Yahoo! Japan is known as one of the largest portal sites in Japan. Their service area is broad, from Yahoo! Search to Yahoo! News, Yahoo! Auction, and PayPay. That is why Yahoo! can obtain rich behavioral data from them, which corporate advertisers can utilize for marketing purposes such as “Yahoo! JAPAN Prediction Funnel.” In this post, I will run an overview of the Yahoo! JAPAN Prediction Funnel and how it can be used for marketing.
1. What is Yahoo! JAPAN Prediction Funnel?
ー What is the Yahoo! JAPAN Prediction Funnel (the Prediction Funnel from hereon)?
The Prediction Funnel is a solution that analyzes audience behavior on the Web and gives a predictive score on a scale of 0.0 to 0.9 for the likelihood of actions such as purchases, searches, and store visits. The higher the score, the higher possibility of conversion. The Prediction Funnel allows companies to predict the likelihood of audience actions for their services, using it as a single metric for ad deliveries and marketing performances.
ー Yahoo!’s rich assets for data analysis
This Prediction Funnel is a solution unique to Yahoo!, covering most of the life scenes with approximately one hundred services such as Yahoo! Search, Yahoo! News, LOHACO, GYAO!, Yahoo! Auction, and PayPay. Because many people live under the Yahoo! JAPAN economic sphere, with 80 million Log in IDs per year, Yahoo! can acquire a wide variety of behavioral data. That is why they can analyze the audience’s potential for action in the Prediction Funnel, guaranteeing both quality and amount of data only possible by Yahoo!.
2. Marketing Use of the Prediction Funnel
ー Visualization of the user portfolio of products and their application to actions
The first is a visualization of the users’ prediction funnel portfolio of products. Simply put, using the Prediction Funnel, you can visualize what percentage of the audience is likely to convert to your product and what percentage is unlikely to. This can be one of the indicators to judge the effectiveness of marketing by looking at the changes in the Prediction Funnel before and after marketing actions such as ad placements.
For example, if the 0.9-score audience volume increases to 120% and the 0.1-score audience volume decrease to 80% before and after the placement, you can say that it impacted the Prediction Funnel. It is important to note, however, that the ad is not the only factor that impacts the Prediction Funnel portfolio; external factors such as the timing of placement may also have an impact. Nevertheless, as direct conversion measurement becomes more difficult due to the ITP, the Prediction Funnel changes should be a helpful indicator for measuring effectiveness.
ー Use with Yahoo! Ads
The second is its use with Yahoo! Ads. By using the Prediction Funnel with Yahoo! Ads, you can place targeted ads based on the prediction score. This allows you to target ad delivery by audiences’ motivation, tailoring the ad for them, including its purpose and creatives based on their characteristics.
On the other hand, Yahoo! does not recommend such ad deliveries to high-scoring audiences for direct response. Since the Prediction Funnel is designed to analyze and visualize the entire funnel, it is recommended to use “Behavior Prediction*1,” “CV Prediction*2,” and “Similarity*3” for delivery aiming to maximize conversions.
*1 An ad delivery solution that uses Yahoo! DMP tags to deliver ads to audiences likely to behave in the same way as the target audience.
*2: An ad delivery solution that delivers ads to audiences with similar characteristics based on CV data.
3: An ad delivery solution that delivers advertisements to audiences with similar characteristics to a specified target list.
3. Case study
From here, I will give two actual cases which used the Prediction Funnel for marketing.
- “nomu.com,” a real estate website by Nomura Real Estate Solutions Co.
By adopting the Prediction Funnel as their new effectiveness metric, they were able to visualize a 114% increase in the volume of the 0.9-score audiences by comparing before and after ad placement. In addition, a comparison of ad contacted and non-contacted users for both programmatic display ads and direct buying display ads on Yahoo! Ads revealed that the score improvement rate for ad contacted users was nearly twice as high as that for non-contacted users, making result assessments on Yahoo! Ads possible.
- Irep Inc.
We at Irep Inc. assessed their ad placements by visualizing the changes in the Prediction Funnel before and after the placement. The comparison showed about 200% to 250% of audiences for both with a score of 0.9 and 0.8, indicating that the audiences with high motivation for conversion increased significantly. This visualization of motivation increase was not possible with just conversion analysis, functioning as a new indicator for ad placement results. We obtained ideas for the subsequent placement, and the Prediction Funnel helped them in our PDCA cycle.
A new data solution by Yahoo!, the “Prediction Funnel,” can be used for a new marketing result indicator and as a reinforcement to Yahoo! Ads. It should become a new reference index as a Cookie alternative due to stronger enforcement of the ITP. We encourage you to consider this solution for your marketing efforts. Feel free to contact us for more information. We are happy to support you with our rich experience with the Prediction Funnel.
Want to know more about LINE Ads?
Contact us to discuss how we can work together.
Irep Inc. is an award-winning global digital marketing agency based in the San Francisco Bay Area. Our headquarters are in Tokyo and our network spans more than 20 countries. In Japan, we are ranked No. 1 for performance-based marketing. We also offer highly specialized market entry, as well as integrated marketing and localization services. Since 1997, our data-driven solutions have effectively led our diverse international clientele to continuous success in Japan, Asia, and beyond.