Aeroplan: Leveraging Machine Learning to Promote a Loyalty Program


Campaign Summary

Aeroplan promoted its loyalty program with ads that featured dynamically designed creative powered by machine learning.



Travel loyalty programs across the board had diminished Canadians' enthusiasm for the category, and, as a result, active program participation had been on the decline. The rapid shift in consumer behaviors in 2020 only further amplified the importance of the travel journey and discouraged an exclusive focus on the destination.

Whether it was frequent business travelers, families, or those who only travel a few times per year, it was apparent that all types of travelers were feeling trapped by unfulfilled travel expectations. With its transformed loyalty program, Aeroplan was looking to change that by delivering a best-in-class credit card with TD that worked and delivered value with new core card features and redemption offers. The TD Aeroplan credit card's key message positioned it as the travel loyalty credit card that travelers could always count on to travel more and travel better. The new program and co-brand card were scheduled to launch on November 8, 2020.

The strategic objective of the 2020 TD Aeroplan co-brand campaign was to drive awareness and ultimately drive interest among the card's core priority target. Aeroplan set out to showcase across digital channels the many ways this new credit card met the unique needs of the various target audience groups with the most relevant credit card benefits and offers. Getting the right creative to the right consumer target with the relevant message was a key objective of the campaign.

Target Audience:

The loyalty program has an expansive target audience that uses and gets value out of Aeroplan for distinct reasons, including adults aged 25 to 50, business travelers, and reward seekers. The campaign was targeted specifically to individuals in Canada, which meant that it needed to run in English and French depending on the specific geography of the consumer.

Creative Strategy:

To effectively tell the transformed co-brand story and capture mindshare for the TD Aeroplan credit card, Aeroplan partnered with AdTheorent to help execute its media strategy and deliver a purposeful, targeted message that highlighted the potential of travel for each audience. The timing of the campaign was strategically chosen to support the initial launch of the card on November 8, with Aeroplan capitalizing on digital channels in the weeks following.

To deliver a specialized media strategy, Aeroplan used machine learning (ML) models to dynamically score and design digital ads in microseconds for each consumer, delivering truly personalized creative. These dynamic ads could consist of an unlimited number of customizable creative elements, such as featured products, logos, colors, backgrounds, messaging, and CTAs.

Simultaneous with the ad design, predictive targeting models determined the right consumer and moment to serve the ad. The ML models scored and constructed the optimal combination of creative elements that were most likely to lead to a successful campaign action for each consumer. Once deployed, the models self-learned from campaign data to improve performance over the course of the campaign with data science oversight.


This was the first year of the campaign. Due to the state of the travel industry and the importance of delivering a campaign experience that demonstrated real value to such a broad audience, it was important that Aeroplan was utilizing industry-leading advertising technology for an industry-leading loyalty program.


Overall Campaign Execution:

The Aeroplan co-brand campaign utilized a total of 48 unique creative combinations in English and 48 in French, four images, three CTAs and four different sets of copy across mobile, tablet, and desktop. The campaign was able to deliver a very high level of customization to its target audience across all digital channels. With the campaign's powerful combination of targeted creative and advanced advertising technology, it successfully achieved the campaign's overall objective. Approximately 69 percent of the campaign budget was dedicated to mobile devices to ensure the target audience was reached wherever they were.

Mobile Execution:

The mobile component played a key role in the overall success of the campaign. When it came to CTR, both mobile phones and tablets outperformed desktop, with tablets accounting for 30 percent of overall delivery.

The amount of data available on mobile allowed Aeroplan to build custom audience models to reach its desired demographic. Aeroplan ran predictive targeting to serve the right combination of creative elements in real time to consumers who had high predictive scores to take desired actions in the target demographic.

This approach allowed the company to reach the right audience within targeting parameters at the right time, with the precise combination of creative elements at scale. This yielded far superior results to standard targeting or rule-based dynamic creative alone.

By leveraging machine learning, Aeroplan was able to find efficiencies as the campaign progressed, which led to a steady increase in engagement.

Business Impact (including context, evaluation, and market impact)

The campaign was successful in driving awareness and interest around the new loyalty program, further solidifying the important role Aeroplan knew mobile would play in the overall media and creative strategy.

Aeroplan's approach delivered a 45.45 percent lift in engagement compared to the randomized control group and exceeded overall expectations for the campaign in many ways.

  • Engagement with new program features outperformed projections within the dynamic ad units.
  • A significant increase in likelihood of signing up for a card in the next three months was noted.
  • The likelihood of applying for a TD Aeroplan credit card increased 21 percent for Aeroplan members and 10 percent for non-members.

As a result of the campaign's success, Aeroplan was able to make a positive impact on the perception of loyalty programs among the core priority target and drive positive brand engagement. The campaign contributed to success that went far beyond media metrics and re-instilled confidence in the Aeroplan loyalty program.

Aeroplan also believed that the campaign:

  • Strengthened the emotional connection with the Aeroplan brand, promoting an image of it as being modern, a leader, attractive and knowledgeable — all of which were drivers of brand engagement
  • Improved perceptions of key features of the loyalty program, including the value of points, ability to quickly accumulate points, ease of redemption, and variety of redemption options relating to travel needs
  • Contributed to closing the gap with competitors (e.g., RBC Avion and WestJet Rewards) on perceptions of "no blackouts" and "goes above and beyond"

Categories: | Industries: | Objectives: Machine Learning and AI | Awards: NA Finalist