Papa John’s serves up AI for more efficient ordering

Chain restaurants of all culinary persuasions have experimented with virtual assistants, which provide another digital channel through which customers may order food. However, few quick-service chains are using artificial intelligence (AI) to help humans take orders more quickly and personalize service.

Here, Papa John’s thinks it’s cracked the code with its AI-assisted call center. The call center initiative, dubbed PapaCall, proved critical during the COVID-19 crisis in enabling consumers leery of human contact to order pizza, beverages, and other food, says Justin Falciola, Papa John’s chief insights and technology officer, who oversaw the rollout of the service.

[ Cut through the hype with our practical guide to machine learning in business and find out the 10 signs you’re ready for AI — but might not succeed. | Get the latest insights with our CIO Daily newsletter. ]

“Like most companies, we were reacting in real-time [to the coronavirus ramifications],” Falciola tells CIO.com. “We had to keep employees safe and do the same for customers, while continuing to provide the best customer experience.”

Domino’s and Papa Johns have paced the pizza-making pack in experimenting with virtual assistant software, such as Alexa or Google Assistant applications, to help log orders. But the results have been tepid at best, as most consumers are more comfortable ordering food online from their computers and phones, rather than using their voice.

The new call center ops recipe

With PapaCall, Papa John’s has put a different spin on leveraging AI to serve customers. The company has built an AI engine that feeds call center agents information when customers phone in an order, reducing the time to process orders and enabling them to focus more on making and delivering pizzas. This approach took some deliberation, as Falciola admits the company has only recently begun to view voice as a digital channel.

“As we started discussing this—both from a how do we lift some of that [phone ordering] effort out of the store and how to do this right from a tech perspective—we knew we had to do things differently,” Falciola says.

Papa John’s turned to long-time partner Cognizant for help using AI and machine learning (ML) models to assist the human agents. “While digital channels have matured, the experience on voice has not kept pace,” adds Sandeep Bhasin, Cognizant’s North American head of digital business operations, who assists clients in retail, consumer goods, and travel and hospitality.

Cognizant engineers helped build PapaCall from scratch, building new business logic and interactive voice recognition (IVR) software tailored to Papa John’s ordering experience, including mastery of the menu and all potential order permutations. “All this info and data is coming together across all these systems, enabling us to try new things,” Bhasin says.

When a customer calls, the cloud-based telephony system and other logic will recognize the phone number, and the agent greets them by name and has customer preferences available to make the conversation contextual. Today, PapaCall is available in over 1,500 North American stores, with the company onboarding more locations regularly.

This CX experience fueled by AI

Papa John’s is further looking to use AI to expand its personalization capabilities, which will empower human agents to process orders, assist customers and make relevant product suggestions.

Such capabilities will enable PapaCall to greet a return customer by name, thank them for being a loyal customer, and make recommendations based on their purchase history, including something as basic as asking whether they’d like to reorder what they ordered the prior week. Order after order, these “Customer Hub” algorithms feed the IVR, agent desktop, and recommendation engine, which will get more tailored to each customer over time.

If the customer’s need is not fully resolved by the IVR, it will route the call to a human agent for resolution, with analytics helping the agent best serve the customer. 

Eventually, the system itself could—like any good human salesperson—try to upsell to customers by recommending new menu items based on prior orders and other factors. These processes require a delicate orchestration between the IVR, machine learning models, and back-end systems.  “The guiding principle is to provide an experience that is customer first, always,” says Falciola.  

The analytics will also incorporate external data to local store context, such as weather patterns and promotions. “It’s about how to make the agent experience feel as local as possible,” Falciola says.

Also critical to the CX is ensuring that the system prompts the human agent with what to recommend versus what not to recommend.  For instance, a customer who has only ordered vegetarian pies will likely not wish to be offered meat lover’s pizza. Such presentation requires continuous A/B testing, learning from the results and adjusting accordingly, Falciola says.

Tech fuels a better CX

Falciola is quick to point out that customers don’t think of the company’s digital channels as “discreet tech interactions” but about how to feed their families. It’s a philosophy that stems from Papa John’s CEO Rob Lynch, who instills in his senior leadership the idea that the company isn’t about building tech for tech’s sake, but about innovating on the menu.

Just as important is considering the whole customer journey map from the moment he or she decides they want pizza through its delivery. Did the customer enjoy their food? Was the experience seamless? What is the likelihood that customers will order again? Papa John’s cares about these considerations above all else. “Internally, we have a phrase: customer first always,” Falciola says.

In polling customers about Papa Call, Papa John’s has clocked a 95% customer satisfaction score. Now that’s a winning recipe.

More on AI and machine learning:

• Reskilling IT for the AI era
• 9 IT projects primed for machine learning
• 6 secrets of successful chatbot strategies
• How AI is revolutionizing business productivity
• 5 machine learning success stories: An inside look
• 10 signs you’re ready for AI — but might not succeed