How To Afford Your Next PLM System Upgrade
Oleg Shilovitsky
CEO @ OpenBOM | Innovator, Leader, Industry Pioneer | Transforming CAD, PLM, Engineering & Manufacturing | Advisor @ BeyondPLM
The original article was published on my Beyond PLM blog
Nothing is more painful in PLM business as PLM upgrades. Historically, PLM implementations took a long time and include many customizations. The result of this practice was a PLM implementation based on a specific version of PLM software, which is customizable to the level that made a system upgrade very difficult and sometimes nearly impossible. The implementations stuck in such status are usually called "release lock". Companies stuck in these situations are suffering because of different reasons. The PLM system is a huge investment and the company uses it in production to support mission-critical processes. At the same time, it is impossible to bring a new release from a vendor to enable new features and bug fixing. For vendors, these implementations are also a big problem. They cause vendors to support multiple releases and do fixes for customers running outdated versions. This is a royal pain and every single vendor is dreaming how to kill them. But a customer is paying six figures maintenance a year and stuck in the old release, as a vendor you have no choice, but to support it. In the PLM world, vendors do it for large customers with a lot of money on the table, but size does matter and you can easily find yourself in the situation when you are in release lock and vendors refuse to support the PLM software version you use.
PLM vendors have been looking at how to solve this problem for a long time. The decision is not obvious and the solution is a mix of technology and business. Technologically you can make an on-premise PLM system to support migration between versions by creating a better data modeling engine and migrating customers to a new version transparently. Some vendors made such claims and made an upgrade part of the subscription model. In my view, Aras pioneered the approach of "upgrade included" for on-premise systems. At the same time, hosted cloud PLM systems did basically the same because hosting means to maintain a PLM system hosted using some IaaS (or dedicated data center infrastructure) and contracts include an upgrade. In both scenarios, customers eventually pay for upgrades directly or indirectly to vendors. Subscription payments supposed to cover the cost of running upgrades and incentives vendors to think about how to make a better technology to support a fully upgradeable model.
Meantime, a new SaaS multi-tenant model evolved for the last decade. The multi-tenant SaaS model redefines the architecture of the system. It includes re-thinking data management layers, adopting new databases and storage mechanisms, defining new tenants and logical models, and laying a new paradigm in how companies are managing and sharing data. Learn more about it in my recent article- Basic Elements of SaaS PLM How does it help to keep the system upgradable? The new architecture of the system and its foundation make two things that fundamentally different - (1) shared resources architecture optimizes the cost and allows to multiple companies to use the same infrastructure; (2) Tenant, Logical, and Customization model is created from the beginning to use frequent releases and DevOps models. Such a new architecture and technologies create the foundation of a continuously upgradeable system, which allows everyone to run on the same version of the system. A more sophisticated SaaS infrastructure can allow isolated tenants and freezing upgrades. Moreover, multi-tenant system infrastructure can be hosted for a single tenant as well, which eventually turns technical problems into a business model. Large companies can consider using multi-tenant systems as an instance for their corporate or supply chain.
Earlier this week, I attended a Siemens webinar speaking about TeamcenterX. As Siemens calls it now - future-ready cloud SaaS PLM. The message about SaaS is that the system is instant on and you don't need to worry about anything - installations and upgrades. The TeamcenterX website provides little information about subscription levels and infrastructure options. So it is really hard to understand if the subscription will provide a shared multi-tenant instance of the Teamcenter database or individually hosted Teamcenter databases will be running on a multi-tenant instance or database (eg. AWS Aurora or similar).
What the future of upgrades means for PLM vendors?
Not to support upgrades will soon become a non-starter for PLM vendors. To address such a problem, vendors will have to re-evaluate their existing platforms and technological options. The upgrade will become a costly operation PLM vendors have to support. It will be a trigger to upgrade infrastructure and to move to SaaS platforms capable of providing flexibility in sharing resources, to support continued development and eliminate costly data upgrades from the beginning.
What the future of upgrades means for manufacturing companies?
The manufacturing environment is changing. The pressure of cost, operation efficiency, new business models, contractors, and suppliers changes. This is a shortlist of challenges manufacturing companies are living every day. Companies cannot afford to run old PLM systems and to upgrade them every 10 years. This is a thing in the past. You cannot run manufacturing using decade-year-old software. Check PLM vendors and ask what upgrade model they have, how often new releases and fixes are coming, who is responsible for the upgrade, data migration, and customization. It can help you to avoid a set of very costly and painful mistakes.
What is my conclusion?
In the current status quo, not all PLM vendors and customers will be able to afford to support the needed level of upgrades, to bring new features, functions, and to evolve with the new development and requirements of security, data management, and business process. It will create a division of PLM vendors. Old, non-upgradable on-premise PLM systems will eventually stall in time and die. A new generation of flexible SaaS platforms will come to support multiple tenant options, release schedules, and new feature introductions. An upgrade will be soon a must-have feature of the system, which exists and supports the options and environments customers need. Just my thoughts…
Best, Oleg
Disclaimer: I’m co-founder and CEO of OpenBOM developing a digital network platform that manages product data and connects manufacturers and their supply chain networks.
Managing Partner and Owner at xLM Solutions LLC
4 年Oleg Shilovitsky the move to saas and a less costly upgrade should be weighed versus the cost of compromised functionality with less ability to customize specific business rules. Such tradeoff has a cost as well. In addition, migrating to saas from on premises solution to start with will be super expensive. Also with saas you are somewhat stuck with a vendor for life. Virtually impossible to change in the future if data needs to move somewhere else. Other consideration about IP ownership and security may play a role as well. Saas is great for some companies but may not be a good match for larger ones. It needs more maturity I think to support larger mfg companies, though it may get there in a few years maybe.
Managing Director for CEE region at SNECI
4 年Simplicity in the implementation, company process (supposed to be stabilized) mapping and regular system maintenance are a key of succesful project further upgrade. Even SaaS platform (eg: Saleforce..., personal experience) is not a guarantee if you miss basic rules of succesful IT project. M. Voskuil is right, M. Hudec is right and M. Shilovitsky is right. First two are bringing the solution to limit damages during the upgrade their own way and experiences. M. Shilivitsky, you were a part as an architect which was extremely difficult to upgrade ?? (Smarteam).
in TECHNIA we have put this topic as number 1 for our customers with all the PLM implementations we did. Therefore we developed Value Components which is additionaly layer of software on top of what is delivered by the PLM Vendor (DS). This way we are able to to full customization according to customer needs with no modification of the core system. It has many advantages but one of the most important is that we allow the PLM system itself to be constantly up to date: https://www.technia.com/software/value-components/
PLM Coach, Blogger & Lecturer and optimist. Passionate advocate for a digital and sustainable future. Connecting the dots.
4 年When a post starts with "Nothing is more painful in PLM business as PLM upgrades" I have the tendency to stop reading. It reflects a mindset that if you have an installed system and you want to move to a new version is the biggest problem. Before reaching that situation you have to implement PLM first in such a manner that what you did in the past is upgradeable in the future. This is a mission impossible for everyone. In my recent blog series, I illustrated that methodologies and data models have been changing. With digitization, we see even a more disruptive set of new best practices as they need to become data-driven. It would be good to study and describe them first before implementing multitenant SaaS solutions, as the same risk will be here for the future. Too much small spaghetti pieces will make it a mikado game - don't touch it as we do not know the impact. To conclude: Having a consistent implementation vision as a strategy is step one - next, be aware with every choice you make, you create a legacy, which can be justified by its business benefits. However, some generations later we will again talk about the (different) challenge of upgrades/changes.