I’m a social scientist who specializes in ethnography. Focusing on the user is at the forefront of my work, and it also seeps into my personal life, because I am a user of so many things myself. As a result of my hyper-awareness of UX, I’m often tortured by my own experiences with products. One area in which I see need for improvement is the experience of mobile ordering versus physically standing in line.

Like most people, I love to avoid lines whenever I can. I was an early adopter of Fastrak, which allows one to bypass the toll booth when driving over bridges here in the Bay Area. I use CLEAR, Global Entry, and TSA Pre✓ when I travel to reduce waiting and make transitions predictable. I also have a United Airlines credit card, which automatically moves me up to Group 2 and takes the stress out of finding room in overhead bins. My goal in using these services is to reduce the unpredictability of wait time and increase the efficiency of my travel experience. I can “cut” the line without physically (and awkwardly) stepping in front of anyone in the process.

One place I’ve always been fairly content to stand in line is at a restaurant. For instance, I go to Chipotle a few times a month, typically for lunch. It is often a go-to choice when I’m traveling for work because it is inexpensive and the menu items are predictable and fairly healthy. A while back, while standing in line at Chipotle, I started noticing that people were cutting in line without actually being in the line. That is, I witnessed the employee who was about to serve the next person in line suddenly working on a complex order from a piece of paper instead. Without witnessing anyone answer the phone, it became obvious that Chipotle was prioritizing the customers sending in to-go orders via the restaurant chain’s app. My mental model of the queue I was standing in — the general societal agreement about the order of lines and the process of being served — was in flux as I suddenly wondered why Chipotle would insult patrons already physically in the store waiting to be served, an experience that is even more aggravating when there are 15-20 people in line.

I did not immediately join the growing movement of customers cutting the line by downloading Chipotle’s app myself. I was hesitant to use an app to order, for a few reasons:

  1. I like to see my food being made while I’m in a restaurant. It reassures me that the food is fresh and hot.

  2. The lines at Chipotle are rarely so long as to warrant what I at first perceived as “cutting.”

  3. I didn’t want to be that person cutting in line, whether people could see me or not.

This last point is an important one. When companies like Chipotle are explicit about allowing customers to “cut in line” by using an app, it causes frustration for their other segments of customers. By explicit, I mean that customers can clearly see that the person who would otherwise be working on their order is now working off of a piece of paper instead. In essence, Chipotle breaks a rule of common courtesy with this protocol.

Of course, restaurants could easily solve this problem by moving the app-orders work process to another area of the restaurant, out of sight of customers ordering in person. For example, Chipotle could have a separate food line in back for employees to fill online orders, and have employees dedicated to this task. But presumably this is not an economically viable alternative for most restaurants.

At any rate, once I began to notice the phenomenon of Chipotle employees taking time away from physical customers to fulfill online orders, I began to notice this same disruption elsewhere. A new restaurant in San Francisco called Pokihub has a similar line setup to Chipotle but a much more complex system of menu choices that results in employees taking quite a bit of time to complete customer orders. And since it is a new restaurant based on a new concept (customizable poki bowls), customers have a lot of questions. Standing in line at Pokihub one day, I witnessed the queue suddenly come to a grinding halt as the employees became swamped with app orders. Once again, non-present customers were getting served first. “Oh no,” I thought to myself. “The Chipotle problem is spreading.”

You’re probably starting to think I have a hangup about standing in line. And you might be right about that. When I go to a coffee shop, I typically order plain old coffee to avoid waiting for a complicated espresso drink. Unfortunately, my partner insists on getting a cafe latte, so I am stuck waiting in front of the barista anyway. During one such wait recently at Starbucks, I noticed a sign advertising mobile ordering. Out of curiosity, I downloaded the Starbucks app and decided to try it out the next time we ordered.

