The First Step in Choosing a Manufacturing Partner – Finding the Right Fit 

By Quentin Samelson, Sr. Managing Consultant, IBM Consulting, Industrial Manufacturing

Let’s suppose that you work at an electronics OEM and have been assigned to help select a manufacturing partner (an “Electronic Manufacturing Services Provider”) for a particular product. This isn’t a trivial exercise: once you’ve started working with a particular EMS company, it can be hard to disengage – so it’s important to get it right. If you research the question on line or ask an industry expert, you’re likely to get a fairly long list of decision criteria to consider (see Table at the end of this blog for a simplified version).

At the very beginning of the process, however, you may realize that there are hundreds or even thousands of potential EMS providers that you could evaluate. How do you identify the companies that should be on your short list?

There is an old maxim that applies to this effort: “Know thyself.” You’ll need to understand, and be able to describe to others, the distinctive features of your company’s product, approach to manufacturing, and the market environment that you operate in. Those three sets of characteristics are likely to have a significant effect on your expectations. If your manufacturing partner doesn’t know and understand them, they are also likely to be a rich source of significant disconnects. The reason is simple: the most critical decision criteria will be about finding a match to your needs, rather than a subjective idea of which company is “best” in a particular area. It seems obvious – but just as you would never try to manufacture pacemakers in a steel mill, no matter how advanced that steel mill was; you wouldn’t want to assemble & configure servers in a cell phone factory.

This can be more difficult than it sounds, especially if you’ve worked for the same company or in the same industry for most of your career. For example: I had worked at three different companies before I realized how completely different the business of making and installing cellular infrastructure was – from every other business I had worked for up till that point. I had gotten used to production lines that finished products within a few hours or a few days, shipped them to customers no more than a week later, and invoiced them the day after shipment. It took me weeks to get used to the idea that finishing the production process was only the first step in many, before we would be able to invoice the customer. You may not even know what’s distinctive about your operations; but a way of starting the process is to develop answers to three questions:

1. Describe the complexity of your product structure. How many components does a product typically have? How many process operations are required to manufacture it? How many subassemblies are used, and are they built in-house or sourced from other companies? How complicated is the BOM?
2. Describe the approach to manufacturing: is it manufactured using a repetitive approach (usually on a production line or lines, usually at high or at least steady volumes); or is it manufactured in smaller volumes, using more of a customized, configure-to-order approach? Are Lean Manufacturing principles used? Where is automation required? What sort of final assembly, configuration & test is needed?
3. Describe the market environment. How complex are customer orders? Are they typically placed direct, or through an intermediary of some kind? How volatile is demand? (Can you count on some weeks or months of relative stability, or are requirements constantly shifting?) Are there industry or governmental agency regulations that must be met in order to be successful?

The goal isn’t so much to perfectly answer these three sets of questions, but to answer them well enough to start a really good conversation. Potential EMS providers will ask you questions to clarify; and their statements will reveal areas that you’ll want to follow up on.

The goal is to find potential manufacturing partners that are capable of supporting customers with similar characteristics to yours. To see why this might be important, try building a matrix for the first two characteristics, as below. You might put Cell Phones in the upper left quadrant (they are quite complex but manufactured in a repetitive environment, at high volumes); but you’d put computer servers & mainframes in the upper right quadrant. (They are just as complex, but manufactured in much smaller volumes, according to customer requests.) Simple radios, headphones, speakers and the like might go in the lower left-hand section; but customized sensors would likely belong in the lower right-hand quadrant. A factory that is perfectly suited for one quadrant is unlikely to be effective in another.

This applies to factories (and sometimes production areas within the factories). You could perform the same exercise in the automotive industry with commercial automobiles (upper left) and farm machines (upper right), with certain automotive parts and accessories in the lower two quadrants.

Many EMS providers already serve multiple industries, and have set up different operations to serve different types of customers and industries. Obviously, you’d want to make sure that the EMS is planning to build your product in a factory that fits your needs.

In terms of customer service models, we don’t necessarily need another 2×2 matrix; but it is important to think through whether the combination of your customer service team and the EMS will not only be able to handle getting your product into your customers’ hands, but also be able to work with the way that your customers share their requirements with you. Inserting another entity in between your customers, your own company, and your suppliers can introduce a significant lag between changes in customer or market demand, and those changes being reflected in the demand that is shared with your suppliers.

(This is commonly called the bullwhip effect, and it can be devastating. I’ve seen a small – 10% — drop in demand create a three-month gap in requirements shared with a supplier because of this. There is a lot of material on line about this phenomenon, but an excellent starting point is an article from 1997: https://sloanreview.mit.edu/article/the-bullwhip-effect-in-supply-chains/.)

Developing an understanding of these three sets of criteria and sharing them with prospective manufacturing partners won’t automatically guide you to the best EMS supplier for your product. But the process will get you to a good starting point. For a full evaluation, you’ll need something more like the table below. Note that these selection criteria are fairly generic and are summarized for simplicity’s sake; a thorough evaluation would likely include at least several dozen individual criteria.

Table: EMS Selection Criteria