The Steak ‘n Shake Milkshake Button — Improving the dining experience

Design Case Study for uIDP

Milkshake button and light on each Steak ‘n Shake dining room table

Short Summary

🍦Here is the short version of the Steak ‘n Shake project if you’re in a rush…

1) The Problem

Steak ‘n Shake wanted to improve the quality of service and interaction between servers and customers in their dining area.

“American consumers are willing to spend more with companies that provide outstanding service, and they will also tell, on average, twice as many people about bad service than they are about good service.” — National Restaurant Association

2) Who I Worked With

I worked alongside three hard-working designers: Mary, Ryan, and Shelby. We worked on this project for 4 weeks in our undergraduate Interaction Design Practice class.

3) What We Designed

We designed a milkshake button and light ordering system based off the research we found that Steak ‘n Shake customers order milkshakes at various times during their visit. From our user tests in a real Steak ‘n Shake we found that the milkshake button and light improved customer service and improved their overall experience while dining.

Left: Milkshake light both on and off — Right: Milkshake button top and side view

The Problem

Steak ‘n Shake wanted to improve the quality of service and interaction between servers and customers in their dining area. There were many different paths we could have taken to help improve this problem but we chose to constrain ourselves to milkshakes and the milkshake ordering process. We chose to focus on milkshakes for two reasons:

  1. We all had experiences at Steak ‘n Shake where ordering a milkshake took forever! We asked around and this was a common theme among everyone we talked to. Milkshakes were a Steak ‘n Shake pain point but also an integral part of their brand.
  2. Steak ‘n Shake milkshakes are delicious and our team wanted to spend 4 weeks drinking as many as we could for “research” 😂

Research and Exploration

Online Research

We started out by doing research online to find what customers were saying about Steak ‘n Shake service, specifically what they were saying about their milkshake ordering experience. We read customer reviews on Google, Facebook, Twitter, YouTube, Yelp, and Reddit. We also visited other ice cream/food places around Bloomington, Indiana such as Jiffy Treat, Dairy Queen, Hartzell’s, and Culver’s.

What we learned:

  • Making hand-dipped milkshakes is a slow process — We learned that most customers don’t realize how tedious the process is. We sure didn’t when we started doing research!
  • Steak ‘n Shake servers make the milkshakes — They also serve their tables and do other small tasks which makes being a server at Steak ‘n Shake really busy and stressful.
  • Mistakes are often made when servers are forced to rush — Being able to staff enough quality servers seemed to be a problem. Servers being stressed and busy led to many customers being frustrated because their milkshake order was messed up or took a long time to come out.

Primary Research

We then talked to Steak ‘n Shake servers, managers, and cooks who we were able to contact through friends or calling around town. Current and past Steak ‘n Shake workers proved to have incredible amounts of insights into what they believed was both good and bad about the Steak ‘n Shake customer experience.

We created a survey that focused on the service at Steak ‘n Shake and the milkshake ordering process. We got an enormous amount of responses (194 responses to be exact!). From this we were able to focus on the most interesting insight we received which was people order their Steak ‘n Shake milkshake at different times during their meal. Some people like to have a shake before their food comes out while others wait until after. Some decide on the spot during their meal. We learned ordering a milkshake was most times a spontaneous decision for customers.

Milkshakes are a huge part of the Steak ‘n Shake experience for customers
“People order their Steak ‘n Shake milkshake at different times during their meal. Some people like to have a shake before their food comes out while others wait until after. Some decide on the spot during their meal. We learned ordering a milkshake was most times a spontaneous decision for customers.”

What we learned:

  • Over 60% of our respondents go to Steak ‘n Shake specifically for milkshakes
  • Customers order their milkshakes at different times of the meal
  • Under 2% of our respondents used the Steak ‘n Shake app
  • Over 90% of our respondents thought that Steak ‘n Shakes service and overall experience could be improved

Our Design Goal

“Decrease the time it takes for a server to know when a table wants to order a milkshake and increase overall customer satisfaction by allowing customers to be in control of their milkshake ordering process.”

Brainstorming Sessions

Early Sketches, User flows, and Brainstorming Ideas

Here are some early ideas and sketches we had when thinking about how to improve the milkshake ordering process. Our brainstorming ideas included drones, milkshake carts, a jukebox, iPhone apps, buttons, lights, milkshake bars, and more! As a team we quickly voted on which ideas seemed to hold the most weight and be a good starting place to explore. We chose the button and light idea.


Our milkshake button/light prototype

We built our physical milkshake button/light prototype in under an hour. All it took was the right pieces, a little bit of creativity, and boom! It was real enough in the sense that it gave the user the satisfaction of pressing a button, seeing a “milkshake” lamp light up, and a server walking over. We now had something we could test with Steak ‘n Shake customers and servers in order to get feedback on.

Usability Testing

Hands down performing our usability test in a Steak ‘n Shake restaurant with real customers and servers was the best decision we made during our project. We worried about whether we would get into trouble with management or distract customers but it turned out everyone really enjoyed giving us feedback and seeing the prototype in action.

Going to Steak N Shake!

We packed up our lamp, button, and a long extension cord 🔌 and headed over to our local Steak ‘n Shake. In hindsight we probably should’ve let the manager know we were coming but the Steak ‘n Shake staff was cordial enough to let us test our prototype in a booth near the back! It was exciting to see our prototype in a real Steak ‘n Shake booth. Our sketches and idea had come to life and it was now ready to see how real customers reacted to it.

