Automated Bidding Helps Rakuten Ichiba Increase Sales and Conversations

Internet shopping mall Rakuten Ichiba incorporates scientific approaches to analyze and continually improve its marketing. By testing automated bidding against manual bidding and ultimately adopting automated bidding, it saw increased volume and value of sales with lower campaign costs.


  • Test automated bidding
  • Grow number of conversions and value of sales


  • Used the target ROAS feature to test automated and manual bidding
  • Responded to test results by converting to full automation and allocating a larger budget


  • In the test, automated bidding produced 12% more conversions and a 14% higher value of sales
  • After converting fully to automated bidding, conversions increased 160% and value of sales increased 190%

Rakuten Ichiba, an online shopping mall, is one of the world's most unique marketplaces, incorporating a B2B2C model. With lots of sellers and buyers to keep happy, it brings a rigorous scientific approach to its marketing campaigns, always in pursuit of optimal efficiency. It had already incorporated automated targeting and dynamic remarketing, but was conducting bidding manually. Rakuten Ichiba relied on its expertise to separate ad groups by product category, recency, and ad size. In September of 2014, it decided to see what automated bidding could do.

With automated bidding, the AdWords algorithm analyzes signals related to the circumstances of each ad auction and sets bid amounts according to the likelihood of a click leading to a conversion. The analysis is not confined to individual signals, but combines many complex signals to set the bid amount. In addition to increasing the number of conversions and value of sales compared to manual bidding, automated bidding can significantly reduce time spent managing a campaign.

Rakuten Ichiba used the target return on ad spend (ROAS) bidding strategy tool to conduct an A/B test between automated bidding and manual bidding. This feature lets users set an ROAS target and automatically adjust bid amounts for maximum effectiveness.

Automated bidding combines many complex signals to set an optimal bid amount.

Rakuten Ichiba’s rewarding results

Rakuten Ichiba split users on its remarketing list into groups of equal size and rolled out campaigns with equal budgets to conduct the A/B test. Automated bidding was the clear winner. Compared to the manual bidding campaign, automated bidding reduced campaign costs by 12%, increased conversions by 12%, and increased the value of sales by 14%.

Additionally, Rakuten Ichiba went all in with automated bidding. In April 2015, conversions were up 160% and sales value was up 190%. With the time saved from not having to manually manage all the campaigns, other tasks, such as user-related strategies and new policies, could take priority.

Test Results: Manual vs. Automated Bidding
September 2014
Change After Automated Bidding's Introduction
April 2015

Continual improvements

"Since introducing automated bidding, we have been conducting our bids with a continuously high level of performance. When we first introduced automation, we got results that were higher than our target ROAS, and we had several issues such as unreliable performance. However, with repeated optimizations by the Google team to improve the signal reading, we have successfully solved those problems and things are running smoothly now,” says Shiho Kadota, of Rakuten Ichiba’s Affiliate Merchandiser Group Display Team. She concludes, "Automated bidding has gone so well, we’re considering conducting tests on search ads and introducing automated bidding there as well.”

Think w/Google APAC on LinkedIn

Join a community of digital thought leaders and be inspired.