Recommended product sets
Encouraging customers to increase the value of their basket is very important in ecommerce activities. To do this, it is worth recommending products not necessarily from the same category as the product customers are looking at, but complementary to it. You can display products that together can create a complete and useful set.
Thanks to this, users can create their own perfect set of products and at the same time, you can boost revenue and the number of products in the basket.
Example of use - Retail Industry
We prepared some recommendation sets for a client in the retail industry. The client used a similar solution on the website before, but made it manually, which was time consuming, especially because of the number of products on their site.
We proposed an AI-driven mechanism offering the possibility to compose their own sets and adding products to carts in a group or separately.
So here we can see AI-recommended products composed for the main product, which is a bicycle.
The user can choose additional products from every category on his own and add them to his set. The price for the whole set is calculated dynamically.
Users can also add the whole set to the cart, using one button.
But if for some reason a client decides that a particular product, e.g. a bike pump, is not right for him, he can choose another one and add it separately to the card as an additional product using the button directly below the chosen one.
It is worth mentioning that each type of bike has different categories of products in their recommendation sets. For example, city bikes have a basket and RTB bikes have materials to keep the bike clean and so on.
- Similar model trained
- Color and series of products in the product feed – as custom attributes
- Tracking code on the website
- OG tags implemented – especially product:retailer_part_no which is the same as <g:id> in the product feed and og:category which is also the same as <g:product_type> in the product feed
How to do it
What is important, each item in the set is another recommendation with some extra filters. Therefore, you have to create an AI recommendation campaign for every product.
Go to Campaign – Recommendations, and create a new campaign.
Use cross-sell recommendation because we want to get products that are most often bought together with our main product.
Remember to define how many products you want to show. In this particular case we will show maximum 6 products per our slot.
Next, define the main category which in this case will be running
With the help of attributes filter you are able to set your recommendation to showing only products which have some specific custom attributes. In this particular case we are using custom attribute which we have in our product feed and its name is the nature of product. It told us what this specific item is. You add here that you want to show only products which are bike pumps.
Do the same for other slots choosing like in this case: bike lights and water bottle holder.
Also remember to add the conditions where and when the campaign should be displayed. You can construct the conditions in this way to make it possible to display the campaign only on specific urls or you can use also OG:tags.