A few things passed through my mind while doing so. First, where would I pick up my order, and how would I identify myself? The solution turned out to be intuitive. My drink was placed on the bar with all the other drinks. The order, attached to the cup as a slip of paper, clearly had my name on it. I also wondered how Starbucks employees feel about juggling various different kinds of orders, and I was pleasantly surprised to be impressed with the friendliness of the Starbucks staff to this order process. My third initial question was about timing. How would I need to time my app order in order to walk in the door and pick up a fresh, hot cup of coffee? The app indicates how long your wait time will be, and I’ve noticed that it’s usually about 5 to 7 minutes. So, I order via the app about seven minutes before I expect to arrive at Starbucks.

The Starbucks app payment is linked to Apple Pay, which makes it very easy to pay without standing in line. Any checkout process that eliminates even a quick point-of-purchase interaction is a win-win, in my experience and also in my professional research. And I have to admit, I love walking into Starbucks and getting my hot drink without having to spend any time in the store or talking to anyone.

My Starbucks experience a success, I decided to download the Chipotle app after all. I logged on to place an order but was dismayed to find that the app estimated about a 15-minute prep time, and since I was only five minutes away, I decided to walk in and order instead. This was at 1:13 p.m. I had my food and was checked out by 1:21 p.m., nine minutes ahead of the app’s estimation. Either the app is not smart enough to know how busy that particular Chipotle is or it intentionally pads time.

These two contrasting experiences have been very insightful for me as a UX researcher. I have come to realize that our socially agreed-upon idea of what it means to stand in a queue is changing as technology provides us with different kinds of queues and even helps us to avoid the feeling of being in a queue altogether. We have all had the experience of peeking into a store, seeing a long line, and deciding not to go in. We use a variety of factors in judging this decision: the number of customers in line, the number of workers per customer, and how efficient those workers are. Previous customer experience with the processes of the establishment help us to decide whether the line will move efficiently or not.

When we move out of the physical space to a technological one, how is the concept of a “line” assessed by the customer who can’t actually see the line? Technology is the only source of information we have about the length of the queue — and like with my experience with the Chipotle app, we can’t entirely trust the app. The most important factor in physical and online ordering is how long it will take; the trustworthiness of this information is paramount. One of the main challenges with mobile ordering is therefore being able to trust time estimations given by apps, as well as the actual time saved (or not) by using mobile ordering. And minutes count.

In this regard, from my own personal experience, Starbucks is doing a far better job than Chipotle with the mobile ordering. Starbucks app orders typically come with a 5-to-7-minute wait time. Chipotle estimates are always beyond 15 minutes. That’s a big difference. Another major difference I’ve noticed between the app experience at Starbucks and at Chipotle is what happens when a customer picks up the order. At Starbucks, there is at least one person dedicated to making drinks only. Whether you order at the counter or via an app, you go to the same place to wait for your drink, and the drink appears when it is ready. At Chipotle, though, a customer who has ordered online must physically interrupt the line of customers waiting to order and pay. In my mind, this is a compromised customer experience for everyone — the app customer, the customers physically waiting in line, and the cashier, who must juggle two conflicting sets of priorities.

What is my takeaway from all this?

  1. Companies thinking about creating a mobile-ordering app need to make sure that existing customer-order streams will be unaffected. Starbucks does a great job at this. Chipotle fails on two fronts: interrupting the flow of the physical line to serve app customers out of order, and asking those app customers to cut in line to pick up their orders.

  2. Time estimates for when orders will be ready are super important. If companies get these time estimates wrong, they lose trust of the customer.

  3. Companies must make it easy for customers to pay using the smartphone app. Starbucks links to Apple Pay, which makes payment fluid and easy.

As a user experience researcher specializing in ethnography, my forte is to immerse myself in a situation, ask questions of those intimately involved, and experience for myself as much as possible. I always see the value of experiential information to inform process or technology design. Companies benefit from ethnography to unearth information they might be overlooking because of their own cognitive bias. What might be obvious to the consumer who is literally or figuratively standing in line for your service is not always obvious to the creator.

James Glasnapp is a senior user experience researcher working in the Innovation Services Group at Palo Alto Research Center.