew into the backbone that connects CAD, ECAD, BOMs, suppliers, and change.

Why PDM Felt Like It Would Disappear

Back in the late 2000s and early 2010s, the signs were clear. SOLIDWORKS had Workgroup PDM and later PDM Standard/Professional; Autodesk pushed Vault; PTC evolved Intralink into Windchill; Siemens unified around Teamcenter; and cloud-native Autodesk Fusion and Onshape embedded PDM directly. If all you needed was “don’t overwrite Alex’s assembly,” PDM-as-a-CAD-feature did the job.

What Actually Happened

The products we build became multidisciplinary, and the data we manage grew beyond vaulting files. Three shifts drove the split:

1. Multi-discipline complexity. MCAD, ECAD, firmware, compliance, and suppliers don’t fit in a single-tool vault. You need lifecycles, identities, alternates/substitutes, and cross-company change visibility.
2. The cloud reset expectations. Teams wanted to work anywhere, invite suppliers, and connect ERP/PLM without six-month IT projects. Cloud PDM/PLM arrived, and enterprise suites added SaaS options- everything messed up.
3. We are moving from documents to data. The Product Data (aka BOM+) became the center of gravity. When the BOM is the backbone, files hang from it. PDM needs to look less like a file vault and more like a graph of items, configurations, and relationships.

Two PDMs Coexist

Although, I can trigger a terminology storm now, I’d say there is more agreement about what PDM software does, rather PLM software. And here is how I see it – two PDMs:

• PDM-in-CAD: Lightweight, fast, opinionated for one authoring tool. Manages versions, references, and daily design hygiene.
• PDM-as-Backbone: System-of-record for items, BOMs, and change across MCAD/ECAD and into ERP/MES/QMS. API-first, shareable, and increasingly graph-fluent.

This separation of concerns is healthy: design teams keep agility inside CAD; operations get reliable product data outside of it.

PDM Systems of the Last ~30 Years

I used AI to generate the list and made slight editing later. If I missed something, please let me know. I probably missed some regional and country / vertical systems.

Note: “First seen” is approximate by decade; many lines predate, were renamed, or merged. The list focuses on widely used and historically significant systems across MCAD, ECAD/EDA, and engineering document management.

What Changed in the Last Decade (and What Didn’t)

• Standardization happened – inside ecosystems. SOLIDWORKS PDM, Autodesk Vault, and Onshape’s native model are “standard” in their worlds. Cross-ecosystem standardization remains elusive.
• PLM swallowed the brand; PDM kept the job. “PLM” signals strategy and governance; “PDM” is still the daily work of keeping files, items, BOMs, and changes correct and connected.
• The “BOM” (or data beyond CAD) became a jungle of data – spreadsheets are everywhere. The interfaces that matter are item/BOM-centered such as variants, effectivity, AML/AVL, cost rollups, and change histories – all lives in a variety of tools. PDM, PLM, ERP… you name it – just one big zoo of data. And it is painful. Larger companies focus on large 3+1 tools and some ERPs, smalls struggle with spreadsheets.
• Everyone watches “AI” magic, but AI didn’t replace structures. This space is very active and work in progress. Probabilistic LLMs models improve search and assistance, but structured queries, reliable rollups and compliant change tracking still require data identifiers, structure, relationships, and states.

Where PDM Is Heading

1. From “vault” to “product data platform.” Services for identity/numbering, immutable file versions, item/BOM graph, change states, and links to requirements/tests/suppliers/cost.
2. Multi-view and multi-company by default. Engineering, manufacturing, service, and commercial BOMs as views over the same graph, with first-class, controlled sharing outside the company.
3. AI as a co-pilot and intelligence, but not a schema. Agents draft changes, flag inconsistencies, suggest alternates, and estimate cost/lead-time using clean PDM objects and histories.
4. Lightweight in CAD, heavyweight at the core. Keep in-CAD PDM minimal and fast; move discipline-heavy processes into a shared, API-driven data platform.

What is my conclusion?

Will “PDM” Make a Comeback (and Leave “PLM” for Strategy)?

My short answer: Yes – at least in how customers talk about what they actually need day-to-day.

• Customers (especially SMB/mid-market) will increasingly say “we need PDM” because it maps to the job to be done: keep product data correct, connected, and consumable across tools and teams. When they say “PLM,” they often mean “PDM that reaches purchasing and suppliers.”
• Vendors will hedge on labels: “PLM” carries executive scope and budget; “PDM” signals fast, tangible value. Expect neutral phrases—product data platform, digital backbone, xBOM services—that deliver PDM’s responsibilities without the baggage.
• Practically, PDM will have a life of its own. It won’t vanish into CAD, and it won’t be only a PLM sub-module. It will remain the operational data layer everything else relies on, while PLM remains the program for strategy and governance around it.

Five-year bet: More teams will adopt a PDM-capable product data platform – sometimes packaged as PLM, sometimes not – and describe PLM as the program that organizes people, product data and process around the platform. That’s a healthier separation than trying to make a vault “do PLM,” or stuffing enterprise rules into the CAD save dialog.

Just my thoughts…

Best, Oleg

Disclaimer: I’m the co-founder and CEO of OpenBOM, a digital-thread platform providing cloud-native collaborative and integration services between engineering tools including PDM, PLM, and ERP capabilities. With extensive experience in federated CAD-PDM and PLM architecture, I’m advocates for agile, open product models and cloud technologies in manufacturing. My opinion can be unintentionally biased