Taking a Brand from Clicks to Acquisitions with Machine Learning

As the world’s fifth-largest low-cost airline, Norwegian Air carried 37 million passengers to over 150 destinations in 2018. Using limited ad dollars, the carrier partnered with AdTheorent in the fourth quarter to boost its reach and conversions in key markets using smart data and machine learning, the company announced today.

Norwegian’s Head of Marketing North America Marina Suberlyak said, “Norwegian’s goal with many advertising campaigns is similar: to drive ticket bookings via the Norwegian website. And we always have specific geographies that we target based on flight routes and bookings goals.”

For last year’s holiday push, the airline identified key markets for takeoff, including New York, Boston, Chicago, Los Angeles, Austin, Denver, San Francisco and Florida. They drafted New York-based AdTheorent to unleash machine learning tailored to each DMA, focusing on travelers most likely to book flights.

“What’s unique about AdTheorent’s approach is its ability to drive efficiency by identifying and targeting only the users most likely to take the desired action of booking a flight among all users within Norwegian’s targeting parameters,” Suberlyak explained. “AdTheorent developed custom machine learning models around our objectives and used learnings and data throughout the campaign to optimize toward our desired outcome, which delivered stand-out performance.”

The campaign results boasted a cost-per-booking acquisition (CPA) that was 170 percent lower than the goal Norwegian Air set for the effort. This concentration on down-funnel outcomes made AdTheorent an ally in achieving the intended goal.

“Many brands focus on clicks or engagement metrics which don’t always translate into a direct business outcomes,” said James Lawson, CEO of AdTheorent. “Norwegian wanted to move beyond basic measurement tactics to demonstrate real-world value.”

Crucially, AdTheorent’s platform used data on the DMAs to target variables that drove higher engagement, information tied to Norwegian Air and its customers that can be used for future campaigns. According to Suberlyak, the brand discovered that performance was strongest at the beginning of the week, and that Wednesdays were the day of the week that drove the most efficient cost per booking.

Additionally, the campaign uncovered that users of WiFi were four times more likely to book flights from home than at work.

Lawson said, “Once a campaign launches, AdTheorent’s data-driven systems begin learning.  We use the data and actions captured to create and optimize predictive models – algorithms – which are tuned to achieve specified advertiser goals, in this case, ticket bookings. Over time, as a result of these optimizations, campaign performance continues to improve.”

According to Lawson, the platform continues to work with the brand “to drive flight bookings at an efficient and value-adding ‘cost per booking.”

He adds, “We are expanding our relationship with Norwegian this year by supporting additional airline routes.”

Total
0
Shares
Related Posts