I like the way BOM discussion usually sparks. From my experience, "BOM management" topic has the ability to ignite conversation almost immediately. Therefore, I wasn’t too much surprised when I got blog attention from Hardi Meybaum, CEO of GrabCAD – most buzzed about company in Boston if you trust Venturefizz. Here is the article – Why single BOM is not the answer for you. Read it and draw your own conclusion. My favorite passage is the one that explains "BOM problem":
Recent posts on bill of materials (BOM) like Oleg’s ‘Will PLM Manage Enterprise BOM,’ make me [Hardi] remember my first conversation about BOM management. It was when I was in charge of IT in a mid-size manufacturing company. Then, and now, Procurement Specialists said that BOM should live in ERP, Engineering Managers said it should live in PLM, Engineers said that it should live in spreadsheets, and PLM vendors said it should live in PLM. This is where today’s BOM problems start.
According to Hardi, BOM management problem will disappear because of SaaS and API availability. He promotes open exchange of data between GrabCAD and other SaaS applications and out-of-the-box integration. It will finally make right information available to everyone in an organization. From what I can see, GrabCAD’s believe that companies will move from integrated large product suites into lean and granular applications. Actually, I like the idea very much. Granularity is a very good thing. Two years ago, Chad Jackson addressed granularity on his Lifecycle Insight blog –Point Solutions, Integrated Solutions and the Granularity Value Proposition. This is how Chad explains the granularity:
The fundamental idea is that you layer on different solutions that each do something very specific and well. Basically it is the point solution approach but from an ecosystem perspective. It would include something like leaving your workgroup PDM software in place. Layer on top of that a workflow. Then add some social computing solution for collaboration. Then you can add in a project management solution. You get the idea. Leave what you have in place. Add in other point solutions where needed. And integrate them as lightly as you can.
Granularity and best-of-breed application set usually comes with two additional price tickets – IT/setup and integration.I can see an interesting and new perspective in the way Hardi presents the advantage of granular solution. It comes together with SaaS, which makes a lot of sense to me. Installation, setup, updates – all these things will go away with SaaS approach. However, I have a doubt about integration. Out of the box integration is a tricky part. I never seen it works well. If you want to create a custom integration, you eventually coming down to nuts and bolts. API plays an important role in the ability of systems to be integrated. I mentioned the importance of Web APIs in PLM last year – Why PLM Need To Lean Web APIs? Even if APIs are available, integration is an expensive thing. From that standpoint, it doesn’t really matter if your PLM and ERP systems are running as SaaS applications or installed in you datacenter. My hunch, traditional RDBMS technologies allows more options to build hardwired integration compared to multi-tenant cloud applications. BOM implementations are complex and required lots of business rules to be implemented to synchronize multiple bill of materials. This is pure integration question and hardly can be handled as custom-made work.
What is my conclusion? SaaS is a game changer in terms of how systems can be deployed and maintained. It creates a completely different view on IT spending and TCO. It is also creates a different perspective of openness and importance of APIs. It was easy to hack ERP or PLM RDBMS with direct SQL call to solve your integration needs. It is completely different for SaaS apps. You cannot do integrations without custom code or flexible integration frameworks and middleware. In my view, the real BOM management problem is not related to the fact your system runs on the cloud or on premise. It is about how to connect multiple BOMs into a single meaningful solution. This task is complex and cost money. Cloud is an excellent foundation to simplify solution available and making information easy accessible. Will it solve BOM management problems across organizational silos. I’m not sure about that. Just my thoughts. What is your take?