recommendations based on visual similarity


MASTERING THE PARADOX OF CHOICE WITH COMPUTER VISION

Show what your clients like
You want to convert your audience into customers and make the most out of your resources? First of all, show your audience what they want to see. If someone is searching for a dress and clicked on a navy blue midi-length dress, the best you can do is to show him more relevant items from your catalog.

esprit_01
esprit_02

Similarity means relevance
Does it matter to show „What others bought“? Rarely. The most relevant suggestions for your shopper will be driven by the visual similarity to the desired item. That’s why we are focusing on the product images, refined with data, search and click behavior of the actual session. In real-time we are recommending relevant visually similar products from your catalog personalized for every visitor.

Individual control and easy integration
Of course you can leverage your know-how and expertise within our algorithms. Feel free to determine which factors are most relevant for visual similarity – shape, colors, patterns etc., simply by using a slider in our client frontend. If you want to fine-tune our results, submit your business values like price ranges, brand lists etc. Have a brief look at how it works!

4

As more and more traffic is coming from mobile devices, this gets increasingly important. The smaller the screens, the more crucial it is to display the most relevant items.

Showing the most appropriate products to each individual shopper results in:
» increasing conversion rates,
» boosting average order size and
» lowering bounce rates and exits.

4

There are several use cases to benefit from our recommendation engine based on visual similarity:
» offer the most similar and relevant items as a recommendation on every detail page
» show next best offers for sold out or out of stock products
» employ the most personalized category listing pages sorted by the visual similarity to the individually selected item


CLIENT’S STATEMENT

We have successfully tested picalike's personalization and recommendation engine based on their unique visual technology in our German online store last year. From now on we will expand to all our 15 country stores enjoying the benefits of their solution.

Mark Oktay Kretschmann - Product Owner Personalization Global, ESPRIT Europe GmbH


YOU WANT TO SEE IT LIVE, WITH YOUR OWN CATALOG?

That’s for free, please fill in:

captcha

Contact information

+49 40 468 95 17 0

Stresemannstr 23, D-22769 Hamburg