Cart recommendations

This type of recommendations allow you to create product offers on the basis of the products your customers have added to the cart. The offers are prepared individually for each user, so each of them receive a unique offer adjusted to their preferences.

By displaying this type of recommendation on your website you increase the chance that the customer will buy more products, as the products recommended are complementary to the ones in his or her cart.


Requirements

  • Product feed delivered
  • Integration of events via AI tracking or API
  • Periodic transaction updates (one week history)
  • Online data stream including different channels if possible (recommended)
  • Historical transaction data (at least 3 moths backwards)

Configuration

In order to start configuration of the recommendation type, you need to follow the path:

Screen presents the main menu in the application with marked path to Recommendations

Campaigns (1) → Recommendations (2) → Create recommendation (3)

Once you choose Recommendations, you will see the list of all recommendations you have created so far. Each recommendation is marked with a tag that indicates its status (either as draft or active). Recommendations can be edited or deleted at any time.

Click Create recommendation (3) to start the process of configuration.

desc

Create cart recommendations in five steps:

  1. Name and type of recommendations
  2. Number of products
  3. Settings
  4. Filters
  5. Additional settings

Name and type of recommendations

Screen presentes available types of recommendations in the application

This is the place where you can choose the kind of recommendation type you want to create. In this case, choose cart recommendations.

  1. Name - Click the pencil icon to assign a name to the recommendation you’re creating.
  2. Recommendation types - You can choose from seven types of recommendations: cross sell, similar products, cart recommendations, personalized recommendations, visual similarity, top products, and last seen.

Number of products

Screen presents the option of switching on exclusion of products from the cart and the option of choosing the minimum and maximum number of products included in the product offer

  1. Exclude product from cart - This option switches on or off the exclusion of products from the cart. When switched off, your recommendations may or may not include the products from the cart. When switched on, your recommendations for sure will not contain the products from the cart.
  2. Decide how many products you want to display in your recommendation frame. Choose the minimum and maximum value.

Settings

Screen presents three type of sliders which makes it possible to diversify recommendation filters

  1. Product similarity - This slider lets you define the similarity level between products in a cart and recommended products, the higher values you set, the higher dissimilarity level you receive.
  2. Popular boosting - This slider lets you define the importance of the popularity of the products included in the offer. In other words, higher value will boost scoring for most popular products.
  3. Recommendation variety - This slider lets you differentiate the range of the recommended produtcs, so higher value will result in a higher variety within recommended product set.

Filters

Screen present the types of filters

  1. Brand filter - This filter lets you limit the range of products included in the recommendation. You can include or exclude specific brands according to your preferences. If a brand or brands are specified in ‘include’ only products from those brands will be recommended to the customer. When a brand is specified in ‘exclude’ products of all the remaining brands will be recommended.

    Mark as elastic is an option that allows you to ignore the filters in specific situations.

    Tip: Mark as elastic option is particularly useful in a situation when your recommendation frame has empty slots. It may happen when there are not enough products to fill the recommendation frame with the products that fulfill filter conditions.

    After you choose the options, you will see the dropdown list with possible options to select:

    • Custom - Results will be filtered by brands specified in include/exclude field. If you choose this option, you will be able to choose brands which are available in the product feed from the dropdown list.
  2. Category filter - This filter lets you restrict the range of products in recommendations. You can include and exclude specific categories according to your preferences. The set of products available for recommendation will be subsetted to the products in the categories specified in ‘include’. On the other hand if a category is specified in ‘exclude’, that category will be excluded from the set of recommended products, which will consist of products from the rest of the available categories. For example, you want to recommend products available in the category in which TV set is placed. TV set is placed in the main category Electronics, and categories in general can take the structure of a tree. In this case, TV set is in Electronics > RTV > TV.

    • Custom - Results will be filtered by categories specified in include/exclude field. If you choose this option, you will be able to choose categories which are available in the product feed from the dropdown list. To choose a subcategory of the chosen category click on the same drop down menu again. To choose another category click ‘add category’.
  3. Price filter - This filter lets you restrict the shown products in the recommendation to products within a specific prince range. You can include and exclude specific product prices according to your preferences.

    • Amount results - Determines the acceptable price difference between the average price of products in the customers cart and the recommended ones (e.g. products in a range from -25$ (specify -25 in the ‘from’ cell) in relation to the price of the average price of the products in the customers cart to +40$ (specify 40 in the ‘to’ cell) in relation to that price).
    • Absolute results - Determines the absolute price range in which recommended products can be found. The prices of the products in the cart are not important (e.g from 20$ to 40$).
    • Percent results - Percentage difference between the average price of the products in the customers cart and the recommended product. For example selecting values: from = -5 and to = 30, gives products, which price is not less than -5% off and not more than +30% over the average price of the products in the cart).
  4. Discount filter - This option lets you filter the products in the recommendation to discounted products. This filter works only if there is at least one discounted product in the customers cart.

    • Select basis - Specify the basis on which the discount for filtering will be calculated. This means the discount of the product(s) in the cart will be used in the format specified by the basis.
      • Percentage - If you choose this option the discount used to filter the products will be expressed as a percentage, and its base value is the average percent discount value of the items in the customers cart.
      • Amount - If you choose this option the discount used to filter the products will be expressed as an amount, and its value will be equal to the average value of the discount of the products in the cart.
    • Select type - Specify the type of discount value the products will be filtered by, basing on the average discount value of the products in the cart.

      • Amount - Typically used with the percentage basis option. It will give you a discount percentage range based on the basis value. This will be explained below, in the examples.
      • Absolute - This will give you the range as in absolute values, not based on the basis value. Example is given below.
      • Percentage - Typically used with the amount option. It will give you a discount amount range based on the basis value. This will be explained below, in the examples.

