2 in 3 consumers now expect personalization as a standard of service, across all touchpoints, a recent study by Redpoint Global and The Harris Poll conducted in the U.S., U.K., and Canada, has revealed. Artificial intelligence and machine learning are recognized as the top game-changing technologies in retail.
Although companies are at different stages of this journey, most agree that finding or developing analytic talent is a key concern and bottleneck for doing more.
With a fully managed and fast implementation service, Recommendations AI poses a disruptive alternative to the common pathways of implementing Artificial Intelligence in retail businesses. Many retailers are using prebuilt Machine Learning APIs, while some choose to train a Machine Learning model to meet their business needs with the help of a specialized team of Data Scientists. But it could take months, if not years, to recruit and hire data scientists for a fully-capable team and to build a custom ML/AI model to achieve what you need.
In contrast, Recommendations AI doesn’t require Machine Learning experience, data preprocessing, or manual infrastructure provisioning, all of which are handled automatically by the system. You can customize it based on your own recommendations scenarios and business objectives and has a simple pricing plan – pay only per predictions you consume.
During the Google Cloud Next ‘19 event in San Francisco, eMAG shared its experience with Recommendations AI, as one of the first retailers worldwide to have implemented the solution at that time. eMAG managed to have a go-to-market model ready in 2 weeks, following an easy integration of the Recommendations AI solution and fast training of models.
When compared to its recommendation systems, eMAG experienced lift as much as:
We can help you integrate Recommendations AI into your business by actively implementing the solution or by guiding your team along the way.
It could take up to one month to have everything ready, so you can start training the model quickly and get results right away. Once the implementation is done, you enjoy a simple pricing model, as you pay only for the predictions you consume. Recommendations AI pricing is based on training and tuning (per node per hour) and requesting predictions (per 1000 predictions), starting from 0.10$ to 0.27$, tiered by the number of monthly prediction requests.
One of the clients we’ve recently supported the implementation of Recommendations AI for is Vivre. Let’s see what our client thinks of their experience with Recommendations AI so far, from Mihai Iova, Head of Product at Vivre:
“Vivre is a company born from a passion for technology with innovation at the forefront of its growth. We found in Zitec an experienced partner with the same enthusiasm for technology. With Recommendations AI we have the freedom to experiment with recommendation scenarios and to offer our customers a source of inspiration of more than 150,000 unique products available daily on vivre.ro. We recorded a 15% increase in transactions obtained from product pages that contain recommendations. We will expand both the scope and the number of our experiments with Recommendations AI.”
Recommendations AI has increased its efficiency with 50% from project-start and is expected to improve, as Vivre continues to build and run new experiments.
If you also want to accelerate your business with the power of AI and deliver product recommendations that speak your customers’ minds, we’re here to help you set everything up. Using cloud solutions in our software since 2007, we have extensive experience in delivering sustainable solutions for various business needs as cloud natives.
Starting April 2020, we’ve been officially listed as one of the first Recommendations AI integration partners worldwide by Google Cloud, a shortlist of trusted partners who can help clients implement the innovative solution into their eCommerce business.
Reach out to us and request a demo – we’re only a few clicks away!