Implemented a natural language customer chatbot in less than 40 hours
Fully automated more than 25% of team member “call off” requests
Manages 50%+ of password reset calls with Customer Engagement Suite
Improves guest experience with automated invoice requesting using chatbot
Hershey Entertainment & Resorts (HE&R) leveraged Google Cloud Conversational Insights AI to analyze customer interaction data. Using that information, it prioritized automating the most frequent customer service interactions (like invoice requests and password resets) and now provides personalized responses to enhance its customer experience.
Hershey Entertainment & Resorts (HE&R) leveraged Google Cloud Conversational Insights AI to analyze customer interaction data. Using that information, it prioritized automating the most frequent customer service interactions (like invoice requests and password resets) and now provides personalized responses to enhance its customer experience.
Curating a relevant and engaging experience for everyone
Hershey has been a household name for well over 100 years. Chocolatier Milton S. Hershey created three businesses, including Hershey Entertainment & Resorts (HE&R), which serves as the stewards of the town he built for his chocolate company workers. HE&R boasts a large amusement park, professional hockey team, historic hotel, and entertainment venues. “The Sweetest Place On Earth” welcomes millions of guests per year from all walks of life and worldwide.
Because these visitors can vary so wildly, the company can’t simply offer the same experience to everyone. It wants to create engaging, relevant experiences for different audiences. “A family of four on vacation has very different needs than a VP looking to book an event,” says Mike McPoyle, Manager of Development and Integration, HE&R. “We want to do more than offer a one-size-fits-all approach to supporting our guests.”
HE&R has used data from customer service calls and web activity for many years to determine how it can improve the customer journey. The team would have to manually extract, review, and analyze data from Google Analytics.
This manual process was time-consuming, yet the data was still incomplete. Manually pulling and reviewing a truly comprehensive data set simply wasn’t possible, which limited the team’s ability to make the best decisions possible. The company’s need to make more informed decisions faster led it to look for more Google Cloud solutions.
“We used to rely on data sampling to review and iterate,” McPoyle shares, “but even if we’re reviewing a few hundred or thousand calls or user sessions, we had millions more that we couldn’t view.” The company decided to expand its use of AI to help ingest, review, and provide actionable direction based on its entire dataset rather than small subsets. “Anything we can do to customize and personalize for our guests improves their experience with us,” he says. “With Google Cloud and the power of AI, we can look at all of our data and find things that we probably would've missed if we were only reviewing a small sample.”
Anything we can do to customize and personalize for our guests improves their experience with us. With Google Cloud and the power of AI, we can look at all of our data and find things that we probably would've missed if we were only reviewing a small sample.
Mike McPoyle
Manager of Development and Integration, HE&R
Creating change based on comprehensive, actionable data
By storing its customer information in one place with BigQuery, HE&R could more easily process, aggregate, and analyze large amounts of data using Vertex AI and Gemini. One previously hard-to-track source was public sentiment in forums like the “Connoisseurs” Facebook page. “We’ve always used social listening to keep a pulse on what’s going on in our community,” McPoyle says. “With Google Cloud, we don’t have to dig. We can get a summary that tells us the overall sentiment is positive today, or that people liked the new Fast Track program changes.” The team also uses Gemini to get a pulse on its 10k+ team members. Reviewing the company’s annual engagement survey used to take dozens of hours of work each year to sort data and identify common themes. Now, the team just asks Gemini and gets a summary in seconds.
With a more holistic view of its customer experiences, the company found more targeted areas of its customer journey it wanted to improve. It started by introducing a natural language chatbot to its website for its guests, powered by Conversational Agents. “We used to manually mine for metrics and A/B test how we interacted with guests,” says McPoyle. “Now we can use tools to tell us which type of interaction performs better, and it can dynamically change content to match which segment it thinks a visitor is in.”
In just 40 hours, the company had a fully functional chatbot helping provide more tailored responses to each customer rather than generalized answers. “We wanted to start small,” McPoyle shares. “We knew we could iterate and grow, and the sooner we can help guests, the better.” This iterative approach led the team to implement an invoicing request option into the chatbot’s logic.
