In today’s fast-paced world, quick-service restaurants (QSRs) are constantly seeking innovative ways to enhance their customers’ experiences while optimizing operational efficiency. In part one of our conversation with Revenue Management Solutions’ Director of AI and ML, Michelle Miller, we discussed the role artificial intelligence (AI) and machine learning (ML) play in the realm of QSRs).
Now, in part two, we will further delve into various use cases of AI and ML in QSRs, such as in-app customer service and AI-powered drive-thru experiences. Additionally, we will take a look at the necessary data requirements to effectively support and implement these technologies.
What’s the most promising use of AI and machine learning in QSRs?
Taking transaction data and creating customer-level insights. Using AI to track data over time, you can see how customer behavior is impacted by new pricing or new efforts (like a remodel). Restaurant brands now have the ability to see much more than we ever could before and so much faster. Operators can access those fine-grained details more quickly, meaning they can focus on implementation and not just data collection.
What are some non-back-office usages of AI and machine learning for QSRs?
- In-app customer service is one. For QSRs with mobile apps, the ability to interact with a customer-service chatbot for issues and questions can be a significant time saver. And as mentioned above, the more use the technology gets, the more it learns. This is key in becoming even more effective and useful for the franchisor.
- Personalized menu recommendations are possible for anyone who’s placed an order in the past. While brands should safeguard user privacy, they will have an order history for guests who have placed orders digitally or are members of a loyalty program. With this added personalization, restaurants can target guests with compelling upsell offers for a new item or the extra items they didn’t get last time.
- Social media engagement is another. Chatbots, powered by ChatGPT and other large language models, are helpful here. They can offer fast and reliable replies to social media tags. Or they can even detect mood or the language someone is using and then flag where a personalized response vs. a computerized one is needed. This use of AI and machine learning also ensures ongoing brand engagement on social media but without a heavy burden on staff.
- AI in the drive-thru might be the most promising use to date. Recommendations driven by daypart or customized in real time as the customer orders are becoming reality.
What kinds of data are required to support the above use cases? Should operators be worried about whether they have this data or not?
Operators most definitely have the data. But how easy it is for franchisees to access it is the question. The solve for this is a customized model which takes a base approach and then learns the brand specifics, such as what the flagship items are or how the brand positions itself in the segment. When a brand has this in place, the data is easily available to franchisees in a way they can act on it.
What will AI and machine learning not replace in the QSR world?
As Sam Altman, OpenAI CEO, said recently, AI-based systems are good at doing “tasks, not jobs.” In other words, AI is great at supplementing human expertise, but it can’t do the job for you. All the models, tools and technologies aren’t helpful if you don’t know how to use them. If you don’t have quality data, and it’s not used a certain way, or not put into a model that performs certain tasks, then it won’t work for you.
Leveraging RMS’ pricing services with AI and machine learning technologies built in, QSRs can move forward rapidly in the most profitable direction.