The Importance of Systems, Part 2

By Ari Weinzweig, Zingerman’s Co-Founding Partner

Understanding these three types of systems—and deciding how they can work for you—is one of the best investments of time in your business that you can make.

In an earlier article, we discussed the importance of creating a corporate culture of appreciation and training around strong systems as well as finding the right system to match your business challenges.

In an article in the Harvard Business Review entitled “When Should a Process Be Art, Not Science?”, authors Joseph M. Hall and M. Eric Johnson described four types of systems including: Broken Processes; Total Industrial, Effectively Standardized Processes; Mass Customization; and Craft Systems. We discussed Broken Processes in the last issue, so in this installment we focus on the three remaining categories. To better help you consider how these three systems would fit in to your business, I’m going to briefly outline how each of them work.

Total Industrial, Effectively Standardized Processes

Last issue we talked about broken systems, now let’s talk about bulletproof ones—completely standardized processes. This type of system is made for work that requires careful, detailed, step-by-step scripting. It is for work that is done best by doing it exactly the same way each time.

When you show this type of system to anarchistcally oriented, independent-minded folks (like me), these standardized processes are often perceived negatively. In a way, this is the kind of factory assembly line system that in the early 20th century caused workers to be treated like interchangeable machine parts. Today, however, these systems can be used in an effective way—if independent-minded folks can recognize it.

Paying proper attention to systems’ design and effective implementation—whether we’re shipping $75 bottles of olive oils or building big engine parts—does pay off. Our Mail Order crew here has spent years studying and implementing the principles of the Toyota system. I’m not the expert on it, but I’ve read a fair bit and seen a lot of it in action here and I will tell you, in the right situation it works well. And contrary to what those who see the system as “restrictive” would say, the approach actually appeals to the intelligence of the people who use it, and in the process, it usually makes their work more rewarding and definitely more effective.

Systems of this sort are perfect for products, or workflow, that’s the same every single time—same stuff in, same stuff out. While variation in those situations may seem interesting to those who feel that unneeded problems are fun to deal with, the truth is that there’s no good business reason to do anything other than design a solid system. That means devising a process that takes into account the most likely problems that can come up, and setting it up to be self regulating in terms of keeping those issues from causing complications for our customers. Packing boxes, lab testing milk for sanitation before cheesemaking, quality testing products before they’re sold or served and so on, all are good examples of where effectively standardized processes are appropriate and helpful.

As an anarchist and a decently creative person with a lot of innate authority issues, I believe that in the right settings—as outlined above—there’s no reason that Total Industrial, Effectively Standardized Processes can’t be set up to be fun to work with, respectful of the people using them, and basically foolproof at the same time. Consistency pays off big time because it reduces time spent on work that’s best done without a lot of variation, and reduces waste so that resources can be applied in other parts of the business that can benefit from more creativity.

Mass Customization

Do you have a product that’s made in significant quantities, where you have a good system/well-tested recipe, but where nature precludes having the product come out exactly the same every single day? If that’s the case then you have a setting that is ripe for Mass Customization.

The artisan breads we make at our Bakehouse fit this mold perfectly. We do have systems and time-tested recipes. We do follow them and we do a good bit of training to help people learn to use them. But the reality is that weather, humidity, temperature and other variables make it so that the bread is going to be a little bit different than it was the day before. To eliminate that modest variation, in my experience we would have to reduce the quality levels at their highest points and make the bread a systems-friendly industrial product that would, wonder of wonders (pun intended), taste the same way every day. At Zingerman’s, though, we choose to accept that small range of variability. In order to have the chance to hit the high note—those few days a year where all the elements align and we produce a 10 out of 10, totally perfect product—we have to accept that some days the humidity, the bacteria and the temperature might mean that we only get a 7 or an 8.

The idea that my Harvard Business Review author-heroes outlined made clear to me that Mass Customization is great as long as our customers are accepting of the variability in the product that this system allows. That isn’t always the case—I don’t think it would work well with car tires, computers or medication. Consumers want and need total consistency in those types of products. But in the world of artisan food, customers have come to expect that there’s some natural, desirable swing in flavor and texture from one day’s production to the next. In fact, they actually like the fact that that variation exists and they understand that, to a degree, that’s what they’re paying for. In other words, if the people who buy our bread like the fact that some days it’s a bit crustier/darker/lighter/softer/etc., then the process is working.

Craft Systems

The fourth type of system that Johnson and Hall wrote about was the one that caused me to open my eyes wide and smile. The question they posed is: When does a well-designed system need to be both systemically sound but still artistic?
I’ve struggled with this for years. I tried redoing any number of systems in a hundred different ways, but no matter what I did, the “people” part of it was just too big. I couldn’t figure out how to guarantee systemically that a counter person, server or phone sales staffer was going to put the right amount of affect into their tone, make eye contact with the guest or bring the kind of energy to the interaction that we were looking for. I wanted to figure out a way to make this work pretty much foolproof. But every way I looked at it, it kept coming down to the reality that while we hire well, train our staff extensively and set up the work involved carefully, at the “end” of the system there was still this human, variable, out-of-anyone’s-control interaction. In these instances, good systems can help, but they can’t guarantee success; like it or not, it’s not possible to take all risk/chance out of a situation that involves two human beings communicating with one another.

Interestingly, in hindsight, I can see where I wasted way too much time searching for the “right” systemic solution. The Harvard Business Review article got me to stop arguing and embrace the reality that while you can script service sequences and tell staff exactly what they’re supposed to say in every setting, the problem is that service doesn’t work that way. When everything is totally scripted, staff usually ends up a) sounding like robots, and b) making silly statements and ineffective decisions because they’re understandably afraid to depart from the script. Service has way too many variables to fit into the Mass Customization category.

Johnson and Hall call this an “artistic system” but I call it a Craft System because art, as I’ve learned the term, basically means you can do whatever you want, and as long as it’s creative and inspired, it’s art. In this case, it’s not all art; there’s a big need for good science and solid systems work.

Regardless of the exaact name, the key, the authors point out with this fourth type of system, is to design specific processes that are rather rigidly used all the way up to the final “artistic” act, which in the case of say, service, would be the actual interaction with the guest. By making the system as consistent as possible in the early stages, we’re able to minimize any unnecessary variability and bring us to the moment of systemic truth in good shape.

Craft systems are exactly what we do in our service work here at Zingerman’s. We want careful and thorough training; we want to make sure that as many parts of the system leading up to the customer interaction as possible are standardized; and we want to train staff how to handle the situations they may come up against. But once staffers go out “on stage,” the reality is that we rely on their ability to adapt, to use their insight and make great things happen as best they can. The fact that we free our staff to do this is a big part of what sets the service work we do here apart from most other businesses. We’ve also learned that it’s an important factor in why our best people enjoy working here so much. We expect people to think and then adapt everything as needed, all in the interests of getting the sort of results we’re looking for—great service to customers and coworkers, great financials and great product quality.

Your Systems

I hope that I’ve given you at the least a bit of new food for systems thought, and at best helped to frame what can be a confusing jumble of organizational material. When it comes to working on our business, the attention to systems design, effective management of those systems, the training to support them and the work of identifying and improving them is as good an investment of time as I think we can make.