Starbucks’s (Dis)Continuous Improvement

Over the last few days, we heard about a new initiative that Starbucks has implemented to improve efficiency and speed in its operations. Here is a quote from a recent Axios article about it. 

“Starbucks unveiled a new strategy today to reinvent itself inside and out, and to expand around the world…Starbucks plans to spend roughly $450 million to put new equipment in its North America locations to speed up service, cutting down on labor-intensive processes for making drinks and food. These new systems can cut the time it takes to make a Frappuccino by almost a third, and reduce time to cook food.”

WSJ provides more details:

“They moved all but the most commonly ordered syrup flavors and now store pitchers closer to where the drinks are made. After learning that topping the drinks with whipped cream and chocolate or caramel drizzle at the drink station was slowing down production, they moved those items closer to where drinks are handed to customers. The changes shaved eight seconds off the 45-second process. ‘Just to top the beverage with whipped cream and drizzle took six seconds,’ Ms. Jordan says.”

Well, the second quote is actually a trick. It’s not from a recent article. In fact, it’s a quote from a 2009 article on the Lean Operations initiative Starbucks was taking at that time. 

One more quote about Starbucks’s effort to improve its speed:

“‘The old Venti scoop didn't give you enough ice,’ Ms. Peterson says. Engineers experimented with ceramic coffee mugs, which then led them to develop one-piece plastic ‘volumetric ice scoops.’ But the handles kept breaking, so engineers had stronger ones made. The new scoops helped cut 14 seconds off the average preparation time for blended beverages of about one minute.”

Well, you guessed it, this is from an even older WSJ article, this time from 2005.

Need for Speed

What drives this initiative?

As I wrote in a post a few weeks ago, Starbucks will have to tighten up its production line since now, it really is a production line. As a recent WSJ article states on the rationale behind these initiatives:

“A clipboard in one Buffalo Starbucks, used to track workers’ drive-through delivery times, said ‘Expectations: Under 50 Seconds,’ ... Now, nearly half of Starbucks’s U.S. sales come from drive-through orders. Mr. Schultz wants 90% of the chain’s new stores to include drive-throughs––continuing a shift that he acknowledged has contributed to new pressures on workers.”

But it’s not only drive-thrus. Upon entering a Starbucks store, usually, there are three people waiting in line, but 15 drinks ready on the counter. These are, of course, ordered through their app or through UberEats. While customers don’t wait on-premise (which should alleviate some of the pressure at the drive-thru where the line is very visible and salient), the fact that customers are buying mainly speed (and some coffee), and not the ambience or experience of a coffee shop, is putting more pressure on Starbucks’s speedometer. 

So the need for speed is one take-away from this recent initiative. 

A History of Lean Initiatives

This brings me to the question of why a firm that prides itself in adopting “lean operations,” needs to implement such initiatives over, and over again. 

And not initiatives that rethink the whole approach to making coffee (by focusing on making better coffee, for example), but numerous marginal improvements. And I mean marginal in the positive sense; small improvements that compound when done over and over again. 

As I mentioned, a few years ago Starbucks adopted a lean thinking approach. The main essence of lean is continuous improvement. Kaizen (in Japanese “Good Change”) is the heart of Lean. It suggests making many small continuous improvements to the “production” line by empowering those on the production line to suggest, experiment, and make changes. Continuous improvement is the opposite of making huge changes every 5-10 years. 

But you can notice where the issue already lies. In the original article of 2009, it is mentioned:

“Pushing Starbucks’s drive is Scott Heydon, the company’s ‘vice president of lean thinking,’ and a student of the Toyota production system, where lean manufacturing got its start. He and a 10-person ‘lean team’ have been going from region to region armed with a stopwatch and a Mr. Potato Head toy that they challenge managers to put together and re-box in less than 45 seconds.”

The same group of engineers is also mentioned in the 2005 article. 

Having a team like that is not a bad idea. But from what I read on how this team of 10 did, the repeated initiatives, and what I see when I visit Starbucks, it’s clear that this team, at least in the way the firm used it, is antithetical to the notion of lean. 

