4 tools used by advertisers to help cost-efficiently accelerate global sales
February 08, 2023
The study highlights four free tools and tactics that the top 50% of multi-country advertisers use more extensively than the bottom 50% of multi-country advertisers. Global advertisers may also want to consider using and investing in these strategies when launching or scaling campaigns across countries.
Executive summary
In a recent study of more than 150 advertisers that expanded to Europe and Canada from July 2021 to February 2022, Amazon Ads found that there are four tools available in the advertising console that top performers adopted to drive performance when launching advertising campaigns in new countries within Europe and Canada. They include suggested keyword localization and custom keyword ranking, multi-country campaign creation, dynamic bidding, and manual targeting.
High-level methodology
Amazon Ads adopted a gradient-boosted decision trees model to identify the most important campaign actions that contribute to the following success metrics: ad-attributed sales and cost per click (CPC). By looking at the performance of the top 50% of advertisers and the bottom 50% of advertisers, we identified that the top 50% of advertisers are able to drive 11x higher ad-attributed sales at a 40% lower CPC on average.1 This shows that by using the right tools, advertisers can generate a better return on ad spend (ROAS) by increasing ad-driven revenue and increase efficiency by bidding on cost-effective keywords.
Indexed performance (Baseline = Bottom 50%)
Top 50%
Bottom 50%
Attributed Sales: 11
Attributed Sales: 1.0
CPC: 0.6
CPC: 1.0
We identified four factors behind this performance gap in generating revenue and lowering cost of advertising. The study demonstrates that the top 50% of advertisers most frequently adopted suggested localized keywords. They also created more campaigns using the multi-campaign creation feature and more extensively leveraged dynamic bidding strategies. Additionally, they utilized manual targeting to a larger extent than the bottom 50% of advertisers.
What did top quartile do differently than bottom quartile that helped them achieve higher attributed sales and lower CPC?
| Actions | Top 50% | Bottom 50% |
|---|---|---|
| Number of suggested localized keywords adopted (baseline = median of bottom 50%) | 1.8 | 1 |
| Number of campaigns using custom keyword ranking (baseline = median of bottom 50%) | 1.8 | 1 |
| Percentage of campaigns created using multi-country campaign creation out of total campaigns | 10% | 5% |
| Percentage of campaigns using dynamic bidding | 93% | 89% |
| Percentage of campaigns using manual targeting | 25% | 12% |
Findings
1. Top-performing multi-country advertisers adopt suggested keyword localization and custom keyword ranking
We observed that the top 50% of advertisers adopted suggested keyword localization and custom keyword ranking more often than the bottom 50% of advertisers. On average, the median number of custom suggested keywords used by the top 50% of advertisers was 80% higher than the bottom 50% of advertisers. Among all the features, suggested keyword usage also showed the strongest correlation with our success metrics, indicating that the most efficient feature to drive success was suggested keyword localization and custom keyword ranking.
Suggested keyword localization and custom keyword ranking allows you to create global manual keyword targeting campaigns in just a few clicks. When using manual targeting with Sponsored Products and Sponsored Brands in a country with a primary language other than your language of preference, you will see translations for the suggested keywords in your selected advertising console language. In addition, the recommended keywords are ranked by estimated clicks and order volume. Simply choose the recommended keywords that best suit your campaigns and advertising objectives, set your bids, and launch your campaign as usual.
Suggested keyword localization tool
It was observed that suggested keywords have on average a 17% higher click-through rate (CTR) than self-selected keywords.2
2. Top-performing multi-country advertisers leverage multi-country campaign creation
The study also highlights that the top 50% of advertisers used multi-country campaign creation more extensively than the bottom 50% of advertisers. On average, 10% of all campaigns created by the top 50% of advertisers used the multi-campaign creation tool versus only 5% of campaigns for the bottom 50% of advertisers.
