Restaurant sales

How AI-powered menu grouping can boost restaurant sales

For restaurant and convenience store marketers, campaign selection is a challenge. They should factor in the cost of guest messaging as well as any discount or reward to include. They must also weigh the relevance of the message and the end monetary goal.

Too large an offer can break the campaign budget, but too narrow an offer may result in low engagement.

Therefore, marketers need to pair their intuition with something more concrete when trying to focus on the sweet spot in the campaign. Artificial intelligence (AI) is a concrete tool that allows marketers to use all the data at their disposal to achieve the desired results:

  1. Reduce the cost of the offering and increase the return on investment.
  2. Activate more personal, relevant and targeted individual offers for each guest.
  3. Easily adjust and replicate these offers to save time.

With AI, marketers can use the mountains of data collected through their loyalty program to launch campaigns that target customers based on their unique behavior, ensuring relevance, savings, and ultimately, a huge return on their initial investment. At Paytronix, we call this “AI to AI” the practice of using artificial intelligence to drive individual action.

This does not mean that the best performing marketing departments are run by computers. Instead, AI is an asset that helps marketers identify underlying patterns and make reasonable predictions about customer behavior. While a brand may choose to let AI determine every element of the campaign, from the guest segment to the offer delivered, it’s more common to see marketers using AI to improve their own engagement. decision.

A human touch is often still beneficial. AI can optimize any variable, but the mind of the marketer must guide what variable it should be.

For example, an algorithm designed to segment alcohol buyers might identify the optimal group such as those who have bought alcohol four times in the past month, but that group may be too small to really move the needle. Instead, the marketer might go for a different segment of guests who have bought alcohol twice in the past month, as they are targeting a much larger group.

Ultimately, the best results come from the collaboration between humans and AI.

Grouping

Clustering, or k-means clustering, involves one of the most popular machine learning algorithms and provides a great example of driving AI to AI. Simply put, AI identifies points in a dataset, then creates closest clusters while keeping each cluster as small as possible. In other words, it is a method of grouping data points to identify similarities that are not immediately apparent.

But what does this mean in practice?

A bundling app allows marketers to choose an item or category they want to develop, such as a new burger on the menu or a high-margin offering like alcohol, and find the purchases that motivate customers who buy it. These guests can then be grouped into groups and receive offers tailored to their specific buying habits in the hope of enticing them to purchase the designated item.

Let’s take a look at a use case.

Creating clusters for a Burger Joint

The “menu grouping” illustrates how this process works. It is the practice of using AI to segment customers into groups based on their past purchasing behavior. There are four main use cases for menu grouping:

  1. Push members towards the desired behavior.
  2. Encourage members to continue the existing behavior.
  3. Position new products correctly.
  4. Match the right reward to the right guest.

In this case, the brand has identified four key segments: customers who buy meals, customers who buy French fries, customers who buy milkshakes, and customers who buy vegetable products. Next, the brand reflected on the offerings that would lead customers in each segment to a desired behavior.

The brand introduced a new limited-time burger. Armed with the segmentation information provided by AI, he launched four unique campaigns to bring customers an offer that made sense to them.

Using guest segments, the burger brand’s marketing team crafted unique offerings that specifically addressed guests in each segment and met each menu grouping goal:

  1. Push members towards the desired behavior. Customers in the dining segment always add fries, but only occasionally add a drink to complete the meal. Since these guests are rewarded with double the points, they are more likely to add the drink to their order.
  2. Encourage members to continue the existing behavior. Milkshake shoppers are high value guests as they add dessert for a larger basket. The brand encouraged these guests to keep buying shakes by offering them a reward on the third purchase.
  3. Position new products correctly. Customers who have only been identified for the regular purchase of French fries likely have a fairly large variance in the rest of their order, making them more likely to try new things. By seducing them with a free accompaniment, the brand encouraged these diners to try their new burger.
  4. Match the right reward to the right guest. A post promoting the new beef burger would be off-putting to vegans and vegetarians. Because the brand knew the eating habits of its members, it was able to offer this group a relevant reward: a free vegetarian burger.

Now for the ultimate question: Were these campaigns successful?

Yes! The campaigns, which only ran in-store, delivered an impressive 61-fold ROI.

The bottom line

Clustering enables brand marketers to better understand their customers and deliver personalized and one-on-one communications and promotions. AI-to-AI campaigns can save money because they cast a smaller net on a more motivated set of guests, delivering offers that are more relevant to them. In turn, this relevance drives customers to act more urgently and to visit more frequently.

AI-to-AI campaigns also reveal patterns of behavior that may not be apparent to the human eye. It allows marketers to communicate with customers in new and exciting ways.

Data analytics expert Lee Barnes leads the Data Insights team at Paytronix, a leading provider of reward program solutions whose customer engagement platform helps over 300 restaurant and retail chains retail to manage and increase customer spending by more than $ 18 billion. Barnes is a self-confessed data geek who can often be found digging into data with his team to optimize customer engagement with over 165 million loyal customers, through mobile, social and digital marketing tools. the most innovative today. His math degree and MBA from Harvard Business School give him the unusual ability to both run complex analyzes and translate the results into ideas for business leaders to use.

CRM covers