Gen3 SaaS PLM – How To Avoid Future “Release Lock-in”

Do you remember what is PLM software Release Lock-in? It happened to many companies and it continues to be a big deal for many others. Industrial companies were using heavily customized Product Lifecycle Management (PLM) and Product Data Management (PDM) systems. It was a good thing to match their business processes and product development processes. However, companies are not standing still and product lifecycle management should reflect changes in product lifecycle and business strategy. Heavily customized systems had a big problem here. Companies were locked in these systems together with customer data, engineering data, inflexible business processes, and substantial business expenses. It leads to situations where it was impossible to upgrade to the new version of PLM software. Designing and building products require modern software.

PLM Software Release Lock-in

Release Lock-in was and still is one of the biggest factors slowing down PLM progress (especially for large companies). Changing what is done is hard, especially when legacy is not supported in the future. My favorite passage about the lock-in of data in legacy PLM systems was coming from the last year ACE2021 – Unlock PLM data for breakfast.

I’ll give you a prediction: no digital transformation project is going to succeed, It’s just not possible if engineering and operations systems are standing between you and real-time data accessibility. If we can’t surface this data and enable it throughout the rest of the organization, and if it’s not part of your digital transformation, then you are really stuck.

We think about typical PLM users. They are on an old version of Teamcenter, 3dexperience, Windchill, or SAP. It’s a very difficult upgrade path to get to modern technology. If we can’t get to the latest versions, we’re not getting to the openness and we’re not getting the technology that has the connectivity. How are we going to work with data that’s in a proprietary, closed system?

For a very long time, customization was a “tabu” word in the PLM software world. Customization was considered bad and many PLM vendors recommended using a new modern way to do so – configuration. Here is an example from the Arena Solutions blog speaking about customization and rev lock-in:

On-premises software models continue to plague customers who want to take advantage of upgrades while avoiding security and firewall issues. Many companies that believe they want the ability to customize soon realize the very real additional time, energy, and expense to make that happen for very little payoff. The cloud makes a big difference. You’re never going to be rev-locked, in fact, you’re always going to benefit from the new features as they come. With each release, the customer gains the newest features and functionality.

However, all these discussions were in the past. With the big progress in cloud and SaaS business models and technologies made in PLM software for the last few years, the debates about future cloud PLM are taking new interesting turns.

Gen1/Gen2 SaaS PLM

CIMdata speaks about Gen1/Gen2 SaaS PLM now. I found the materials sponsored by Aras Enterprise SaaS. Read the article and draw your own conclusion. In a nutshell, Gen2 is SaaS PLM is something that is sometimes called “enterprise Saas” and it basically a modern version of hosted PLM solutions. Here is a passage from CIMdata.

Gen 1 SaaS PLM solutions generally offer basic capabilities including document, part and BoM
management, limited configurability, and are well suited for companies with simpler products
that make extensive use of standard parts and are assembled by contract manufacturers.
Gen 2 SaaS PLM solutions offer broader capabilities, (adding CAD integration, manufacturing,
simulation management, and others), and significantly more configurability. Companies with
more complex products with a high design content or operating in a regulated environment
should evaluate these solutions.

Gen2 SaaS PLM Examples

Examples of Gen2 SaaS PLM can be easily found by navigating to all large PLM vendors. Teamcenter X, Windchill+, or Aras Enterprise SaaS are the best examples I found. All large vendors did a great job transforming their on-premise PLM software packages to enterprise SaaS. Here are some examples from Siemens and PTC.

The following picture is also very interesting because it gives you an idea about the architecture of Gen2 SaaS PLM. You can clearly see the reflection of CIMdata summary.

Future Gen3 SaaS PLM and Open Architecture

What is presented as Gen1/Gen2 SaaS PLM architecture is basically the evolution of existing PLM systems. Earlier SaaS PLM according to CIMdata (eg. Arena Solutions is 20 years old) were limited in flexibility and functions, but developed with the SaaS model in mind are meeting existing on-premise PLM platforms that receive SaaS PLM boosters, but limited with their single tenant data model capabilities.

Both Gen1 and Gen2 SaaS PLM cannot be changed. From product architecture, data management, and other fundamental elements of the system, both these systems are at the top of their productivity. From the lifecycle standpoint, these systems are like old airplanes before jet engines were invented. You can think about the IC engine, which still remains one of the most efficient ways to power cars, but eventually will be replaced with electrical motors.

Both Gen1 and Gen2 are PLM architecture dead ends. So, what is the future? All PLM vendors are in the race towards Gen3 SaaS PLM now. It might be not loudly pronounced now, but if you look at roadmaps of the companies will see it there happening.

Here are two examples. PTC is developing its Atlas Platform, which is based on the Onshape acquisition. The core element of this platform is the multi-tenant Onshape model, which has an enterprise option. I’d expect the Atlas platform to evolve to add more PLM functions. Where they would be coming from? They can come from Gen1 SaaS PLM (Arena) and from Gen2 SaaS PLM (Windchill). It is not clear yet, in my view. Both platforms have their limitations to be integrated with Atlas.

Another example is Autodesk cloud platforms. A recent AU 2022 event introduced the next generation of Autodesk clouds and Autodesk Platform Services (former Autodesk Forge). Autodesk cloud is one of the oldest services developed by Autodesk for the last decade. Autodesk was acquiring SaaS PDM & PLM software, which by CIMdata definition is most probably Gen1. Autodesk Fusion is a new cloud for design and manufacturing software.

Gen3 SaaS PLM: Data Scale, Flexibility, and Openness

Both Gen1 and Gen2 SaaS PLMs are capable solutions, but none of them provides a reliable path in the future. Two fundamental problems are related to the most critical elements of PLM architecture – (1) how to provide a data management platform capable to serve multiple companies; (2) how to support data portability. Solving the first problem is needed to provide an answer to many problems that exist today in enterprise organizations and supply chains. Solving the second problem will allow to companies to build their solutions using multiple services without the limitation of one vendor = one platform.

What is my conclusion?

The PLM industry is adopting SaaS and cloud, which raises many questions and demands decisions about future solution architecture. Where to invest and how to get a fast ROI. It doesn’t matter if you’re implementing your first PLM system or planning to replace PLM legacy software. It is very important how not to make mistakes that will lock your organization in the next round of “PLM release lock-in”. For industrial companies, the choice of the right PLM architecture will directly impact the success of the digital transformation. For PLM vendors, it is the opportunity to introduce a future resilient platform. Just my thoughts…

Best, Oleg

Disclaimer: I’m co-founder and CEO of OpenBOM developing a digital cloud-native PDM & PLM platform that manages product data and connects manufacturers, construction companies, and their supply chain networks. My opinion can be unintentionally biased.