“We saw in BigQuery that a lot of people were asking for help locating an invoice, but the chatbot couldn’t do that,” he says. “But we realized we could easily ask for an email address, check if a user with that address stayed with us in the last 60 days, and send them a PDF. The agentic AI handles the request autonomously based on a single customer input.” This shift to simplify the user experience has reduced the number of calls coming into the company’s call center—McPoyle’s next area of focus.
We used to manually mine for metrics and A/B test how we interacted with guests. Now we can use tools to tell us which type of interaction performs better, and it can dynamically change content to match which segment it thinks a visitor is in.
Mike McPoyle
Manager of Development and Integration, HE&R
Using conversational AI to support customers and employees
As HE&R continued its search for more ways to use its expanded datasets, it found additional Google Cloud solutions to close newfound gaps. The beauty of Google Cloud is how well all the tools mesh. We can take our Dialogflow interactions, push them to BigQuery by answering three questions in Dialogflow, and then go to Looker Studio to create a report in 30 minutes. That interconnectedness makes it easy to use, so we can focus on other work.” The company’s contact centers manage thousands of daily calls from park guests, event attendees, and even its team members. Before Google Cloud, the team was still monitoring the types of calls it received and how cases were resolved. However, much like its other customer-related changes, updates were based on small sample sizes. “With these tools we're able to look at everything and just bubble the important stuff up to the top,” McPoyle says. This call center analysis led HE&R to build ‘Agnes’, an agentic AI Interactive Voice Response (IVR) solution with Customer Engagement Suite.
In its data review, McPoyle’s team found that company employees, especially seasonal employees who work at HE&R’s amusement parks, were frequently reaching out to call-off, notify they’d be late to work, or have their schedules read to them. This would take time from another team member answering the call, looking up their account, and making changes. Now, Agnes is fielding more than 25% of call-offs by using a series of webhooks to identify callers by caller ID, referring to the caller by name, and being able to ingest and share updates back to HE&R’s internal systems. “That’s a full eight hours of employee work per day. By saving those hours, our people can spend time innovating and solving more complex guest problems.”
Customer Engagement Suite saves a full eight hours of employee work per day. By saving those hours, our people can spend time innovating and solving more complex guest problems.
Mike McPoyle
Manager of Development and Integration, HE&R
Another common cause for calls the team uncovered was password resets. By using different webhooks to connect Agnes to HE&R’s vast enterprise tech stack, it can identify and anticipate customer needs based on behavior. “Since Agnes is plugged into our infrastructure, it can see things like account lockouts, failed logins, or open support tickets,” says McPoyle. Instead of a user choosing what they want to do from a list, Agnes can recognize the phone number attached to their account and respond with something like, “Mike, I see you’re currently locked out of your account. Please check your email inbox for a password reset. Is there anything else I can help with?” Agnes now manages more than 50% of calls related to password resets, saving customers time and improving their experience with the team.
This shift allows HE&R to focus its attention on continuous improvement. “We eventually want to get to a point where guests can use our solutions to make reservations with us using natural language. Just type or speak a request, and then we'll fill a shopping cart with everything you requested,” McPoyle says. “For a long time, I thought AI was oversold, but now I’m seeing the possibility of what it can do for our customers, and I’ve started telling everyone they have to try it for themselves.”
Hershey Entertainment & Resorts is a world-class entertainment and hospitality company that owns and operates or supports many of the sweet attractions, resorts, and entertainment venues across Hershey, PA. With a core purpose to honor and fulfill the dream of founder, Milton S. Hershey, and a mission to provide exceptional experiences for guests, Hershey Entertainment & Resorts is truly in the business of making life more fun and enjoyable one moment at a time.
Industries: Consumer Packaged Goods, Tourism, Hospitality, and Entertainment
Location: USA
Products: Google Cloud, Customer Engagement Suite, Conversational Agents, BigQuery, Gemini, Google Analytics, Vertex AI