What I Learned About TPS From a Coffee Empire

To gain a deeper understanding of the Toyota Production System (TPS) and Lean, I dove into a book called “Steady Work” by Karen Gaudet. It was referenced in Jeffrey Liker’s classic “The Toyota Way”. Karen Gaudet was an ex-Starbucks regional director responsible for about 100 Starbucks stores in the northwest region of the United States. Her book chronicles two significant events during her career at Starbucks,

1. The circa 2007 introduction of Lean thinking at Starbucks inspired by the Toyota Production System and its trailblazing experimentation and rollout in a number of lighthouse Starbucks stores in her region.
2. The Sandy Hook tragedy of December 14, 2012, its impact on one of her Starbucks stores in Newtown, Connecticut and how Starbucks’ Lean-inspired playbook was used to weather the significant 3X increase in demand at the Newtown Starbucks as the world descended upon it in the aftermath of the tragedy.

It’s a great account of how a non-manufacturing business and industry was able to leverage lean and TPS concepts from the very top on down. Transforming not only how they managed their business but perhaps more importantly, transforming how they regarded and treated their employees (referred to as “partners”), especially the front-line baristas.

Three personal insights caught my eye and raised my eyebrows.

1. Standardized work has its place.
2. A “Playbook” doesn’t have to be prescriptive.
3. A validation that “Best practices” don’t exist.

Standardized Work

I used to think that the moment you standardize something, say a process, you stop focusing on what the objective or outcome of the process was and start mindlessly focusing on executing the steps of the process exactly as specified. Losing sight of the why behind the process. The baristas at Starbucks saw standardized work as an affront to their humanness and creativity. They didn’t want to be treated like robots. The reasons for standardizing the work of preparing the over 87,000 different coffee drink combinations at Starbucks included,

• Reducing customer wait time (system lead time) for their order
• Improving the consistent quality of the customer’s order

There was also a third reason which ultimately became the primary reason,

• Spending less time on making coffee and more time on human contact: engaging and interacting with customers.

After all, according to a Starbucks VP,

Starbucks is not in the coffee business serving people; Starbucks is in the people business serving coffee.

The baristas and other skeptics in the organization slowly realized that by standardizing the work, it actually freed up time to make more human contact!

If I extend this thinking to knowledge work, there will always be a combination of “Standard” work (common, low to no variability, highly predictable, easily repeatable) and “Non-Standard” work (special, unique, high complexity). Taking time to standardize the standard knowledge work could have two benefits:

1. Standardized work that is manual in nature is a natural gateway to automating that work to further reduce process and lead time.
2. Standardized work frees up our cognitive load to focus on non-standard work.

Playbooks

I used to cringe and have a very negative and visceral reaction to the mere mention of the word “Playbook” in the agile space. The aforementioned playbooks were usually authored by what Martin Fowler referred to as an Agile Industrial Complex, otherwise known as one of the top ‘x’ consulting firms. Those playbooks usually involved documenting and prescribing a set of ‘best practices’ to implement or install Agile ways of working in an organization. For me, it was tantamount to prescribing how to be a human. After reading Steady Work and its account of how it used its ‘playbook’ for running a Starbucks store, I’m seeing the term ‘playbook’ in a whole new light. One which balances the rigidity of prescription with the flexibility of customization or tailoring. To begin with, Starbucks differentiates between “Standards” and “Methods”.

• Standards: Its playbook captures and defines the corporate standards or the “what” that is to be achieved. And, rather than being developed in a vacuum by process experts far removed from the work, those standards were actually field developed and tested with the people doing the work.
• Methods: The people doing the work define the methods for “how” the playbook is implemented in each store. This enables the playbook standards to be tailored to the unique culture, customers and needs of each store.

It kind of reminds me of the difference between the universal set of guiding values and principles as defined in the Agile Manifesto and the actual set of frameworks, methods and practices which Agile teams and organizations choose to use.

Best Practices

The book reiterates why the concept of ‘best practices’ is misleading. Something that I’ve always subscribed to when I exclaim to my clients,

“There are no best practices, only good practices and patterns”

What one organization develops as its practices is based on specific conditions and its environment at a point in time. Too often, another organization, seeing the success of those practices will try and copy those exact practices without appreciating or even knowing what those specific conditions and environments were. And then, they are surprised when those practices don’t generate the same results for them.

I love how the book captures this sentiment with the following pithy quote:

“Best thinking is preferable to best practices”

Karen Gaudet

Copy that!