Over the weekend, I was catching up on the Mendix World online event. Mendix, which is Siemens Industrial Software darling, recently got in focus because of the growing trend of low code development in PLM and manufacturing companies. Check the website- it gives access to tons of videos with presentations and discussions from the event. There is a good set of videos about customer user stories demo how the Mendix platform is used to develop applications for a variety of enterprise companies. My favorite presentation is about building intelligent workflows. Unfortunately, Mendix World doesn’t allow embedding videos of their presentations. So, the presentation here.
My earlier article – PLM Honeycode raised tons of discussion online and offline. It struck a nerve with many people looking at how to build custom applications tailored to specific needs. Low code is an interesting trend and recently it is often coming up as a technology that is supposed to solve the problem of complex customizations.
There is always a good reason to customize your PLM solution. In reality, flexibility is a big deal and no PLM system was able to survive out of the box. It is a question of company policies, business practices and organizational structure are factors that impact the way PLM vendors are customizing their software. Nothing wrong with that approach. Companies are looking at how to move from holistic “company” wide customization towards specific “customer” (read – user) needs. PLM software should be more personal and, by doing that, to attach to specific user functions in a very unique way – to provide value, be available everywhere, anytime, and in any form. This is the future.
But, there is one problem with that. Customizations aren’t the sexiest thing you can think about in enterprise software. The thing can go messy, custom projects can take much longer and much over the planned budgets. Finally, heavily customized systems are a barrier for the future PLM platform progress. The projects to upgrade thousands of people using outdated legacy PLM systems are well known and they fail in the most spectacular way burring budgets, times, and carriers.
So, what to do? PLM configurations became a way to solve the problem. In many articles and examples, PLM vendors explained the difference between configurations and customizations, promoting modern flexible user interface and capability of the system to configure a specific behavior, user interface appearance, and business logic. Nothing wrong with that – a good configurable user interface is the right way to go. However, still it cannot solve the problem of writing custom applications, getting specific data reports, business workflows, and handling complex levels of enterprise integrations.
What is my conclusion?
Low code is an interesting trend. Behind sexy names, you can see interesting application development platforms, the ability to customize data appearance, business logic. Low code platforms solve the problem of mobile applications getting access to legacy data and simplified experience. Low code became a sexy name to bring technologies to combine data and business logic into re-usable applications. Inside, the technique and tools are not much different from web application development, enterprise integration platform, workflow builders, and cloud delivery platforms. The mix is interesting. But the real question is how to connect these tools to the data. Without that, low-code platforms won’t be any much different. Connecting to sources of product information is the tricky part because most of these tools are old and not cloud and web-friendly. Before jumping into low code, you better make an assessment of what is the way to connect meaningful data to these applications. Just my thoughts…
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.