Personalization

Better search results and recommendations with minimal effort

The more FACT-Finder modules you use, the greater the effect of personalization. In addition to search results, it is also possible to personalize recommendations and category pages. Furthermore, it is even possible to personalize product sets that are highlighted in a campaign. One of our clients subjected FACT-Finder’s Personalization to a 30-day A/B test. Just within this short test phase, the retailer recorded a conversion increase of 15%, and sales increased by 25%.
No additional integration is required to add this module to your store starting with FACT-Finder 6.11 or better. If the tracking features have been implemented, you can use Personalization immediately. Because of the ease of implementation, users are normally able to achieve a very fast ROI.

Starting from version 7, FACT-Finder allows you to adjust search results based on the preferences of your customers. FACT-Finder determines these preferences bases on the Tracking.

FACT-Finder uses various parameters for Personalization which can be specifically edited:

Active (1): You can decide on a global level, if the Personalization module should influence your results or not.

Influence (2): Adjust the scope of enhancement by FACT-Finder for the preferred product. If you choose a low value, then Personalization will only have a minor impact. Increase for a stronger influence.

Max. record limit (3): Choose if Personalization should affect a limited amount of records or the whole search result. A limitation reduces the processing requirement for FACT-Finder. Especially with search requests with a large amount of hits you can improve performance through this option. Please note that this limitation also means that only a limited amount of records can be influenced by Personalization. If products fitting the preferences are outside the defined range, then they will not be affected.

Max record count (4): Enter the max number of displayed results before cutoff.

Weight (5): Determine how FACT-Finder should evaluate the Click, Cart and Purchase events. The default is 1 for Click, 10 for Cart and 20 for Purchase. This means, that the purchase of a brand article will raise the preference for this brand by 20, while a click on a product of a different brand will only raise the preference by 1. By entering different values you can adjust the results used for preference determination towards one another.

Field weights (6): You can assign a weight to each field of your product data after you have added the field. No weight equals a value of 0, maximum weight equals value 100. If you mainly wish to personalise the brand field you can for example set the value to 20. If all other fields are low or 0, then FACT-Finder will give the brand field a high importance for the personalisation of search results. You can also use multiple fields. If they get similar or almost similar values they will be of equal importance. We recommend working with text fields, because similar text will lead to better preference determinations than numbers, for example. If you combine multiple fields with only slightly diverging values, then the effect is not so easily recognizable. The results will therefore appear much more realistic.

Personalisation has been developed so far, that pushed products can be personalised in campaigns, too. For details, see: Campaigns.