Patented Portfolio of Algorithms

No matter how clever it is, there is no single algorithm that will give the best recommendations for every retailer – online retailers and shoppers vary way too much.

"MyBuys provides the customer profiling and merchandising know-how to deliver the personalization that we need in order to stay relevant in a competitive market."

Erik Martinez, Director of eCommerce at Peruvian Connection

It Really IS Rocket Science

That’s why we don’t rely on just one. MyBuys has developed  a patented portfolio of algorithms that was developed by e-Commerce personalization experts from Blue Martini, BroadVision, MarketLive and Microsoft, as well as the head of data mining at NASA. We then apply a unique “recipe” that combines different weightings of the algorithms in each situation where we deliver recommendations. Depending on what the shopper is doing, what his or her price affinity is, what products are in your catalog, what merchant events are going on, and how you choose to merchandise, the mix of algorithms changes.

Getting the Right Mix

For example, one online retailer’s shoppers may be influenced more by price than brand or color.  On another site, the shoppers might be so brand focused that making recommendations outside of their favorite brands would make no sense whatsoever. A third retailer’s site may benefit primarily from a mix of algorithms based on inventory information. But even then, there are always times when we need to make changes to the recipe, either temporarily or permanently. For example, when sales are going on, promoting the on-sale items might be most important.

Our sophisticated multi-algorithmic approach to recommendations accommodates all of these scenarios, and more. We do this for both the recommendations made on the website as well in emails.  Sometimes each requires a different approach; we can ensure the right mix is used in any situation.

What Next?

Learn about recommendations. Learn how we are always optimizing. See the results we deliver.