Operations Research at Starbucks

A columnist at Inc.com is blaming operations research for some of Starbuck's ill-advised policed, in its all-out drive to optimize profits. (Thanks a lot to the reader who pointed out the link!) The author, Joel Spolsky, describes a situation at a Manhattan Starbucks store on Eighth Avenue at 58th Street where an "expediter" walks down the customer line and asks patrons what they are planning to buy to rush their  coffee order. Since the queue is first-come-first-serve, it is unlikely the barista gets started on the drink that much earlier,  but customers in line can hope that they won't have to wait twice - once at the cashier's and once at the pick-up counter. Because the expediter behaved in extremely rude fashion the day Spolsky visited the store, he decided to explore the issue further and ended up on Starbucks's gossip site, where employees discuss various store policies. As it turns out, at least according to the grapevine, the "expediter" isn't there to make the customer experience more satisfying at all. He (or she) is there to make the patrons place their order earlier so that they will feel an ethical obligation to wait until they get their drink, instead of balking at the long lines and getting their coffee elsewhere.

According to Spolsky, this evil policy is the result of operations research in action - he writes : "All of this fancy optimization stuff is called operations research. It's what Michael Gerber talks about in his best-selling book The E-Myth Revisited . If you're planning to expand your business to a certain scale, you must first establish procedures and build systems to get predictable outcomes so that your employees can produce decent results even when they're not having a great day. It's a real academic field of study, and it's really hard and really important. You need to hire pretty smart people to do studies and experiments and collect the statistics and then figure out what it all means." My first reaction was: getting more visibility for operations research is great, but we certainly don't need bad PR. (The prices of airline tickets already provide plenty of that.) So can we pass the buck? Here comes the spin minute. As much as I loved the "really hard and really important" comment and the "pretty smart people" one, if it is about directing customer behavior, it is a lot more about operations management than operations research. (Remember what I wrote last week about Pennsylvania blaming New Jersey for all its troubles and then some? We all need a cousin to dump our problems on.)

Wikipedia defines operations research as "an interdisciplinary branch of applied mathematics and formal science that uses methods like mathematical modeling, statistics, and algorithms to arrive at optimal or near optimal solutions to complex problems." In other words, it deals with the quantitative models behind decision-making. Academics who work in operations research are typically found in engineering schools, often in industrial engineering departments. OR graduates are hired by companies like Amazon.com to satisfy customer order at the lowest cost (for instance, they develop large-scale algorithms to analyze sales data and decide from which warehouse to dispatch the goods.) Operations management, according to the same source, "is an area of business that is concerned with the production of goods and services, and involves the responsibility of ensuring that business operations are efficient and effective. It is the management of resources, the distribution of goods and services to customers, and the analysis of queue systems." OM people are typically found in business schools, and tend to have a more "integrated" approach to decision-making, in the sense that they take more into account the fuzzy factors behind the problems at hand (customer satisfaction, supplier relationship, and the like) and are less mathematically-oriented. (This is, of course, a generalization, and as in any generalization, you can find people who don't fit neatly in my simplified overview of OR versus OM, like this one.) They also tend to interact more closely with top management at companies. In OR, you analyze the data and design computer models to optimize profits by assigning values to different input variables. Computers can't output "create a new type of employees to make your customers feel guilty".

Devising policies that rely on customers' good heart (no, I will not let poor little Starbucks waste a latte by leaving the line!) is not quantitative enough to count as OR. Opening upscale stores so that customers are willing to pay more for coffee isn't quantitative enough either. (Nice spin, huh? Life in the Lehigh Valley has served me well.) Operations research would be about the optimal shipment strategies for coffee beans and staffing policies in the stores - you might need only two employees off-peak but four in the morning or late afternoon when people start or leave work. OR quantitative techniques are what is behind Starbucks's former executive vice president Ted Garcia's success story: "Through his prior experience developing manufacturing and logistics networks, as well as information systems and technology platforms, he was able to establish programs for Starbucks that have delivered savings totaling more than $250 million."

But the fact is, OR and OM people are in this boat together - they help companies become more profitable. OR/OM success stories are case studies where academics help business make a lot of money. They only benefit the general public in the sense that the extra cash flow creates or maintains employees' jobs, brings more dividends to shareholders, which might include your pension fund, or keep prices down by helping companies stay in business despite competition. Good decision-making in business seems, at first glance, to benefit companies and hurt the average customer - the goal is, after all, to make clients pay as much as possible. Blaming greedy big businesses is also a popular theme in some media outlets, especially local newspapers. In working to help OR/OM gain more visibility, we risk letting others give it a bad name, unless we relentlessly point to concrete benefits such as low prices (Wal-Mart is an operations research expert too). But having OR in the limelight also means we get to take all the credit in articles such as this 2004 piece in the Boston Globe: "Operations everything [..] a primer on the most influential academic discipline you've never heard of." In it, the associate director of the school of operations research and (then) industrial engineering at Cornell University defines OR as: "the effective use of scarce resources under dynamic and uncertain conditions." No mention of pop psychology to make customers stay in line. The Globe article is a great read because it takes the average person's vantage point and shows how OR benefits his or her life (when the person books a flight on Orbitz and heads to Disney World, for instance), and it correctly identifies many of the trends that have made OR so popular over the last few years. ("After decades in which the field's progress was mostly theoretical, computers have finally gotten powerful enough to collect the data and deliver the problem-solving solutions that O.R. has been promising since the heady days of the New Frontier.") We need more articles like that.

Comments

[FF]

1- not all customers buy stuff which need to be made by the barista.
2- reordering the drinks ordered by customers can help the barista to make them faster. It is easier to make two cups of Mocha and then two Iced Skinny Cinnamon Dolce Latte, than making one cup of Mocha, then one cup of Iced Latte, then another cup of Mocha and finally another cup of Iced Latte.

Because of the above, the expediters or as they used to be called floaters do expedite the process.

[Aurelie]

Grouping orders wouldn't necessarily be a good business practice because customers would become annoyed if someone who arrived after them got his order first.

[FF]

I agree, but the expediter is used when there is a long line. Now if the drink of the 12th customer in line is made before the drink of the 11th customer, by the time they pay the cashier, both drinks are ready and they won’t notice anything.