What is better than free PLM upgrade?
Oleg Shilovitsky
CEO @ OpenBOM | Innovator, Leader, Industry Pioneer | Transforming CAD, PLM, Engineering & Manufacturing | Advisor @ BeyondPLM
The original article was published here. More articles on Beyond PLM
There are two things in business that you can find very hard to compete with. These are simplicity and free. Think about it for a few minutes you will find many examples in your personal and business life when you have been completely sold on these two things. PLM industry darling Aras is famous for several very important innovation in the last decade of PLM system development. Both are around "Free" stuff - (1) Free downloads (aka enterprise open source); (2) Free upgrades. I've been writing about Aras Enterprise Open Source a lot in the past. If you missed, check these two blogs - PLM Open Source Conundrum and What is the real difference between Aras and traditional PLM system TCO.
Today, I want to talk about very interesting Aras innovation - free PLM upgrade. I came across the following article - Are Free Upgrades too Good to be True? Read it, since it contains many data points about Aras free upgrade system.
At the first place, some interesting statistic about upgrades. PLM systems are stuck to upgrade to a new system.
According to the consulting firm, Kalypso, “eight out of 10 dollars spent on PLM go to systems maintenance support.” And you thought your license costs were high?
Each year, companies that find themselves in this “instant legacy” predicament may consider an upgrade, but it tends to not happen for a variety of reasons. Usually, it comes down to the fact that the benefits, which aren’t easily quantified, cannot be justified due to the costs of an expensive upgrade and the disruption to the business. As the years roll on, this is no longer an upgrade, but a migration, and by that time, the business is feeling the negative impact of the legacy PLM system. Since all those costs over the years are now sunk costs, your costs are similar to a re-implementation, and this is where Aras factors into the equation.
And Aras "aha moment".
The true “Aha” moment, belongs to Peter Schroer, and the team, who years ago had both tremendous vision and discipline to be what Forrester has called, “the disruptor in the PLM market.” The upgrades, which are available at no extra charge for any subscriber, would not be possible without an Open Platform, and the discipline of an incorporation strategy that rebuilds all applications on services within the platform. This means you don’t end up with a hodgepodge of different rigid architectures, which is why so many legacy PLM systems get stuck.
“. . . it is important to note that the guaranteed upgrade of a highly customized PLM solution is not only possible but is, in fact, commonplace among Aras subscribers. Many reports that the upgraded database is typically returned from Aras within two weeks. Aras says the technical part of the upgrade generally takes a few hours, with the remaining time consumed with validation testing. Customers still need to validate integrations and perform their own acceptance testing.” (CIMdata, May 2018)
As much as I'm very impressed with Aras architecture and discipline, I'd like to put "but" after Aras guarantee upgrade system and ask you to think about elements of system upgrades. There are 3 main elements when you attempt to upgrade a complex enterprise system.
1- System or database upgrade. This is a small set of function usually impacted by changes in storage, data management, and system administration. I have to acknowledge that Aras team build a very robust and compact core flexible data management core and XML based abstraction layer. This layer allows you to do Aras magic - to upgrade it and run tests. As far as Aras will keep the core system and preserve the abstraction layer, the magic of Aras will continue. However, the more deviation will be done, the chances that complexity of upgrades (and cost) will increase. And super large PLM deployment is always the perfect reason to make a dent in a perfect architecture.
2- Solution upgrade. Solutions developed on top of Aras core are not part of Aras core and can potentially introduce new logic, configuration, and functions that can be incompatible with the previous version of the same solution. While it does not prevent Aras core upgrade, it doesn't mean that new solution upgrade will be as painless as core update. I'm very much interested in how Aras solutions can be developed, deployed and upgraded. I didn't find much information about it, so please share if you have some links
3- Processes and training upgrade. There is an entire world of process development, business logic, integrations and they way end users of PLM system are trained and use the system. As soon as you're going to introduce a new business process or new product (solution), the cost of retraining 1000s of users will be much more than cost database upgrades.
So, what are the modern trends of enterprise software development? With all respect to PLM industry vendors, they are not in forefront (unfortunately) to discover new trends. Manufacturing companies and PLM systems are known as the biggest laggards of the industry with code base aged north of 15 years. What are the trends of enterprise platform development?
In the world of a modern platform, development vendors prefer to separate two spaces - features and platform. While the first is very important for customers, the last one preferable won't be exposed to customers at all. Why customers should be involved in database migration, switching of hosting providers or introducing new storage system? You don't want your customers to know any of these... But feature management is important. User experience is important. The ability to experiment with new functionality is important.
The trend of feature management is actually very interesting. Check this article to compare systems for feature management.
Feature management tools are generally used by DevOps and Product teams to facilitate code releases, product experimentation, and feature rollout. Common feature management tools such as flags, toggles, and switches allow for the decoupling of code releases from feature deployments.
If you want to deep your heads and hands into feature management, I recommend you this article about Feature Toggling.
It will introduce you to an entire world of continues system development, features rollouts, A/B testing, releases and many others. The main point- features and functional planning of what system can do is separated from infrastructure development and breaks a traditional "upgrade" cycle into no upgrades.
What is my conclusion? A free upgrade is a great innovation. It is much better than old fashion enterprise software locking your company in an old on-premise software without alternatives. So, compared to version locking, the free upgrade included in your subscription is progress. However, what is next? At the same time, the future we won't see any upgrades. This is how modern enterprise software is built already today. Software vendors will separate infrastructure and features. Feature management will become a new way to launch a new functionality for a specific set of customers or vendors. And upgrades (even free) will be a thing in the past. Just my thoughts...
Best, Oleg
Disclaimer: I’m co-founder and CEO of OpenBOM developing cloud-based bill of materials and inventory management tool for manufacturing companies, hardware startups, and supply chain. My opinion can be unintentionally biased.?