What top-performing beauty advertisers do differently on Amazon.ca
By: Gyan Harshvardhan, Senior Analytics and Media Manager, and Ashton Brown, Technical Writer
Repeat purchase rate and brand loyalty contribute to brand success and brand growth in Amazon’s store. In this 2021 study we explore three tactics top-performing Beauty advertisers use that differentiate them from lower performers and help increase repeat purchase rates.
Story highlights:
In this article, we highlight three tactics top-performing Beauty advertisers on Amazon.ca (Canada) used more often than lower performers. This comparison of advertising actions taken by top and lower performers can help advertisers inform new strategies to increase brand loyalty, as measured by repeat purchase rate in this analysis.
To perform our analysis, we select over 700 Beauty advertisers on Amazon.ca with similar attributes (brand size, ad spend, industry, etc.) and then use machine learning to group advertisers into four clusters. Clusters are determined by the number of repeat purchasers, with top performers (cluster one) having the highest number of repeat purchasers and lower performers (cluster four) having the lowest number.
We find that top performers achieve a 1.6x higher repeat purchase rate than lower performers, and are differentiated from lower performers by three tactics, which we explore in greater detail here.
Top performers are 3.7x more likely to use remarketing audiences
The first tactic used more often by top performers is audience remarketing, which allows advertisers to drive consideration and conversion with groups of previous shoppers. On average, 33% of top-performing Beauty advertisers used remarketing, but only 9% of advertisers with the lowest repeat purchase rate use remarketing.
Percentage of advertisers that used audience re-marketing
Top performers
Lower performers
When remarketing products, advertisers can consider:
- Marketing during high traffic events and campaigns: Tentpole events such as Prime Day and Black Friday can help potential customers discover products. Advertisers can also remarket through brand campaigns or product deals.
- Reengaging audiences who viewed other products: Cross-selling or upselling to shoppers browsing products is another tactic commonly used to help drive brand engagement.
Top performers have 2.5x more customer reviews
On average, top performers had 15 reviews per product, compared to six reviews for lower performers. Customer reviews are a key factor in purchase decisions. In addition to reengaging with customers, advertisers should consider improving the number of customer reviews on their products.
Customer reviews per product
Top performers
Lower performers
Amazon Vine is a program sellers can use to increase the availability of customer reviews on newly launched products.
Top-performing advertisers are more likely to adopt Amazon DSP
Top performers have a 20% higher Amazon DSP adoption rate than lower performers. Amazon DSP allows for greater audience reach by expanding campaigns to third-party sites and apps, which can help advertisers engage audiences at scale.
Top performers
Higher DSP adoption rate
Conclusion
As seen in our analysis, in combination with our supervised machine learning model, we identified three differentiators between top- and lower-performing Beauty advertisers on Amazon.ca. First, top performers are more likely to use remarketing audiences. Second, top performers have 2.5x more reviews than lower performers; and third, top performers are more likely to adopt Amazon DSP.
These three tactics serve as differentiators between lower-performing beauty advertisers and top performers, who achieve a 1.6x higher repeat purchase rate on Amazon.ca.
Methodology
We first use a supervised machine learning model to identify a list of attributes that help improve the composite score among advertising attributes. Specifically, we follow a five-step process to create a suite of success metrics, measured by repeat purchase rate, and then identify the top advertising and retail strategies to help increase the success metrics with machine learning algorithms.
- Select advertisers: In this study, we analyze 732 advertisers in the Beauty category on Amazon.ca between July 2020 and June 2021.
- Create success metric: We create a success score, measured by repeat purchase rate, and then identify the top advertising strategies to help increase the success score with machine learning algorithms.
- Identify effective ad or retail actions: We then use Pearson Correlation and XGBoost to assign features weights. We then use this list of attributes and perform cluster analysis among advertisers, so advertisers in the same cluster are similar in ad and retail attributes, while advertisers in different clusters are different in ad and retail attributes. Advertising attributes include: always-on measures such as Sponsored Products/Sponsored advertisers/Sponsored Display activeness, Amazon DSP, and more.
- Group advertisers: Campaigns are grouped by their success metrics (repeat purchase rate) and assigned into four clusters. Leveraging advertising actions, we then analyze their impact on the success metrics.
- Compare brand groups: Last, we identify which strategies top-performing advertisers (Cluster One) use to improve repeat purchase rates, compared to strategies that lower-performing advertisers (Cluster Four) are or are not using.