Enhancing Brand Consistency With Natural Language Search (NLS) In DAM

Abstract

This graduation report presents the results of a project for the Master Strategic Product Design at TU Delft, in collaboration with Bynder, a digital asset management company. As Bynder's digital asset ecosystem expands, a strategic opportunity emerges to innovate its brand governance tools. Discrepancies have appeared between the words employees use in Natural Language Search (NLS) and the system-generated tags that enable asset retrieval. This project explores how Bynder can evolve NLS from a generic search feature into a brand-aware capability that strengthens brand consistency. The research integrated a double diamond framework with a Research-through-Design approach, using literature reviews, competitive analysis, and stakeholder interviews. The primary finding is that while Bynder's taxonomy provides a strong foundation, it could be enhanced to better accommodate the dynamics of contemporary brand language. A secondary finding points to a market opportunity for DAM systems that leverage intelligent automation to enforce brand governance. The resulting design solution consists of three parts: a Brand Dictionary Configurator for admins to add brand-specific terms; a Guided Prompt Framework to help users phrase brand-aligned queries; and a Brand-Language Feedback Loop to help admins manage and update search terms. Strategically, this framework evolves the Bynder platform from an asset management solution to an active brand partner. Operationally, it provides clients with superior asset discoverability, ensures greater coherence in brand storytelling, and delivers a measurable decrease in the resources required for governance.