How An Apple Is Changing the Quick-Service Restaurant Industry

Human-based sales processes represent an important dimension where Quick Service Restaurant (QSR) brands have a lot of space for improvement, despite all the work done to increase efficiency over the last decade.

The digital revolution is driving a transformation in how QSR brands sell. However, the rapid pace of the transformation and the role that aggregators are playing in this process make predictions blur together and cause the predictions to be very speculative.

Aggregators like Uber Eats and Grubhub have become a critical element in the commercial strategy of QSR brands, mainly because they bring in customers and sales; however, at the same time, they strongly erode profitability. Although brands are trying to be creative in optimizing their own ordering apps to compete with aggregators, the difference in scale (number of customers reached) between the two is immense. One major challenge faced by brand-owned ordering apps is the imposition to customers to have multiple brand apps installed on their mobile phones versus having all of them available in one single aggregator app.

Although many challenges still remain to be addressed for a comprehensive digital transformation in QSR sales, big changes are on the horizon. Humans will have their dominance diminished in the QSR industry, like in many other industries. However, the question is how quickly things will change. In my view, the replacement of humans in the QSR industry will move slow, even in more mature markets like the US and Europe. One such digital technology that is said to be around the corner for QSRs is intelligent virtual assistants. Intelligent virtual assistants are meant to enable machines to help customers in the ordering process, especially in the drive-thru channel. This is definitely a breakthrough. However, in order for this solution to be ubiquitously effective, a very complex system is required to be in place, which depends on important technological pieces that still need to be developed and orchestrated. In other words, this and other disruptive digital technologies will take some time to develop for QSRs.

Realistically, I believe the brave cashiers will keep their hegemony on doing the best they can to execute arduous sales strategies for a reasonable amount of time, whether in the counter or in the drive-thru. But what does arduous mean? 

Value meals (group of menu items offered together at a lower price than they’d cost individually) is a key strategic marketing instrument for QSR brands in any market in the world since they inspire customers to buy. The tactic is highly efficient but, by principle, there’s no free lunch in this industry. Lower prices mean lower margins which impose brands the need to cross-sell in order to increase ticket size and offset discounted prices.

Although the understanding of the challenge is clear, addressing it in an effective way is a very different and complex story.

Effective cross-selling in the QSR industry means recommending relevant products to customers over the sale process. Yes, doing it right imposes that recommended products must be relevant which means that different combinations of customers and contexts (hour of the day, day of the week, weather, etc.) require different products to be recommended. And to make it even more challenging, recommendations need to be executed quickly so speed of service is virtually not impacted.

Although digital sales have skyrocketed due to the pandemic, a major share of QSR sales is still executed by humans which are, by essence, unable to effectively operate in this highly complex sales environment. And this is a common problem for all of us, humans. Managing effectively a highly diverse range of possibilities in a time pressured environment is, unfortunately, impossible for any of us, regardless our level of education, training or IQ.

From a business point of view, the problem of optimizing product recommendations on each order is real (the unaddressed need), critical (improves sales and profits) and complex (the solution involves rapid contextualization and personalization decisions).

I was presented with the opportunity to address this problem in 2017, when I started conversations with a global QSR brand operating in Brazil. After engaging with other brands both in Brazil and abroad, I was convinced that this was a common challenge for every single QSR brand in the world. It was the trigger for a new ambitious AI-driven solution development journey.

Understanding QSR sales processes and developing a scalable and easy to adopt solution took me a couple of years. Convincing the first major QSR brand to test it was also exceptionally tough, but it happened.

If you are connected to the QSR industry and looking for something very pragmatic to make a difference to your business, I invite you keep reading.

xSell is an AI-based solution specific for the QSR industry that provides highly relevant cross-sell recommendations for every sale in every channel without impacting speed of service.

At the beginning of this journey, it was apparent that QSR brands did not clearly understand how powerful AI could be in optimizing complex, critical recurring decisions. It took some persistence to convince a brand to jump into the confusing and complicated world of AI models and digital transformation.

Eventually a group of brave executives in a very dynamic and innovative QSR brand in Brazil embraced the idea of running a first experiment at a drive-thru store. It was an uncharted world for all of them and a unique opportunity for the technology to prove its value towards increasing sales.  

