Skip to content

Apptus Personalization for Aimless and Directed Shoppers

July 9, 2013

Apptus's More Like This  function gives the aimless many natural paths to follow. In this example, the shopper is shown items from the same brand, category or lifestyle, selected by Apptus based on the his interests. Every item has a "more like this" button which will create a whole new page of things to look at. And so on.

Apptus’s More Like This function gives the aimless many natural paths to follow. In this example, the shopper is shown items from the same brand, category or lifestyle, selected by Apptus based on the his interests. Every item has a “more like this” button which will create a whole new page of things to look at. And so on.


Apptus is a software company intent on eliminating irrelevant information in communications with customers, via the tactics and technologies of personalization. It offers self-learning solutions for digital marketing and retailing. Aptus’s eSales solution offers targeted search and navigation, recommendations, merchandising and promotional content. Apptus was founded in Sweden in 2000, and now has clients in Europe, North America and Australia. Customers include CDONgroup, Bokus, Hittahem.se, Lensway, Nelly.com, and Würth.

Apptus is of interest because it is unusual in the recommendations market for providing its solutions both as SaaS and on-site licensed software. Key strengths are its non-directed shopper experience, automated optimization; and its math research collaboration with University of Malmö.

The non-directed shopper does not have a specific purchase in mind. She is not thinking, “red cashmere cardigan sweater with lapels and gold buttons.” She is wondering, “what sort of fashion does this site offer, and is there anything that would freshen up my look for the coming season?” While the directed shopper is fairly well-served by, and fairly accustomed to, travelling up and down a hierarchy and back and forth from lists to details in hot pursuit of the right sweater, the non-directed shopper lacks the motivation for all this running around. Apptus’s “More Like This” navigation gives the aimless shopper multiple paths that are natural and interesting. Apptus identifies products that are similar based on product attributes, and then selects which items to show first based on each visitor’s interests. The More Like This trail is created on the fly, based on what the customer indicates is interesting. Any click has the potential to surface any product in the catalog – and every retailer wants to display more of the items in the catalog.

More Like This is not a replacement for directed search. It supports a natural experience for the non-directed shopper, just as search and category navigation support what has become a natural experience for the directed shopper. Apptus supports both types of experience, personalized for each visitor.

eSales tailors customer experience with search and navigation based on customer interests, determined from behavior (such as what content does the customer spend time with) and customer profile (such as what region does the customer shop from). E.G., A customer who always buys French wines will see France first in the list of navigation facets. It offers four types of recommendations: tailored to visitor, alternative products, complementary products, and generic (such as top sellers). Search, navigation, merchandising and recommendations all use the same core services for optimization and prediction.

Apptus has an interesting slant on assessing customer interests, in pursuit of high performing content. It not only observes the 10 items a visitor clicked on, it observes the 100 items the visitor ignored. If an item is repeatedly ignored, Apptus will allocate that real estate to something – anything!—more likely to be interesting. This is the core of Apptus’s automated optimization: automatically and continually testing various alternatives in order to select the most effective content. Marketers may, if they choose to take control of specific situations, set rules and kick off multivariate tests to measure results.

Apptus scientists collaborate with Malmö University’s Department of Computer Science to find better algorithms for solving problems in combinatorial optimization – known to retailers as the problem of very quickly selecting a set of items that maximizes the probability of purchase, and to mathematicians as the Maximum Coverage problem.

Advertisements
Leave a Comment

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: