Research for decisions, not content for clicks.
Commerce Lab examines one consequential Amazon operating decision at a time, using measured data from accounts under management — published with its methodology, its confounders, and its limitations in full view. Every study states what the data can support, and stops there.
Five tracks, one standard
Variation structure, hero-ASIN concentration, SKU rationalization, and catalog architecture decisions.
Branded spend, advertising dependency, and what advertising actually contributes beyond what would have sold anyway.
Subscribe & Save economics, repeat purchase behavior, and post-acquisition contribution.
Creator commerce, TikTok Shop, and the branded-demand halo across channels.
Internal versus outsourced team cost, channel overhead, and contribution margin structure.
Every publication carries its evidence label
The label sits next to every headline claim, so a quoted finding carries its own qualification with it.
Based on publicly available marketplace information.
Based on private account or campaign data, published with the account holder’s permission and anonymized.
Produced through a deliberately designed experiment with a stated hypothesis and comparison.
Aggregated across multiple businesses or datasets, with sample and selection method disclosed.
Each study documents its own design, confounders, and limitations in the body of the publication — not in a footnote. If the data cannot support a conclusion, the conclusion changes.