As expected in every transformational process, the most challenging part was related to change management, but at some point, all the key pieces between IT, Marketing and Operations were aligned, and a pilot was enabled. Sales increase emerged right away. 

As results and expectations were confirmed, executives became more involved and familiar with how the technology worked. From there, business-driven questions and challenges naturally started to arise and at one point early on I got a very unusual request: to change the recommendation algorithm to stop xSell from recommending apples. Yes, apples. It turned out that the AI model had learned from historical data that in some ‘not-so-unusual’ situations an apple was the best product to be recommended. Although overall confidence in the solution was high at that point, recommending apples to customers was seen as a mistake or defect. 

Given we’re all “complicated” people with preconceived notions and biases, this sort of assumption is not surprising and, in fact, very understandable. Apples are not what most of us think about when we visit a QSR store. Luckily, on top of complicated, people are also reasonable and open (not all of us, we need to acknowledge that) to give the benefit of doubt to reasonable explanations, and that’s what enabled the recommendation algorithm to remain intact.  Fortuitously, the following day, a business partner supporting the pilot on-site was approached by a cashier with a smile from ear to ear who said, “I just sold a recommended apple!” He said the apple recommendation appeared on his screen at the end of the order and asked the customer if he would like an apple to complement his meal.  He was convinced the customer would laugh and say, “Why would I want an apple?” Instead, the customer had a quizzical look on his face and said, “OK, that sounds great.” This and other experiences like it helped to convince cashiers, our partners and executives that the data-driven xSell solution was not only quick but extremely accurate.  Over time, the confidence among stakeholders has grown and they have been convinced to let the data talk.

Image source: The Week Magazine

Although the apple case as well as all other recommended products sold are compelling signs of an important transformation on how QSR brands can improve sales, they represent only one dimension of the value. Since any change in a human-based sales process is complex and very sensitive, the pilot relied on what we at A2Go call “primary approach”. This approach provides recommendations at the end of the sales process, which makes the process simple. Cashiers are oriented to simply exchange the common final question “Anything else?” by a value-inspired question like “How about an ice cream for $1?” where the ice cream is an xSell recommended product that pops-up in the POS screen. This approach virtually avoids any change in the existing culture and sales process, which makes adoption a very straightforward and smooth process.

Results achieved with the primary approach suggest that xSell can consistently generate 3%-5% increase in sales to offline channels. For digital channels, a pilot with a different customer is providing the evidence to show the reward is even greater.

Once the primary approach is in place, initial value is unleashed and sales increase is confirmed, it’s time to expand the value by taking xSell to the whole offline sales process through what we call “complete approach”.

In the complete approach, xSell is fully embedded in the sales process managing all products to be recommended to customers over the whole sales journey. By doing this, brands become able to operate a very systematic process where humans are responsible for embracing and serving customers, while AI provides the best products to be recommended over the journey. This is a strong example of collaborative intelligence where we combine the best of AI (analytical capability) with the best of humans (creativity and socialization) to optimize the outcome. Collectively, the AI-human collaboration, automation and monitoring of real-time granular performance indicators comprise the landscape for sales increases that can reach 10% or more. Since the sales increases are purely enable by technology and not reliant on marketing investments or operational changes, the incremental gross margin increase is even bigger. 

This level of sales increase may sound unrealistic for a mature industry like QSR, but it is exactly what xSell is about.  It’s set to enable all brave and innovative brands to benefit from a very complete, easy to deploy and use QSR specialized AI-based solution.

There is no way for me to deny that this article can be seen as a piece of marketing, but believe it or not, this is not my main intention. I have been involved with amazing brands and brilliant people in the QSR space for a few years now, and I'm impressed with how hard companies and people work and struggle to maximize their commercial performance. More than that, it’s heartbreaking to see how cashiers are strongly and frustratedly pushed to deliver something they’ll never be able to deliver. Improving this environment both from a human and a business point of view is the essence of xSell.

If you are interested or even intrigued to know more about xSell, I'll be delighted to talk about it. 

Rafael Fanchini - Chief Product Officer - Analytics2Go

rfanchini@analytics2go.com | +1 (305) 763-1905