The power of Datatrics is to let the algorithm do the work. It decides when to show and what kind of products. Sometimes you would like to give certain products an advantage because you get more profit from them, or you would like to clear the stock because it is the end of the season. With recommendation ruling, you can adjust the algorithm and show those products more often.
Potential attributes to use
The more diverse your content feed is, the more options you have to set a specific rule to promote products. This inspirational help document about a feed could help you extend the feed you already have.
when a new fashion season starts, you want to recommend these products on the homepage sooner than older products. Therefore, you can use certain targeting to prioritise these products. In the example below, U Digital used this for their customer to promote products that are new this season. This recommendation is based on the attribute "NEW IN". It will show earlier if the customer is interested in a specific type of product and this product has a newer variant.
In Datatrics, you can set this via recommendation ruling. You search for the correct attribute and give it a higher weight for the algorithm.
New in products:
Underneath, you can see the product recommendation before the recommendation rule.
As you can see, an attribute with the name “SALE” instead of “NEW IN”.
And after you set the recommendation ruling, you get different products, as you can see below.
These products contain the attribute “NEW IN”.
You can also do this for a certain brand because this brand got a higher margin, for example. In the example below, you can see how this works.
Before the new ruling, you could see this product below.
And afterwards, you will see a product that came along with the targeting. The algorithm took the ruling into account and showed a product from the brand “Virtufit” based on the ruling.
The same you could do with color. If a customer is looking for black shoes, you would like to make the color black more important than other colors, which the algorithm can take into account. So if you upper the weight for color, it is more likely to show a black shoe on the product page as a recommendation. In the example below, you can see how this works.
Before the recommendation ruling, it is possible to show different colors of shoes.
After the ruling, it is more likely that the recommendation contains black shoes, as you can see below.
Some other examples you can use without an example:
At the end of a season, you want to get rid of your stock. You can set stock as a rule and target products within a specific range of stock you would like to promote more often than other products with a higher/lower stock.
Price can be important for some customers, especially the economical customers. With the ruling, you can give the price below a certain amount a higher weight. If a customer on the website is more likely to purchase a product with a low amount of money, the algorithm will consider the price even more.
The category could also be essential to show on a product page. The algorithm will take the category page of a product into account when deciding which products are the most relevant for every customer.