McDonald's
Making McDonald’s coupons rain with ad optimizations on TikTok
The Objective
How about a discount coupon?
It is nothing new to anyone that McDonald's is one of the favorite brands of Brazilians. But more than hamburgers and fries, they also have an app – where they offer discount coupons to buy snacks and combos – and a strong strategy centered on it.
On TikTok, publicizing the app helps the brand in searching for people interested in downloading it and also the retention of these customers with new coupons. And in this strategy, their goal was to make these ads as efficient as possible.
The Solution
A mix of optimizations and machine learning
At the beginning of their campaigns on TikTok, McDonald's strategy was just of awareness and consideration, with low investment in performance. But with the good results in the early stages of the funnel, the brand decided to increase this dedicated budget at the end of the journey, focusing on optimizing it for app installations and coupon redemptions.
The first few months of delivery have already brought higher-than-expected numbers to the brand. This is a good sign. But a question remained: if investment escalated and the strategy remained the same, would these good results continue? They would had to get our hands dirty to test it.
That's exactly what the brand did, in an action led by the agency Galeria. Galeria structured a whole schedule of tests and learning, to analyze both the creative factor and the use of the optimization and machine learning resources of our platform. Do we need to say that it was (and has been) a success?
The Results
The lowest redemption cost!
In the first quarter of testing and optimizations (the first quarter of 2022), the cost per redemption dropped 89% compared to the average of the last quarter of the previous year – even with a 200% higher investment.
And it didn't stop there: this reduction was consistent. In the following months, they continued to see a great improvement in results. In June 2022, six months after the start of the tests, the redemption cost was 179% lower than that attained in January. In the end, the average cost of redemption for the first half of this year was 600% lower than the average for the last quarter.
The secret to success lies in creative optimizations and machine learning, of course. After the first months of testing and learning, for example, the brand followed the recommendation of the TikTok team and activated the Automatic Placement feature when setting up ads. Thus, our algorithm started to automatically take them to the placements where they brought more results, choosing between TikTok and Pangle.
Maintaining and running great campaigns aimed at branding, awareness and consideration on TikTok also contributed a lot to the result. A full funnel strategy helped building a solid foundation for the brand's message at every stage of the funnel and, as a result, made it perform even better.
"TikTok showed that it was possible for us to advertise in a much lighter, fun way and still generate great results. Our main objective in the App Engagement strategy is to generate coupons at the most profitable possible value. We had a profitability in 2022 vs 2021 of almost 70% and after the beginning of Placement Optimization, we have already made another 50% in the value of generating coupons per user."
-- Raphael Lima, Performance Media, McDonald's Brazil
"TikTok is an essential partner in our performance media strategies, because in addition to ensuring positive results, it helps us maintain the coolness of the McDonald's brand in an innovative platform with an audience that is demanding with the content consumed. In this strategy, creating a daily monitoring and optimization routine with the TikTok team was essential to reduce our learning curve and leverage our results. During these interactions between TikTok, Galeria and McDonald's, we were able to define our goals and KPIs and bring creative and placement optimization solutions that helped monetize our costs of installing and redeeming coupons in the app.
