Use product recommendations to make sure your customers see dynamic, personalised recommendations at the most optimised time in their customer journey. Use correctly, and you'll notice improved product discovery, increased conversions, and higher order value.
Once you create a product recommendation, it becomes available in EasyEditor. From here you can start to add a recommendation to your email campaigns, pages, or even insert them on your website.
1. Create a product recommendation
Go to Content > Products > Recommendations.
Select CREATE RECOMMENDATION.
Hover over the recommendation you want to use and select SELECT.
Under the Product catalogs side-panel, choose a catalog and enter a Recommendation name
Select Create.
2. Add Inclusion and exclusion filters
You can add include and exclude rules to filter the recommendations displayed into a smaller subset of products based on various attributes.
The product recommendation filtering combines `and` and `or` conditions to create any rules you want. Your filters can target any product attributes, including enrichment tags.
Learn more about enrichment tags in Product data enrichment.
You create filters based on three points: Filter field, Filter operator, and Filter value.
| Filter field | Filter operator | Filter value |
Example | Categories > Name | is equal to | SnowYo |
Add Include rules
On the edit recommendation page, under the header Include, select Click to select product data.
On the Edit rule side-panel, expand the filter field drop-down menu and choose the data field you want to filter by.
Expand the filter operator drop-down menu, and select an operator for your filter.
In the value box, enter the value you want to add to your filter.
If you want to add additional rules, under and, expand the filter drop-down menu and start again.
Select APPLY.
Add Exclude rules
On the edit recommendation page, under the header Exclude, select Click to select product data.
On the Edit rule side panel, expand the filter field drop-down menu and choose the data field you want to filter by.
Expand the filter operator drop-down menu, and select an operator for your filter.
In the value box, enter the value you want to add to your filter.
If you want to add additional rules, under and, expand the filter drop-down menu and start again.
Select Apply.
3. Add a fallback
If a product recommendation doesn't return enough recommendations to fill a block, we show the fallback recommendation instead.
Understanding fallbacks
It is a hard switch between the primary and fallback recommendations. For example, if a block needs four products but the primary recommendation only returns three, we switch to the fallback for all products. Hard switching makes your revenue attribution simpler.
You might use fallbacks if:
You create very granular filtering, and you may not always find enough products.
You have a high turnover of products or low stock levels.
You use long-running automation programs. This way if you forget to update your recommendation, โ for example, if products are no longer available โ we can display fallback recommendations instead.
To add a product fallback:
On the edit recommendation page, under If no recommendations are found, select Click to select fallback.
On the Edit fallback side panel, select Select to view your fallback recommendation options.
Find and select a fallback recommendation.
If you need to change your fallback, select Change fallback. To delete a fallback, select the red X.
When complete, select Apply.
4. Preview product recommendations
You can preview the customer-specific product recommendations while building your recommendation.
To preview a customer's product recommendations:
On the edit recommendation page, select Preview.
Find and select a customer.
You'll only see this option if you're previewing a customer-specific product recommendation. For generic recommendations, you'll see a list of products.
On the recommendation side panel, you can see the ranked recommendations for the customer you selected.
Understanding the recommendation preview
For AI-powered, customer-specific product recommendations, the recommendation preview shows a ranked list of the products that our AI suggests are a good fit for this customer.
The ranked order is the same order that the products appear in the recommendation block โ the affinity score decides this order. The affinity score is a percentile score that tells us how confident we are that the product is a good fit for the customer. The higher the affinity score, the more likely it is the customer will make a purchase.
Learn more in Understanding the affinity score.