5 Considerations for a Successful Adobe Target Product Recommendations Implementation

Are you considering ways to personalize and optimize your website? Adobe Target Product Recommendations can be a powerful tool that allows you to serve relevant products, or other content, to your visitors based on their actions, attributes, and defined business rules. To get the most out of this personalization engine, it is important to spend the necessary planning time up front to ensure a successful implementation.

Include these 5 considerations in your planning toolkit:

1. 1. 1. 1. Recommendation Type
            The first step is to determine what type of content you will recommend with your “Product Recommendations.” Although the name suggests you would be recommending “products,” you can use the feature to recommend other types of content depending on what you need for your site. Pro tip: If you are going to recommend traditional products, consider your product catalog.
            • Q – Do your products have different variations that would each need a unique product or SKU, but ultimately roll up into the same master/parent ID?
              • For example, a size Large Black T-Shirt and a size Small Black T-Shirt have different IDs but are in the “Black T-Shirt” grouping
            • A – If so, consider utilizing the master/parent ID as your main product id (entity.id) to optimize the Adobe Target algorithms and increase the product variety in your Recommendations.
         2. Recommendation Placements
            Knowing which website pages you plan to implement Recommendations can determine the types of algorithms you will want to use, and in turn, further help you determine the parameters to pass to Target on each page. Common places for Recommendations include your Homepage, Product Category Pages, Product Detail Pages, Shopping Cart, and Deals page.
         3. Recommendation Content
            Identifying the content you would like in your Recommendation zone will guide which parameters you pass to Target throughout your site or in your Product Feed. You will want to include the basics like a name, image thumbnail, and pricing, but don’t forget about the descriptive content that will make a product recommendation more compelling to your customers. For example, does your product have:
            • A star rating
            • Promotional messaging
            • Discounted pricing
            • Notable inventory levels
         4. Recommendation Logic
            In addition to the content you share with customers, consider the type of product to display – do you want to:
            • Exclude products that are on clearance?
            • Only include products that are on sale?
            • Exclude products that are out of stock?
            • Give preference to products with a higher margin?
            • Promote house brands?
            • Ensure variety in your Recommendation zone by using certain characteristics such as product parent, product type, price level, etc.
         5. Recommendation Data Source
            Consider where each piece of information you identified above will be sourced. Is it something you can pass to Target via a parameter upon page load? Is it something that should be uploaded in a product feed on a scheduled basis? Will it come from both? If you are going to use a product feed, explore your options for getting this set up:
            • • Is there an existing product feed that you can leverage?
              • Which available method(s) will you use? To name a few options:
                • Google Product Search feed – popular for many retailers
                • Custom CSV feed
                • Adobe Analytics Product Classifications – note, this may not be recommended due to data delays

            Does all of your data need to be in a single feed? Or would it be more effective to have multiple feeds based on the type of data you include? For example, you may have an automated feed that populates each night with updated product data from your commerce system. In addition, it might be valuable to have a manual feed that Marketing can load with key promotional details

            • Multiple feeds might be necessary if product inventory information is housed in a separate system from general product information (product detail page URL, thumbnail URL, name, brand, etc.)
            • Keep in mind the following if using multiple feeds:
              • Each feed must use the same common product ID
              • If the same field is in multiple feeds, the product catalog within Adobe Target will show the most recent value for that field – note, having the same field in multiple feeds is not recommended

Considering any of the above in your planning process will help to ensure you get Adobe Target set up properly to enable effective Product Recommendations. And then the fun begins – designing your Recommendation zones, defining product catalogs and custom criteria, testing algorithms and design to optimize results… but those are topics for another day.

Happy optimizing!

About the Author
Caitlyn Phillips is a member of Ansira’s Data Science & Insights Center of Excellence and an Adobe Target Certified Expert. She pushes herself to think outside of the box to truly understand what drives consumers and how to translate those answers into actions to help clients.