The Enterprise Partner Supplier Customer (EPSC) Model sits as a core concept of Enterprise Architecture. It is so much at the core of everything we do that we seldom question it. Is that healthy? This post will discuss the core idea behind the EPSC model (differentiation by control) and alternative ways to think about enterprise boundaries.
First off, we only name things when we want to differentiate them. As the old expression goes, “the fish discovers water last.” In EA, we tend not to discuss the fact that we assume a particular model of enterprise identification and enumeration on a regular basis. That’s because the model is built in to the things we say and do. It’s built in to our business models and our service models. It’s built in to the way enterprises create policies and budgets and govern efforts. It’s so deeply ingrained that we rarely question it. Well, it’s time for this fish to discuss the nature of water. And to do that, we have to name it. I’m naming it the EPSC model, which is an acronym for “”Enterprise Partner Supplier Customer”. It looks like this:
The view that an enterprise is a bounded thing, with suppliers feeding in, customers getting the benefits, and partners in a peer-to-peer relationship… that’s the EPSC model.
The underlying assumption of EPSC is control. In this model, there is typically OWNERSHIP CONTROL over the enterprise. “Ownership control” means the same people OWN an influential number of shares in each of the organizations inside the box. That is not necessarily controlling interest. It is sufficient interest to ensure that the all the businesses inside the box get along well with each other. It works because employees do what their bosses tell them to do. Take that fundamental notion and expand it to the enterprise level and you get the assumption of ownership control.
Another form of control is ECONOMIC CONTROL which is typically the case when there is a huge size disparity between the suppliers and the enterprise itself with respect to the end customers. This happens in retail a great deal. Walmart is a textbook example of “economic control” since their supply chain requirements can drive massive costs into their suppliers without any substantial backlash. The fundamental model above is the same so I’m not going to redraw it. It’s still EPSC. Just with a really big “E”.
Why do we need alternative models?
The Internet has introduced some things we expected, and some things we didn’t. We expected the introduction of easy communication and easy transmission of data. What we didn’t expect: the creation of the commercial ecosystem as a differentiating factor in strategy. In other words, the creation of a product by one company that is combined with another product to be consumed by a customer dependent upon both to solve the needs of a customer that is totally unaware of either one. This is common now in the mobile applications space. A mobile app may be created from unique capabilities of four or more companies that are not just peers, they are collaborating peers, all focused on producing the mobile application. Yet the customer is unaware of the grouping.
The EPSC model is completely broken for understanding this space. It simply fails. Because there is no boss telling you what to do. There are only customer opportunities. It is organic and bottom up in its very nature.
And the moment we examine the “more than one” condition, we have to open the door to the possibility that there are more than two or three or ten. How many alternative models are there? I do not know. No one has enumerated them and drawn distinctions (hey, doctoral candidates… want a dissertation idea? Enumerate these).
I will brainstorm a couple of alternatives. This is not a thoughtful investigation of models. It’s a brainstorm. Take it with a grain of salt. But I encourage you to use the ideas to expand your own thinking.
The Dynamic Collaboration Model
The dynamic collaboration model involves a series of companies that have no common ownership but who collaborate on a very frequent basis to create positive value for customers that is achievable through the combinations of their products.
The key to success in this model is to focus on that dynamic collaboration and to build excellent feedback mechanisms with the customer. This kind of model can fall apart of the customer loses confidence in the collection of companies to meet their needs, and it is vulnerable to attack from alternative collaborative groupings that build better feedback mechanisms.
What other models exist?
My knowledge is not wide enough to suggest that I understand other possible models. Perhaps recognizing more than collaborative self interest is necessary to even perceive them. I know that there are more than one, and I’m guessing that there’s more than two. These are the two that I can see. Please post your own ideas and we can collaborate.
What does this matter?
Well, I’d suggest that the strategy of Microsoft under Bill Gates reflected the dynamic collaboration model, but that the strategy under Steve Ballmer reflects the EPSC model. We can see the gradual deterioration of value and innovation during this period. Microsoft under Satya Nadella has shown signs of moving back towards the dynamic collaboration model. Time will tell. He inherited a very weird beast.
But just as important as understanding Microsoft, what does this say about Google, Amazon, Force.com, IBM, Oracle, and the hundreds of other competitors (and open source initiatives) that fill the technology space?
- Oracle seems to play in the EPSC model. What does that say about the future of Oracle?
- Amazon clearly plays in the Dynamic collaboration space? Does that ensure a bright future? How important are each initiative in Amazon when vi
ewed in this context? While delivery to the neighborhood is necessary, are the drones needed? Or is that just good buzz?
- Google… plays in both. (Kind of like Microsoft). The EPSC model drives their revenue, but there’s a lot of initiatives in the dynamic collaboration space. Is this an intentional transition, or just opportunism?
- Facebook is primarily an EPSC business, with a very large base of users. (See economic control above). Will they make moves toward dynamic collaboration? Can they survive if they don’t?
This kind of differentiation is useful for making these kinds of observations.