- Referencje
- Norwegian Cruise Line Holdings
Leading cruise line personalizes passengers’ onboard experience in real time
SAS helps create targeted messaging campaigns for customer engagement
280%
increase in campaign engagement
Norwegian Cruise Line Holdings achieved this using SAS® Event Stream Processing on SAS® Viya and SAS® Machine Learning deployed with SAS managed services
Norwegian Cruise Line Holdings uses SAS to transform onboard experiences for passengers with prompt personalized communications
Norwegian Cruise Line Holdings (NCLH) is a global leader in cruising experiences, having been awarded No. 1 status in the Mega Ship Category by the Condé Nast Readers’ Choice Awards and named Europe’s and North America’s Leading Cruise Line by World Travel Awards.
Headquartered in Miami, NCLH began sailing in 1966. Today it is the fourth-largest cruise line in the world and sails to over 450 destinations.
The company operates three leading brands: Norwegian Cruise Line, Oceania Cruises and Regent Seven Seas Cruises. It is known for innovative ship designs, diverse itineraries and a focus on delivering exceptional guest experiences across its fleet of 32 ships. Over the next 10 years, it plans to add 13 additional state-of-the-art ships to its fleet. Furthermore, the company owns and operates two private islands – the Great Stirrup Cay in the Bahamas and Harvest Caye in Belize.
During the coronavirus pandemic, NCLH needed to quickly implement a way to perform contact tracing and keep passengers safe before or as events occurred. To accomplish this, they turned to SAS Event Streaming Processing (ESP) on SAS Viya. However, once the emergency passed, NCLH wanted to find a way to use its software to continue improving the guest experience with personalized onboard communication in real time.
On our shore excursion message, we saw an average uplift of 26% between the target group versus the control group. We did the same experiment on our Wi-Fi upgrade, and we saw an average uplift of 75%. Then on CruiseNext, which is one of our best-performing campaigns, we made an average uplift of 280%. Ethan Rasti Manager of Data & Analytics Norwegian Cruise Line Holdings
Creating a cohesive data ecosystem
Transforming onboard communication took careful planning. First, NCLH began identifying what guests were already looking for during their time on board.
“We have a lot of activities and offerings,” explains Ethan Rasti, Manager of Data and Analytics at Norwegian Cruise Line Holdings. “Each guest’s experience might be different, so how could we identify what entices each guest and deliver to them specific activities that they are interested in?”
NCLH decided to use machine learning and prioritization modeling to identify and meet its guests’ needs. However, first it needed to solve a data integration problem.
“One of the most challenging tasks we have in this industry is creating a centralized data ecosystem,” says Rasti. “When it comes to a shipboard project, we always have challenges on how to maintain the same code and the same logic across the fleet – and also how to monitor everything across ships.”
To solve this issue, NCLH created for every ship a mini data mart, referred to as the onboard data platform (ODP). “We feed the data points from the onboard data platform to SAS Event Stream Processing, and we also leverage the same data for our dashboard creation and monitoring,” Rasti says. “Then through satellite, the data comes back to our shoreside SAS-managed environment and gets combined with all the reservation data and the shoreside data. Then that's where we do our machine learning and prioritization models.”
Quickly implementing this centralized data solution was key for NCLH, so it chose SAS Managed Services to accelerate implementation. Within this connected environment, the company can integrate a variety of customer interactions for more reliable data modeling. Plus, the line receives alerts notifying it of any issues with the onboard data platforms for efficient monitoring across ships.
“It's a very unique environment for us to make sure all of this data is in sync because each of these environments has to be able to hypothetically perform in an offline environment,” continues Rasti. “We cannot always rely on satellite connectivity.”
Rasti continues, “If anything is wrong, we use the same technology to make the code change. Then with literally one click, we can deploy it to all of our ships at the same time.”
Norwegian Cruise Line Holdings – Facts & Figures
450
destinations around the world
32
ships in the fleet
41,000
employees worldwide
Using machine learning so guests find what they need when they need it
After overcoming this logistical challenge, NCLH began testing new ways to engage with guests on board. The line chose three types of test messages: a customized welcome aboard message, a reminder about specialty dining credit and a casino message.
“Each message had different versions based on the guest’s loyalty level and who they were traveling with,” says Rasti.
“In our initial pilot, we learned that out of more than 3,000 guests on this ship, roughly 1,700 of them had the mobile app and were actively using it,” says Rasti. “Roughly 77% of the people or the guests who received the message read it, which is important for us. Also, 41% of the guests who read the message performed our desired action.”
After the pilot program’s success, NCLH began expanding this technology throughout the fleet. “We currently have more than 20 messages or campaigns active with a long list of new messages and ideas in our backlog that we continuously groom and go through with the team,” says Rasti. “The messages could be either operational, safety or marketing.”
With the expansion of the pilot, NCLH relies on machine learning algorithms to track guests’ needs and determine which messages each guest should receive. “We don't want to spam our guests messages,” says Rasti. “So how do we understand our guests better and deliver the right message without spamming them during their vacation? This is one of the most important things for us, and we want to make sure we don't cross that line.”
Monitoring the guest experience
To further refine the results, NCLH also set up several A/B tests. Group A was targeted via the machine learning algorithm, and group B was the control group that did not receive the message. Quickly, it became evident that the targeting was effective.
“On our shore excursion message, we saw an average uplift of 26% between the target group versus the control group,” explains Rasti. “We did the same experiment on our Wi-Fi upgrade, and we saw an average uplift of 75%. Then on CruiseNext, which is one of our best-performing campaigns, we made an average uplift of 280%.”
After seeing the impressive results of its personalized messaging testing, NCLH sent a survey to customers to gauge guests’ reactions to the messages. Working alongside the guest satisfaction team, the company created a survey asking guests to rate how satisfied they were with the types of messages received during their cruise. They also asked about things like the time of day messages were delivered, how easy messages were to understand and the overall number of messages received.
“More than 60 to 70% of our guests indicated they were happy with messages they received during their cruise, which shows that we got it right,” says Rasti.
Sailing ahead
Moving forward, NCLH is continuing to find innovative ways to improve the customer experience, including a new mobile app. Next up? Taking their personalization to the next level. “One of the SAS solutions that I got to see at SAS Innovate was SAS Customer Intelligence 360, which I believe will enable us to expand this effort beyond the ship environment,” Rasti says.