What I learned this week … came from some reflection on a workshop I conducted on enterprise systems for a small division of a large A&D contractor. We discussed their needs for functionality across a number of different solutions including ERP, PLM, QLM, and EPM. 
Other than a craving for alphabet soup after all of the acronyms flying around the room, I took some time to think about the enterprise systems ecosystem and how disjointed it is. I couldn’t believe the number of different, disconnected solutions they needed. It begged the question, should we fight best-of-breed or embrace it?
My Observations
On the one hand, I realized that there was no one system that could handle all of their needs, and that they would have to address overlaps and conflicts between the systems they needed. Their business strategy requires capabilities including:
- ERP
- PLM (Product Lifecycle Management) – including a significant amount of document/content management
- EPM (Enterprise Program Managment) – including project accounting and billing for government contracts, which might not really fit cleanly under the EPM label
- QLM (Quality Lifecycle Management – including process management for things like ISO, CAPA, and others
As they begin to look at solutions, I am hopeful they find and ERP that meets their program/contract accounting needs. I am also hopeful that either their ERP or PLM will be able to give them a start on their quality and risk management requirements. I am also thankful that they don’t need customer relationship management (CRM), supply chain management (SCM), or service lifecycle management (SLM) right now – just thinking about integrating all of that makes my head spin.
Even without the other solutions, this was a big laundry list of solutions for a small manufacturer to tackle. I was concerned about the best-of-breed approach we were arriving at, although I am not aware of any single, integrated solution that would meet their needs. Then, I also realized that integrating the collection of solutions they need will be a lot easier than it would have been even five years ago. In many cases, integration today involves connecting web services. It still takes work to map out processes to the appropriate solutions and define cross-application workflow (reminds me of an older article I wrote on Business Processes Cross Application Boundaries, still interesting to read I think). But the technical job of integration has gotten much easier. While I don’t expect this company to develop a lot of sophisticated composite applications, simply tying capabilities together in a portal and creating hyperlinks between information may take them a long way.
Implications for Manufacturers
I believe that the trade-offs between a fully integrated solution and a best of breed approach have shifted. Integration is still not easy, but the need to trade off critical functionality for integration has diminished. For example, in the integration of ERP with PLM we have seen a lot of advancement, and I see a lot of standard integration offerings between vendors. And as one participant in a recent study on ERP and PLM integration indicates, even solutions from a single vendor typically require integration work because one size does not fit all. But if we stick with best-of-breed, don’t we perpetuate the lack of standard application boundaries and overlapping scope issues we find with the different applications? Clearly, there is still a trade-off to be made.
So those are some of my thoughts on best-of-breed strategies, I hope you found it interesting. I hadn’t really given enough thought about how much has changed until putting it into context for this manufacturer. What do you think? Is best-of-breed returning to favor as the preferred approach? How much functionality are you willing to trade for integration?