McDonald’s is the epitome of fast-food, and their survey is the quintessence of fast-food surveys. Typical of the restaurant industry, their survey focuses too much on generating useless scores and market research data instead of improving customer experiences. In doing so, they contribute to customer survey exhaustion. It hurts their chances of generating actionable insights, and it hurts others in the industry by wearing customers out.
In fairness to McDonald’s, they’re not the worst offenders. I like McDonald’s and my home location has been serving me well lately. In fact, the quest that inspired this article was to pay a compliment not a complaint. Instead of making it easy, McDonald’s and their research partner, SMG, offered me 29 thought-provoking questions. I’ve served up the lessons for you, so you don’t have to spend 15 minutes taking the survey.
Barriers to Entry
Like almost every other survey printed on a receipt, this one starts by asking the user enter a code. These codes are a necessary evil; they often are the only way to connect a response with the physical experience (time, place, products purchased, etc). McDonald’s asks users to key in the Big Mac of all survey codes, it’s a Supersized 26-digits long! That’s absurd! That’s almost a septillion (yes, really) possible combinations.
99,999,999,999,999,999,999,999,999 would be a lot of surveys!
If each of McDonald’s 37,241 worldwide locations processed 5 trillion transactions every second of every day, for the next 1,000 years, they’d still have unique numbers left over. Some data entry is unavoidable, but there’s simply no justification for a transaction identifier that long. If you must insist on keeping the long number, at least provide a QR code for easy entry.
Let Them Do the Work
McDonald’s has been slow to adopt self-service technologies in the U.S., at least when it comes to food service. When it comes to surveys, they insist you do their job for them. Even though you meticulously entered a survey code long enough to encode every attribute of your receipt, you still have to tell them how the order was placed, how you dined, if your order was customized, what you ordered, and if you ordered from the value menu.
It may have been cost prohibitive to link all of these attributes to the ridiculously long survey code, but that doesn’t justify offloading the expense on your customers. If it’s not worth it to pay for the integration, why would it be worth your customers’ time?
10 Things I Hate About You
On a scale of highly satisfied to highly dissatisfied, how many questions does it take to understand your customer experience? For McDonald’s, the answer is 10.
- “Please rate your overall satisfaction with your experience at this McDonald’s.”
- “Please rate your satisfaction with… the friendliness of the crew.”
- “Please rate your satisfaction with… the accuracy of your order.”
- “Please rate your satisfaction with… the quality of your food.”
- “Please rate your satisfaction with… the speed of service.”
- “Please rate your satisfaction with… the taste of your food.”
- “Please rate your satisfaction with… the temperature of your food.”
- “Please rate your satisfaction with… the ease of placing your order.”
- “Please rate your satisfaction with… the exterior cleanliness of the restaurant.”
- “Please rate your satisfaction with… the overall value for the price you paid.”
Asking for satisfaction levels in multiple categories is one way of identifying areas where improvement can be made, but it puts a huge burden on the customer if they are going to put any thought into their answer.
Another strategy is to start with an overall satisfaction, loyalty, or effort question and offer a list of check boxes for attributes that affected the score based on the first answer. This saves happy customers from monotonous questions, makes it easier for less happy customers to point out where things could improve, and makes it easier for analysts to spot trouble. It’s a technique I’ve implemented myself with great success.
To make matters worse, many of the questions should have been answered another way. Speed of service is objective and empirically measurable by timers, video surveillance, and other technologies. Once the brand has established a standard, using quantitative data to track performance is much more effective than the qualitative, biased data received by asking this question. The exterior cleanliness of the restaurant would be better tracked by managers on regularly scheduled walks of the property with a company approved checklist. What is acceptable and what is unacceptable has already been established elsewhere. Asking this question in a transactional survey is unlikely to bring any more to the table. It’s another example of how the company’s work has been offloaded to the customer.
Order accuracy is an excellent example of trying to fit a good metric into a bad question. Orders are either accurate or inaccurate. It’s pass or fail. GO or NO GO. This isn’t a high-school English class, there’s no burger rubric. How should a customer respond on a 5-point scale? If I ordered a double hamburger with no pickle and received double hamburger with pickle, is that a 4? What if I got a double cheeseburger without pickle, does that make it a 2? God forbid, they give me a cheeseburger with a pickle! That’s definitely a bottom box score. When it comes to order accuracy, there’s no partial credit. Looking at the average of a 5-point scale won’t tell you how often you actually get it right or wrong. It’s a useless metric.
What’s Your Problem?
Finally, after completing 16% (12 questions) of the survey, they asked if I had a problem. It was a yes or no question, so they’re doing a little better. On the other hand, they asked this question after a very thorough interrogation of my satisfaction levels. I assured them, repeatedly, that I was “highly satisfied.” Experience tells me that customers with problems don’t give you those kind of ratings. Even if I’d been “highly dissatisfied,” why bother asking if I had a problem at this point? It should be pretty obvious that something wasn’t right.
It would have been much more effective to ask this question first. Customers with fresh wounds don’t have the patience for your market research or the desire to give you a fair, honest evaluation of your service. If the customer’s motivation for taking the survey was a problem they experienced, arranging this question deep in the survey makes it more difficult for them to accomplish their goal.
We Expect the Highest Quality Comments
I was already a little suspicious that this survey had been designed by my high-school English teacher; she was a stickler for details. The mid-survey open response essay confirmed it. “Please tell us in three or more sentences why you were not highly satisfied with your McDonald’s experience.”
This question confused me for a few reasons, mostly because they had received very good ratings from me on this occasion. I didn’t know I wasn’t highly satisfied until they told me so. Second, what counts as a sentence? Do compound or complex sentences count as one or two? How about clauses that aren’t complete sentences, will they count at all? Will I lose points for improper grammar and punctuation? Fortunately, the “Comment Detail” gauge offered some insight to how my hamburger essay would be evaluated.
How much or how little customers are share is their prerogative. I love reading what customers have written for me, and I really appreciate it when they take the time to write detailed responses. Demanding three complete sentences seems a little bossy, and I’m tempted to write, “you’ll get what I give you and like it, much like your drive through.” If your customers aren’t writing enough, ask better questions.
In closing, remember, the customer’s experience taking your surveys is your customer experience. You must be as thoughtful and deliberate in the delivery of your survey as you are with your product. If you ever need help, I can be lured with bacon, egg, and cheese biscuits.
Photo courtesy of Mike Mozart (CC BY 2.0).
Andrew Gilliam is a passionate customer experience innovator and change agent. His vision: deliver Amazing Customer Service and Technical Support™. Learn more at andytg.com, follow @ndytg on Twitter, and connect on LinkedIn.