Recommendations Analytics

In the analysis area of epoq Inspire, epoq Connect & epoq Stream in the Control Desk, you will find all the important key figures for the individual recommendation widgets.

You can analyze recommendation widgets by identifying optimization needs and deriving actions to improve performance.


Widget Overview

In the analysis area you will find an overview of all active recommendation widgets.

There you can see at a glance where the widgets are placed in the store and how they perform.

All Inspire widgets that are active in the selected period are displayed here with title and ID. At the top of the page, you can select the time period in which the recommendation widgets are to be analyzed. You get key figures about views, visibility, clicks and purchases of the respective widget in absolute numbers and in percent. You can also view the description of the widget, if one is stored. Via "View configuration" you can directly access the rule editor and adjust the rules for the widget.


Popular Recommendations

In the recommendation analytics in the Control Desk under Services > Inspire > Analytics > popular recommendations you can see the top 100 most popular recommendation products per widget. These products were recommended the most, clicked as a recommendation, and purchased as a recommendation in the selected placement. The table is sorted by purchases by default.

Select a placement you want to analyze from the drop-down menu. The most popular recommendations are displayed in a table:



Based on the most popular recommendations, the products of a widget that perform particularly well are highlighted. With the help of the key figures, you can, for example, recognize products that are recommended frequently, but are not clicked on or purchased very often. For this product, a more detailed analysis (product description, product image, price, etc.) is worthwhile. Through the most popular recommendations you can recognize brands and products of great interest. For these, further measures can be taken in the Control Desk and thus promoted in the online shop. For example, you can create a banner entry for popular brands or create a Theme world. You can find more information about the individual features in the corresponding training courses.


Tips & measures to increase performance

Low click rate of the recommendation widgets

If a widget has a low click rate, a first optimization can be to place the recommendations further up the page. There they are more visible, because the shop visitor does not have to scroll down.

If two recommendation widgets are placed on a store page, of which the lower widget has higher key figures, it is worth conducting an A/B test. This way, the order of the widgets can be checked.

Positioning of the recommendation widgets

In addition, the position of the recommendations can be tested in different places (product detail page, shopping cart, category page). Experience has shown that, for example, cross-sell recommendations work better at the shopping cart level than on the product detail page.


Rule combination in the Rule Editor

If the recommendations are well placed from the start, optimization can also be achieved by a different combination of rules. Because sometimes just a rearrangement of the rules is enough to be more relevant for the visitor and thus more successful in the online shop (e.g. price filter to limit the maximum price of the recommendations).

Popular recommendations

Identify products that are frequently recommended but little clicked on or purchased. For these products, a more detailed analysis is worthwhile (product description, product image, etc.).

Popular products or brands can be further optimized using features (Banner Management, Result
Management, etc.) in the Control Desk.

Was this article helpful?