Expand my Community achievements bar.

AEP Modeling to Drive Engagement and Revenue Through Adobe Campaign


Level 3


Company Name:Bounteous, in co-innovation with a leading clothing retailer

Company URL: https://www.bounteous.com

Your Name: Jim Jurich

Your Title: Technical Architect


Describe your company, the customer experience and business challenge(s) you set out to solve with Adobe Experience Cloud products, and how long your company/organization has been using Adobe Experience Cloud products.

Company is an American outdoor and workwear retailer that creates sustainable products for hard-working, hands-on people.  Bounteous was contracted to implement Adobe Experience Platform to gain better insights into customer behavior across all touch points.  Bounteous was also engaged to implement Adobe Campaign Standard to streamline email marketing and stand-up SMS as a new channel.

Retailer had several goals:

  • Improve customers’ overall digital experience
  • increase digital marketing engagement
  • Drive revenue
  • Leverage the Unified Profile and its data to expand the reach and relevance of their email program
  • Create personalized recommendations for users and build out more Dynamic Content Blocks in their existing Adobe Campaign workflows
  • Create a single model versatile enough to use for omni-channel orchestration
  • Increase campaign efficiency through automated workflows syncing profile and segment data across AEP and Adobe Campaign

There were challenges to overcome to realize these goals:

  • Although using Adobe Target’s out-of-the-box Product Recommender Model for onsite personalization, there was no obvious approach to extend this level of personalized content into their email program
  • Complex business rules were employed in Adobe Campaign, but the rules were not sophisticated enough for their use case and were a challenge to maintain at scale
  • To measure the value of these efforts, they needed A/B testing that would allow for the preservation of test vs. control audiences across multiple sends (extending beyond the out-of-the-box testing capabilities of ACS).​

Through the implementation of Adobe Experience Cloud solutions, the retailer and Bounteous have focused on personalization and insights to drive toward realized value: amplifying existing client wallet share, increasing individual basket size, improving sales conversion, and expanding customer reach & awareness.


Describe how you have integrated and used multiple Adobe Experience Cloud products to solve these challenges to improve and personalize the customer experience/journey. Please provide information that will be helpful in understanding your integration (e.g. architecture diagram, step by step process integration flow, etc.).

Bounteous built and deployed a custom product recommender using machine learning to drive 1:1 personalization with refined audience segmentation and content targeting to improve email campaign relevance and engagement. 

The model selected and ordered products to be recommended on an individual customer basis, and the model data from AEP was used by Adobe Campaign to drive the recommendations displayed in the emails.


Deploying this model in AEP not only enabled layering personalization capabilities within a dynamic email but also the added capability for A/B testing of AEP models versus generalized product rankings from Adobe Analytics.  We had two use cases where we leveraged A/B testing capabilities to determine the success of the model in-market. We structured A/B testing within AEP to designate each user randomly into Test vs. Control (from Adobe Analytics data) groups and maintain each user’s designation over time. Thus, if the test email campaign extends over various sends, each user receives the same experience every time. Not all tools with built-in A/B testing capabilities will do this, but our solution removes confounding signals and bias from the testing results. By implementing this advanced model, Bounteous enabled the retailer to achieve their objectives.






AEP Product Recommender Model

The product recommender model leverages historical user behaviors, including both online and offline purchases, and recommends up to 20 relevant products for each of the retailer’s customers. The model applies various filters downstream that even further customize the products being shown to match different intents based on the specific email campaign. Because results are fed into the Unified Profile, the model can be scaled across multiple channels in the future.​


Only customer purchases for the last year were considered, and products no longer available were excluded. Bulk buyers, i.e., those who purchased 115 or more items, were also excluded from the model.




Email Delivery

The Product Affinity email is sent to customers who purchased online 7 days ago.  For each customer, the email presents the top product recommendation as the hero, and four additional recommendations as thumbnails.  The email is suppressed if the customer received the Product Affinity in the past seven days, or if the customer received Abandoned Browse or Abandoned Cart emails in recent days.


Based on your successful use and integration of multiple Adobe Experience Cloud products, describe how it has transformed the customer experience/journey, and the value, business impact, and results your company/organization has realized. Please cite both qualitative and quantitative results as applicable.

Bounteous identified two use cases for proving the value of these new data capabilities, and orchestrated A/B tests for two distinct email campaigns. In both use cases, the retailer was able to deploy custom A/B testing frameworks to share data between AEP and ACS.  In one case, the Unified Profile allowed us to expand the reach of retargeting by 3X.

The first use case was a Daily Promotional Email Campaign with a Dynamic Banner surfacing relevant products for each individual user. This campaign has a wide breadth and drives upper-funnel engagement and awareness of brands and products.  This test measured the impact of our product recommender versus no personalization, which drove 10% more users to purchase over non-personalized content, leading to an additional $245k in revenue from just three email sends. 

The second use case was a Post-Purchase Retargeting Email Campaign with a Dynamic Banner surfacing relevant products for each individual user. This campaign has a smaller reach, but drives repeat engagement for users to purchase more products, more frequently. This test measured the impact of our machine learning model up against a rules-based model, which drove a 5% lift in basket size over rules-based personalization for users who purchased, all powered through AEP and ACS integration.

Aspire Campaign Experience Platform