      Examples:

      1. Setting a discount filter as -5 percentage points to +15 percentage points to the average percentage discount of the products in the customers cart. First you set the ‘basis’ to ‘Percentage’. Next you choose ‘Amount’ as the ‘type’. For the values you set from = -0.05 and to = 0.15. This means if your basis value of the average percentage discount of the products in the cart is 20% you will filter out the products that are discounted from 15% to 35%.
      2. Setting a discount filter as from 6% to 12%, not referring to the average discount in the cart. Set the ‘basis’ to ‘Percentage’ and ‘type’ to ‘Absolute’. Set values from = 6 and to = 12. This will give you an absolute discount range to filter the products as 6% to 12%.
      3. Setting a discount filter as from 10$ to 20$, not referring to the average discount in the cart. The ‘basis’ to ‘Amount’ and ‘type’ to ‘Absolute’. Set values from = 10 and to = 20. This will give you an absolute discount range to filter the products as 10$ to 20$.
      4. Setting a discount filter as -2% to +7% of the average discount amount of the products in the cart. First you set the ‘basis’ to ‘Amount’. Next you choose ‘Percentage’ as the type. For the values you set from = -2 and to = 7. This means that if the average discount amount of the basket was 100$ you will filter products to the ones that are discounted from 98$ to 107$.
  5. Gender filter - This filter lets you restrict the shown products in the recommendation to the products intended for a particular gender. You can include and exclude a specific gender according to your preferences.

    • Male - Results will present the products specified as relevant for males.
    • Female - Results will present the products specified as relevant for females.
    • Unisex - Results will be filtered by products with unisex gender.
  6. Attributes filters - This filter lets you restrict the shown products in the recommendation to the products that contain specific attributes in the product feed. You can include and exclude specific attributes according to your preferences. Additionally, you can choose range of attributes and indicate their values.

    • Include - Enter the attributes you want to include.
    • Exclude - Enter the attributes you want to exlude.
    • Range - Select the range of the included attributes.

Additional settings

Screen presents additional settings of the recommendations

  1. Boosting - Now your products are selected by means of filters. They are arranged according to a score given by the model. You can influence the arrangement by boosting your popular products score. You can choose from 5 types of metrics. Then use the slider to determine how much you want the metric to influence the score of the products.
    • Sold items count in the last 30 days - The system will favor products that were sold in the highest quantity in the last 30 days.
    • Sold items value in the last 30 days - The system will favor products which summed value, that is how many were sold but also how much they cost, is the highest in the last 30 days.
    • Page visit count in the last 30 days - The system will favor your popular products, based on the amount of times the products page was visited in the last 30 days.
    • Conversion percent in the last 30 days - The system will favor those products which conversion rate in the last 30 days is the highest.
    • Conversion percent after clicking in the recommendation in the last 30 days - The system will favor those products which conversion rate after clicking in the recommendation is the highest in the last 30 days.
  2. Sorting - In the settings of the campaign you specify how many products you want it to return. When the products are filtered, and scores given by the model (with the boosting influence - if chosen), the specified amount of products will be returned. You can choose to sort the resulting products according to some metric. The definition of the metrics is as above.
    • Sold items value in the last 30 days
    • Sold items value in the last 30 days
    • Page visit count in the last 30 days
    • Conversion percent in the last 30 days
    • Conversion percent after clicking in the recommendation in the last 30 days

Saving

Screen presents three possible options: preview, finish later and save

  1. Preview - You can check the preview of the chosen recommendations. In the preview you can create an artificial cart and check what kind of products are recommended to the products in this cart. Type the name of the product and add them to the cart. To do so, click the plus button on the dropdown list next to the name of the product. On the right side you will see products that match all the settings you have configured earlier.

    Screen presents the preview of recommendation

  2. Finish later - If you don’t have enough time to finish the configuration of the recommendation, you can save it as a draft by clicking this button.

  3. Save - If you are ready with the recommendation, you can activate it by clicking the button.

Now, when the recommendation is saved, you can add it to the following campaigns:

  • Email
  • Dynamic content
  • Webpush
  • Mobile push

Adding the recommendation to your campaign

You can add recommendations to emails, webpushes, mobile pushes and on the website as a dynamic content. Decide what kind of campaign you want to incorporate the recommendation to.Regardless of the campaign type you have chosen, proceed to the content of the campaign (let it be an email). You can create a template from scratch or use an existing one. After you make your selection, continue to the code editor.

Screen presents a code editor and arrow marks the folder entitled inserts

Choose Inserts.

description

From the available categories in Screen presents a pop up window with inserts, choose AI Cart Reccomendations and choose the one you have created from the dropdown list.

Screen presents the code string of a recommendation)

Once you click the recommendation you will see the code. Click the option Get HTML code. This option is useful if you want to customize the recommendation. You may also use the basic code presented in the illustration.

Screen presents the HTML code of recommendation

After you click the button, you will see the code. Copy it to the clipboard.

Screen presents HTML code pasted in the code editor

And paste it in the HTML section.

Tip: You need an aggregate that counts the products added to cart in a particular date range.

In the highlighted place in the illustration, paste the ID of the aggregate. You can find the ID of the aggregate in two places:

  • Go to Analytics → Aggregates and open an aggregate you need to include in the campaign. Its ID will be included in the URL of the aggregate.
  • Find Aggregates in Inserts, choose the one you want to use in the campaign, and the code of the aggregate will include its ID.

😕

We are sorry to hear that

Thank you for helping improve out documentation. If you need help or have any questions, please consider contacting support.

😉

Awesome!

Thank you for helping improve out documentation. If you need help or have any questions, please consider contacting support.