Google launches ‘Product Discovery Solutions’ including search and recommendations for online retailers
Google joins the club of companies offering recommendations-as-a-service (RaaS). In a press release, Google announced the launch of “Product Discovery Solutions for Retail, Bolstering Personalized Online Shopping”. The ‘Product Discovery Solutions‘ consists of three modules, namely ‘Vision Product Search‘, Search for Retail (private Preview), and, most importantly for the readers of our Blog, ‘Recommendations AI‘, i.e. recommendations-as-a-service.
Recommendations-as-a-service is not to be confused with a Recommendation API as e.g. offered by Amazon. For using an API, a company still needs notable knowledge in recommender systems and Web Technologies / APIs / Web Programming. Recommendations-as-a-service makes integrating recommendations into an e-commerce shop as easy as integrating Google Analytics.
Google’s demo video showcases a neat web interface that allows choosing from the most important options. This includes the business objective (CTR or CVR) as well as the type of recommendation (‘Others you may like’, ‘Frequently bought together’, and ‘Recommended for you’). How exactly the integration is done remains, however, vague. If anyone knows whether still an API is needed, or if simply a JavaScript snippet needs to be integrated, let us know in the comments.
Prices vary between 10 US Cent and 27 US Cent per 1,000 predictions (see table below).
| Prediction requests per month | Price per 1000 predictions |
|---|---|
| Up to 20,000,000 | $0.27 |
| Next 280,000,000 | $0.18 |
| After 300,000,000 | $0.10 |
Disclaimer: I am involved in the business start-up Darwin & Goliath, which also offers recommendations as-a-service.
About The Author
Joeran Beel
I am the founder of Recommender-Systems.com and head of the Intelligent Systems Group (ISG) at the University of Siegen, Germany https://isg.beel.org. We conduct research in recommender-systems (RecSys), personalization and information retrieval (IR) as well as on automated machine learning (AutoML), meta-learning and algorithm selection. Domains we are particularly interested in include smart places, eHealth, manufacturing (industry 4.0), mobility, visual computing, and digital libraries. We founded or maintain, among others, LensKit-Auto, Darwin & Goliath, Mr. DLib, and Docear, each with thousand of users; we contributed to TensorFlow, JabRef and others; and we developed the first prototypes of automated recommender systems (AutoSurprise and Auto-CaseRec) and Federated Meta Learning (FMLearn Server and Client).