Testing the Server

We started off by explaining our idea to our server and seeing if she would play along. She was too busy to go through the entire interaction and usability test for a server but she ended up giving us some really insightful feedback about the prototype.

Some feedback we received from servers was:

  • “This would be a cool way to be alerted that someone wants a milkshake if we were staffed properly.”
  • “As long as people don’t abuse it, such as kids or drunks abusing the button and light, then I think this is a great idea.”

Testing the Customers

We then started testing real Steak ‘n Shake customers to see how they reacted to the button and light. The feedback and experiences from the customers we tested were overwhelmingly positive!

Some feedback we received from customers was:

  • “This is a fun idea and it gives me more incentive to order a milkshake.”
  • “The server could be annoyed by multiple hits. Also, this seems like it would not just be specifically for milkshakes but I could order anything I want from this button that alerts my server.”

Getting More Feedback

After we tested our idea in Steak ‘n Shake we created another survey with our video attached above demonstrating how the ordering process would work. We then attached a few questions asking people specific things about the light and button idea to solicit additional feedback.

What We Learned

We learned so much from testing our prototype in a real Steak ‘n Shake. We learned and began to consider the following:

  • Button abuse was a huge concern — How could we give the server power over the button being abused? How could we make sure the server wasn’t slave to the lights going off all over the restaurant? How could we ensure the button wasn’t used for anything other than ordering milkshakes?
Have more than 2 people in a booth? It starts to feel crowded with the stuff on the table
  • Dining table was crowded — As soon as we sat down and set up our prototype in Steak ‘n Shake we realized a huge problem. The addition of the lamp and button crowded the table which already felt full with the condiment caddy and advertisement stands. How could we solve this problem?
  • Steak ‘n Shake servers are super busy — By visiting a Steak ‘n Shake we saw, firsthand, how busy each server was with their tables. Would the button being pressed put more stress on the server?
Busy Steak ‘n Shake employees

Final Design Solution

Below are the final designs we created for our milkshake button/light idea. We used the insights and feedback we received from our continuous user testing to reach our final design.

Increased Empathy for Servers

Milkshake light before and after button has been pressed

Its our hope that our milkshake light design helps Steak ‘n Shake customers be empathetic to the Steak ‘n Shake servers because they will be able to look around the restaurant and see how many milkshake lights are on and how busy their server is at any given moment. For our milkshake light we focused on using the existing retro milkshake glass and then lighting it up with the Steak ‘n Shake vintage neon red lights already seen around the restaurants nationwide.

Milkshake button design from top and side view

Servers have Control

We solved many of the button abuse issues by creating a locking mechanism that allowed customers to press the button once. After the button has been pressed the server must unlock the button from their touch screen in the back. This gives the server ultimate control of the buttons for each table, while still giving the customer the ability to notify their server when they are ready to order a shake. We chose to combine this touch screen into the already existing ordering system currently in place at Steak ‘n Shakes nationwide. Finally, we designed a milkshake vector and placed “order” on the button in hopes of better identifying the button as a milkshake ordering device. This was another way we tried to protect against the button being abused.

Combined Light and Condiment Caddy

Combined condiment caddy and light

In order to not overcrowd each dining table we designed a condiment caddy that also doubled as the base for the milkshake light.

Zoomed out view of our final design

Future Ideas to Implement

The hard thing to swallow is that there are always more ideas and features to implement. Time only allows for so many ideas to be fully tested and used. Its part of the beauty and difficulty when designing digital products. Here are a few ideas we could see being implemented in the future:

  • Server screen that identifies which table pressed the button first and who should be served first. It could help the servers stay organized.
  • Expanding the button for all drink orders or even anything a table needed a server for at a given time.
  • A milkshake light that adjusts its brightness depending on how long the table has been waiting for their shakes.

What I Learned

  • Invest time in creating realistic prototypes and do it fast — Before this project and before I read “Sprint” by Jake Knapp, I undervalued creating realistic prototypes. I assumed testing users with paper prototypes was good enough to gather insights. Boy was I wrong! Taking the time to create a realistic looking and “working” prototype (even if you are faking the backend) is worth it hands down. However, its important not to waste weeks building a realistic prototype. Being creative and making something in a few hours is a great approach.
  • Test your prototypes with real users in real situations — Similar to my last point, taking that realistic prototype and testing real users in real situations changes the game. If you take your realistic prototype into the actual environment it will be used in, you will get honest feedback. The flaws and shortcomings of your design become glaringly obvious when you test it in real situations with real users.
“Its about illusion. You’ve got an idea for a great solution. Instead of taking weeks, months, or, even heck, years building that solution, you’re going to fake it. In one day, you’ll make a prototype that appears real, just like that Old West facade… Your customers — like a movie audience — will forget their surroundings and just react.” — Jake Knapp, Sprint
  • Don’t underestimate a simple idea — We were initially afraid to design a simple button and light concept. We were tempted to try and solve the customer service problem with a flashy new app for Steak ‘n Shake. However, we found out early on through our research that nobody used the Steak ‘n Shake app. I realized through this project that you can never have to simple of an idea and usually the simpler the better! Also, I realized that when we approached the problem by thinking outside the technology box, we were able to come up with a solution that worked and then see how we could make that solution better with technology instead of the other way around.

Thanks for reading hopefully you aren’t craving a Steak ‘n Shake milkshake too much right now! Feel free to reach out or following me at any of the links below…

Website | Dribbble | LinkedIn | Twitter

If you enjoyed this article help me share it with others by giving it a 💓!