Audit, improve, and standardize legacy content modules.
Context
Amazon’s A+ Content Manager (ACM) modules were used to guide sellers in all A+ content creation, serving as a starting point and container for a product’s rich content development. They had evolved over time across multiple teams, workflows, and priorities, and most modules contained inconsistent placeholder content, outdated examples, and unclear guidance.
Because of the integral nature of modules on both the front and backend of ACM, I made my first project at Amazon to conduct a large-scale module audit.
Audit findings & recommendations
After my audit, I determined that the use of modules should be deprecated in favor of a more cohesive content creation experience where sellers could build content holistically. Modules were ultimately outdated, fractured, and confusing, and the experience couldn’t compete with modernized B2B content-management competitors. I recommended moving toward WYSIWYG moveable block components.
Due to technical considerations and business priorities, we couldn’t overhaul the whole module system and content creation architecture so I refocused my approach:
- Sellers should be able to access all features that users expect in digital content experiences.
- Modules used in nearly all content layouts (image/text/video) should be available for both basic- and premium-tier sellers, whether for products or brand storytelling. (Because of the piece meal nature of how modules had been added, there was an arbitrary line between seller tiers.)
- Any improvements to how a module functions should be standardized and applied anywhere relevant. This includes high-impact elements that significantly improve content engagement, like videos, a key component in rich content experiences.
- Priority on versatility and reusability: modules that can be adapted for different content types without needing a separate version.
- Modules that serve a unique purpose but aren’t universally required should be built as “enhancements” that are universal to the UI rather than module specific.
Some of the most used modules were built to be “AI-ready,” which created both duplicative modules as well as an unnecessary discrepancy between modules. If the AI aspect could be treated like an enhancement or additional functionality until modules could be deprecated completely, sellers could use gen-AI across multiple modules rather than having to create separate content within individual modules; the module setup didn’t allow for all the content pieces built for a product to “speak each other” in a useful and intuitive way.
To bridge the gap between individual modules and holistic drag-and-drop content, I recommended consolidated modules with flexible layouts, where text, images, and videos could be added similarly to competitive software.
Goals
Separating modules from enhancements and features required another interim step—cleaning up and modernizing the module system with a focus on improving:
- Clarity around “AI ready” modules
- How generation worked
- What sellers were responsible for reviewing
- Where generated assets were saved
- How to regenerate or refine outputs
- Duplication on the backend, causing loading delays and coding confusion
Doggone it
The use of dog imagery and copy wasn’t useful to sellers when setting up product content, and these legacy modules no longer reflected how the system was evolving as a whole.
Module examples used inconsistent (and ultimately unhelpful) placeholder content that required restructuring and standardization:
Replacing placeholder content with structured guidance
I worked with product design to update module examples and guidance to better display product info and increase seller clarity around content creation. This was less about rewriting individual modules and more about improving the coherence of the system.
Combining modules with low usage & overlapping purposes
We factored in usage data when deciding which modules to prioritize and which to deprecate.
A comparison chart of comparison charts—old designs vs. updated designs demonstrating instructional tooltip upon hover:
Designing for future scalability
As AI-assisted features expanded, module guidance needed to take into account generation workflows, structured prompting, modular reuse, and scalable education patterns.
(In an adjacent project, I worked with product design on multi-image generation flows, image selection behavior, regenerate/refine loops, and generation settings exposure, helping bridge older publishing systems with emerging AI workflows.)
Final designs featured updated lifestyle imagery, clear AI-integration instructions, and the proposed module selection and preview experience inside A+ Content Manager.
Takeaways
The module system reflected years of incremental growth, where new functionality was layered onto old structures without reexamining the underlying model.
This project was a reminder of how often product teams find themselves designing for two realities at once: the system that exists today and the system they’re moving toward. My role was helping bridge that gap—improving the current experience while identifying opportunities to support more flexible, scalable content creation in the future.