Case Studies

Behavior-driven article recommendations for an investor education library

Written by Gavin Garrison | Sep 28, 2026, 6:25:31 PM

The challenge

This global asset manager had a rich library of investor education articles on topics like retirement planning, college savings, emergency funds, and investing fundamentals. After finishing one article, investors were left to sort through an unfiltered list to find the next.

Readers were already showing clear patterns of interest, and the education team saw a chance to put that behavior to work. The obstacle was technical: the articles lived in a custom AEM component, and a save-feature constraint made AEM unsuitable as the delivery layer.

What we did

Trellist designed a cross-platform solution that bridged AEM content with Adobe Target delivery.

  • Readership analysis. With the education team, we extracted and analyzed article readership data to find behavioral patterns and cross-topic interest signals.
  • A constraint caught early. Our assessment of the custom AEM component surfaced the save-feature constraint and prevented a costly technical misstep.
  • Target as the injection layer. We designed an alternative architecture that uses Adobe Target and Velocity syntax to serve recommended articles dynamically on the current page.
  • An 'Articles You Might Like' zone. A new zone at the bottom of every article page surfaces relevant recommendations based on real readership data.
  • Tools the client already owned. The full-stack solution fits the existing content ecosystem and required no new platform investment.