With multi-country campaign creation and copy, you can create Sponsored Products auto-targeted campaigns in multiple countries at once. It’s also possible to copy existing campaigns to additional countries, without spending time setting up and translating campaigns for each country where you want to advertise. Amazon Ads automatically transfers and localizes the campaign settings, so there’s no need to worry about setting up targeting, converting currencies, or choosing the right products to advertise.
Multi-country campaign creation tool
In 2022, more than 100,000 Sponsored Products campaigns were launched for the first time in new countries using the multi-country campaign creation and copy features worldwide.3
3. Top-performing multi-country advertisers use dynamic bidding
Our study also shows top-performing advertisers used dynamic bidding more often than the bottom advertisers: 93% of campaigns of the top-performing group included dynamic bidding, i.e., 4% more campaigns than the bottom 50% of advertisers.
When you create a Sponsored Products campaign, you can choose from four bidding strategies to help ensure your ads get shown in the auction-based placements: fixed bids, dynamic bids (down only), dynamic bids (up and down), and dynamic bids (rule-based bidding). Dynamic bidding is a system designed to improve your campaign performance automatically and in real time, and it takes the guesswork out of adjusting bids to achieve your marketing strategy. Every time your Sponsored Products ad is eligible to appear, Amazon predicts the likelihood of a click converting to a sale.
Dynamic bidding strategy
Advertisers that changed their Sponsored Products campaign bidding type to dynamic bidding (up and down) during Prime Day saw a 5.2x increase in ad-attributed sales, compared to those who used down-only dynamic bidding or a fixed bid.
4. Top-performing multi-country advertisers utilize both automatic and manual targeting
We also found that the top 50% of advertisers favored manual targeting compared to the bottom 50%. Indeed, top-performing advertisers utilized manual targeting for 25% of their campaigns, i.e., 13% more than the low-performing advertisers.
The right keyword targeting strategy can help get your ads in front of relevant shoppers. Automatic targeting is the easiest and quickest way to get started. Amazon Ads does the work for you, matching your ads to shopping queries and products. For advertisers with more targeting experience, manual targeting gives you greater control over when and where your ads appear. With manual targeting, you can choose the specific keywords and products you wish to target and set bids accordingly. Using learnings from automatic targeting campaigns, you can take the next step and set up Sponsored Products manual targeting campaigns. In that case, you’ll manually select the keywords or products that you want to target—and using the best performing ones will enable you to be more refined in your targeting strategy.
Manual targeting
Detailed assumptions and data sources
We first defined two campaign metrics as success metrics: ad-attributed sales and CPC. Once the success metrics were defined, we identified the top campaign actions performed by 151 multi-country advertisers in Europe and Canada that contributed to those success metrics, thanks to a gradient-boosted decision trees model. This method helps us understand which advertising actions are the most important to drive strong success metrics.
We also used a quantile methodology. We ranked all campaigns by their success metrics and assigned them into four equal-size groups called quartiles. Leveraging more than 10 advertising actions, we then analyzed their individual impact on the success metrics. Advertising actions included suggested keyword localization and custom keyword ranking, multi-country campaign creation, dynamic bidding, manual targeting, and so on.
We then compared the top 50% of advertisers and the bottom 50% of advertisers. We examined the differences in the campaign strategies used by the top quartile as compared to the advertisers in the bottom quartile. This helped us understand what the most successful advertisers were doing differently. After identifying the most important actions that drove higher success metrics for the advertisers in the top quartiles, we came up with recommendations to help advertisers improve their success metrics, i.e., increase their ad-attributed sales and lower their CPC.
1 Amazon internal data, Europe and North America, July 1, 2021, to February 28, 2022. This analysis highlights the greatest differences between advertisers with the highest and lowest success metrics, and does not predict performance or claim causality.
Pearson correlation, XGBoost, and subject matter expert suggestions are used to assign features weights.
2 Amazon internal data, worldwide, August 2022. Results are based on past observations and are not indicative of future performance.
3 Amazon internal data, Europe and North America, August 2022