How to engage visitors?
Similar products
Truly relevant similar products recommendations help both new and returning visitors by taking their attention, encouraging them to spend more time discovering products and helping them in making a choice.
Deep content analysis
Kea Labs analyses a meaning of textual information of items, such as description or product name.
Product characteristics
Kea Labs evaluates importance of characteristics in each particular category, distinguishes their values and use them in comparison.
This is incredibly useful for complex products, such as Electronics.
Visual product recommendations
Recommendations go even deeper and analyze product images. Kea Labs compares product shapes and color palette.
This is very efficient for Fashion, Apparel, Shoes or Furniture stores.
How to increase sales?
Related products
The most efficient way to increase revenue is to sale related products to customer. But most of recommendation engines do not provide enough relevance.
Naive algorithms recommend products only based on statistics and user actions (such as views or purchases). In contrast to them Kea Labs uses various techniques to improve the relevance:
Enhanced product matching
Product matching algorithm combines analysis of users actions with the power of items comparison algorithm.
It groups products and detects relations even for non-popular or new items.
Consumable products and components
You may have often seen the cases when recommended products do not really fit each others.
We respect product characteristics and detect equipment and consumable products which fit each other.
Fast learning
With a statistical-based Recommendation Services it’s impossible to recommend related products while algorithm has not yet gathered sufficient amount of data.
We have predefined models for most of the general niches and adjust them for each store during the integration.
Later these models get automatically trained by the actions of every user.
How to sell more items?
Automated
Product Bundling
“Bundle” sales encourage customers to buy a package or a set of items instead of a single product. Similar to cross-sell or related products, but with a more complicated algorithm, as bundles have to contain matching products or products from the same set or collection.
Market basket analysis
Kea Labs automatically bundles products by discovering of regularities and stable connections between them. Bundles may also contain services, like installation or extra warranty.
With a help of our Dashboard or a simple data feed you can define custom bundles or adjust automatic ones.
Collections
Fashion designers rarely make stand-alone pieces and usually create collections of items.
Kea Labs uses multiple techniques to detect and recommend products from the same collections. Those algorithms also work in other e-commerce niches.
Actionable analytics
Kea Labs near in real-time analyses the whole path of users before and after search, and identifies insights to your management and merchandisers.
Search analytics allows you to evaluate search quality, measure impact of search to conversion, understand customer demand, and helps to guide your business decisions