Lean is about empowering those who do the work on the “shop floor,” to improve the work they do by finding better ways to do it. It’s not about a team coming in with a stopwatch and then finding better ways. Reading the description of the team, it sounds more like what we call “time and motion” studies or Taylorism. And while these studies are equally important and necessary, they should not be confused with lean. 

They can definitely be a good way to ignite lean. They can be a good way to educate the people on the shop floor on how to do lean. But this is not what we see at Starbucks. What we see is that the engineers came, measured time, came up with suggestions, and then froze these improvements by suggesting new processes, new tools, and new technology. But no new culture and processes to empower those on the floor to continue to improve. 

And this is the main issue here. It’s not only about educating the people on how to do the work better, but rather creating an organization-wide culture of continuous improvement. 

I have mentioned before that the essence of employee involvement in the continuous improvement of a process is the notion of the Andon Cord: enabling the employee to pull a cord that will stop the production line, and alert others to problems. Starbucks doesn’t have anything like that. 

So one issue I have with this new initiative is that it exposes the fact that Starbucks’s “lean” was never about real bottom-up lean, just traditional top-down process re-engineering. 

Continuous Pressure to Improve

The second issue I have with Starbucks’s lean approach is that it was never all that continuous. It’s been always good enough, rather than good —just like their coffee (ok, I’ll stop). 

If one pillar of lean is continuous improvement, the second one is that “yesterday’s solutions, are tomorrow’s problem.” I don’t get the sense that Starbucks is continuously pressing forward and not settling for “good enough” service or speed in a world where consumer preferences and technology continuously evolve. 

And this brings me to my favorite concept in lean operations: The River Analogy

Imagine the organization is a boat on a river, and the problems it faces are the rocks. Some rocks can be issues in and of themselves (e.g., poor quality), while others are created to resolve some of these original issues (e.g., 100% inspections, late deliveries, etc.). To cover these rocks, and allow the organization to function and deliver products, most organizations fill the river with water. This is done by carrying a lot of inventory or implementing longer waiting times (or having employees work overtime). 

The main problem is not the fact that the organization now carries more inventory (or has longer delivery times), it’s the lack of visibility of the main problems, the lack of urgency to improve, and the lack of problem-solving skills among the workers.

The solution is to continuously lower the water level (by reducing inventory, or promising better and better delivery times) as a way of creating “exploratory stress” —essentially creating visibility by exposing the “rocks,” thus creating urgency (since now there are issues to resolve), and building problem-solving skills among front-line workers. 

One of the main advantages of improving continuously, is that you don’t have to spend $450M every few years for a major implementation. You can just continuously improve by making many small (and usually low-cost) adjustments that lead to overall improvement. 

And it’s not that you don’t need to rethink things completely every few years, but many of the changes suggested in these longer intervals are things that should’ve come up on a more continuous basis. 

It’s very clear that this River Analogy is alien to Starbucks’s idea of lean. I don’t see any problem-solving skills on the front line, and while there is a lot of urgency to deliver coffee quickly, there’s also a lot of inertia in how things are done.

One might say that I am being too harsh on Starbucks, so it’s important to point out that many (maybe even most) organizations behave like Starbucks. Most organizations arrange a major Kaizen event every few years, and just let things deteriorate in between. 

So why do most firms adopt this approach rather than continuously improve? 

Well, because it's hard. Very hard. 

Creating a culture (any culture) is hard. Creating a culture that requires and supports people’s effort to continuously think about both the work they do, and ways to do it better, is nearly impossible. 

Continuously educating your employees with process management tools is difficult and costly. 

Creating a culture that supports employee efforts among managers, is hard. 

All these things are difficult and require strong leadership. 

So when we think about it in the grand scheme of things, these are just difficult activities. And because of that, what most firms ultimately end up doing is inviting a consulting firm (and paying an enormous amount of money in the meantime) every few years to try to improve. Sadly, this happens only to revisit a few years later.

Indeed, you need to have both continuous improvement (at store level) and radical improvement (at HQ and the engineering level). 

But, yes, they both have to be more continuous. 

Going back to the 2009 article, one quote still resonates:

“After a few minutes, his map showed a jumble of lines from the espresso machine to the sink, over to the ice chest and back. Shaking his head Mr. Heydon said, ‘We're on a never-ending quest to get to perfection.’”

Never ending it is!