Omnichannel Recommendations Action Requests

when selecting recommendation strategies for this type of action, be aware of the following conditions related to the recommendation algorithm configured as part of the recommendation strategy when using the most viewed (product detail page) , top selling by purchase count , top selling by gross revenue , or other noncollaborative recommendation algorithm , you don't need to send any product specific information in the request when using the viewed and also viewed or other collaborative recommendation algorithms , the option selected from base recommendation on determines what context events you must include in the request the following table contains guidance for which context events you must include in the request when the recommendation strategy uses a collaborative recommendation algorithm, based on the option selected from base recommendation on in the strategy configuration base recommendation on event required item(s) viewed on current page model monetate\ context\ productdetailview or model monetate\ context\ productthumbnailview first item viewed on current page model monetate\ context\ productdetailview or model monetate\ context\ productthumbnailview last item viewed in any session no additional events required so long as the customer's unique identifier is associated with a previous completed purchase item(s) viewed in past sessions no additional events required so long as the customer's unique identifier is associated with a previous completed purchase item(s) currently in the cart model monetate\ context\ cart abandoned cart items in past sessions no additional events required so long as the customer's unique identifier is associated with a previous completed purchase last item carted in any session no additional events required so long as the customer's unique identifier is associated with a previous completed purchase item(s) purchased in current session no additional events required so long as the customer's unique identifier is associated with a previous completed purchase item(s) purchased in previous sessions no additional events required so long as the customer's unique identifier is associated with a previous completed purchase last item purchased in any session no additional events required so long as the customer's unique identifier is associated with a previous completed purchase item group id(s) in custom variable model monetate\ context\ customvariables see user identity persistence in requests for more information about passing unique identifiers for customers in engine api requests request examples for noncollaborative standard algorithms this request example is for an omnichannel recommendations action configured with recommendation strategies that use the most viewed (product detail page) , top selling by purchase count , or top selling by gross revenue recommendation algorithm example request for omnichannel recs action using a noncollaborative algorithm { "channel" "a 927c8120/p/example com", "events" \[ { "eventtype" "monetate\ decision\ decisionrequest", "requestid" "11111", "includereporting" true }, { "eventtype" "monetate\ context\ ipaddress", "ipaddress" "188 221 179 186" }, { "eventtype" "monetate\ context\ pageview", "url" "https //www example com", "pagetype" "index" } ] } request examples for collaborative algorithms this example is for recommendation strategies that use a collaborative recommendation algorithm with recommendations based on first item viewed on the current page or item(s) viewed on current page it includes the monetate\ context\ productdetailview event example request for omnichannel recs action using productdetailview event { "channel" "a 927c8120/p/example com", "events" \[ { "eventtype" "monetate\ decision\ decisionrequest", "requestid" "11111", "includereporting" true }, { "eventtype" "monetate\ context\ ipaddress", "ipaddress" "188 221 179 186" }, { "eventtype" "monetate\ context\ productdetailview", "products" \[ { "productid" "product72", "sku" "product72 large green" } ] }, { "eventtype" "monetate\ context\ pageview", "url" "https //www example com", "pagetype" "productpage" } ] } like the previous example, this one is also for recommendation strategies that use a collaborative recommendation algorithm with recommendations based on first item viewed on the current page or item(s) viewed on current page this one, however, includes the monetate\ context\ productthumbnailview event example request for omnichannel recs action using productthumbnailview event { "channel" "a 927c8120/p/example com", "events" \[ { "eventtype" "monetate\ decision\ decisionrequest", "requestid" "11111", "includereporting" true }, { "eventtype" "monetate\ context\ ipaddress", "ipaddress" "188 221 179 186" }, { "eventtype" "monetate\ context\ productthumbnailview", "products" \[ "123", "456", "789" ] }, { "eventtype" "monetate\ context\ pageview", "url" "https //www example com", "pagetype" "index" } ] } this example is for recommendation strategies that use the viewed and also viewed and the viewed and later purchased recommendation algorithms based on the last item purchased in any session you must send the product id of the last purchase example request for omnichannel recs action using viewed and also viewed, viewed and later purchased algorithms { "channel" "a 927c8120/p/example com", "events" \[ { "eventtype" "monetate\ decision\ decisionrequest", "requestid" "11111", "includereporting" true }, { "eventtype" "monetate\ context\ ipaddress", "ipaddress" "188 221 179 186" }, { "eventtype" "monetate\ context\ purchase", "productid" "product43", "purchaseid" "123456789", "purchaselines" \[ { "sku" "product72colour2", "pid" "product72", "quantity" 2, "currency" "gbp", "value" "24 00" } ] }, { "eventtype" "monetate\ context\ pageview", "url" "https //www example com", "pagetype" "index" } ] } this example is for recommendation strategies that use the viewed and also viewed and the viewed and later purchased recommendation algorithms based on the last item carted in any session you must send the product id of the last item added to the cart example request for omnichannel recs action using viewed and also viewed, viewed and later purchased algorithms { "channel" "a 927c8120/p/example com", "events" \[ { "eventtype" "monetate\ decision\ decisionrequest", "requestid" "11111", "includereporting" true }, { "eventtype" "monetate\ context\ ipaddress", "ipaddress" "188 221 179 186" }, { "eventtype" "monetate\ context\ cart", "cartlines" \[ { "sku" "product72colour2", "pid" "product72", "quantity" 2, "currency" "gbp", "value" "24 00" } ] }, { "eventtype" "monetate\ context\ pageview", "url" "https //www example com", "pagetype" "index" } ] } request best practices by implementation type if your site uses a types of implementations , keep in mind these best practices when configuring an omnichannel product recommendation action you only have to send product ids if you base recommendations on a specific product (for example, a recommendation strategy configured with the viewed and also viewed recommendation algorithm based on the last item viewed in any session or the first item viewed on the current page) model monetate\ context\ productdetailview , model cartline , and model purchaseline events are obtained from the monetate tag you're not required to send these if your site uses a types of implementations , keep in mind that you must send all productdetailview , cartline , and purchaseline events