Here's why good detail pages matter for new product launches

In this 2021 study of over 320K new products or Amazon standard identification numbers (ASINs), we compare the performance metrics of top-and lower-performing new product launches on Amazon.

On Amazon, new and existing brands are constantly innovating by delivering new products for customers. But what is it that helps make a product launch successful? In this study, we attempt to answer this question by quantifying the success of over 320K product or ASIN launches.

Before performing our analysis, we first needed to define a success metric to which all unique products or ASINs that are supported with Amazon Ads solutions could be compared. Using input from internal Amazon Ads experts, we developed a composite score comprised of three factors: (1) the number of ad impressions delivered; (2) the average ad click-through rate; (3) the average retail unit growth rate.

With the success metrics defined, we could group ASINs into equally distributed quintiles (20% of total sample size for each)—with Quintile 1 having the highest composite score and Quintile 5 having the lowest composite score. Next, we selected 35 media and retail attributes, ranging from Amazon Ads products to customer reviews, in order to train a machine learning algorithm designed to test ad strategy combinations that could improve the composite score.

Here, we compared how “top-performing ASINs” (Quintile 1) and “lower-performing ASINs” (Quintile 5) performed in advertising strategies that we had observed to have the highest impact on increasing the composite score of impressions, click-through rate, and average retail unit growth.

Our analysis found that, when compared to lower-performing ASINs, top-performing ASINs receive 7.1x more ad impressions, a 2.1x higher ad click-through rate, and a 3.8x higher average retail unit growth.

We first used a supervised model to identify a list of attributes that help improve the composite score among 35+ media and retail attributes. Specifically, we followed a five-step process to create a suite of success metrics, including branded searches and detail page views, and then identified the top advertising and retail strategies to help increase the success metrics with machine learning algorithms.

1. Select brands: We looked at more than 320K advertiser brands across all industries and seven countries (United States, United Kingdom, Germany, France, Japan, Italy, Spain) between October 2020 and September 2021.
2. Create a success metric: We calculated new ASIN launch success according to the click-through rate percentage, the retail unit’s average growth (for the period observed), and the total number of impressions delivered.
3. Identify effective ad or retail actions: Using inputs from industry experts, we identified 35 attributes that might help improve the success metrics (a higher number of ad impressions, a better ad click-through rate percentage, and a higher average retail unit growth). We then used the Pearson correlation coefficient to reduce attributes that are highly correlated for further study. The remaining attributes are ASIN quality measure, such as ASIN readiness code; Sponsored Products campaign features, such as spend/budget; Sponsored Products targeting, such as keyword targeting; and retail attributes, such as customer reviews and average rating.
4. Group brands: We then adopted a quintile analysis to rank the ASINs by their success metrics. All of the ASINs were assigned to five equal proportional groups based on the final rank.
5. Compare brand groups: We compared the top- and bottom-performing quintiles and concluded what are the key attributes that differentiate their performance on impressions, click-through rate, and average retail unit growth. This helps us understand what the most successful ASINs do differently. We provided recommendations on the most important advertising and retail actions to help advertisers improve their ASIN activeness and overall performance.