Can we transact? Should we transact? Can we fulfill the promise profitably and reliably? Lu Zhang uses those questions to describe a recurring decision problem she encountered across years of work in merchant operations, international fulfillment, and platform transactions. She began formalizing the patterns behind those decisions into a framework for determining how commercial and AI-initiated actions should be governed.

“The same kinds of decisions kept appearing in different environments,” Zhang said. “The products changed, the companies changed, and the operating conditions changed, but the underlying structure kept showing up.” That realization became the foundation for the Commitment Decision Framework for Transaction Systems, an independently authored approach to governing when commercial and AI-initiated actions should proceed, under what constraints, and with what traceable decision record.

The framework grew from recurring problems Zhang encountered while working across merchant systems, transaction infrastructure, and cross-border fulfillment. Different companies presented different operating conditions, yet Zhang kept encountering decisions whose outcome depended on information scattered across multiple parts of the commercial system. What interested her was whether those recurring patterns could be separated from the individual products where they first appeared and expressed in a more reusable form.

“If the same class of problem keeps appearing in different systems, solving it once for one company only gets you so far,” Zhang said. “I became interested in what could be abstracted and carried into another environment.” Her work shifted from addressing each operating problem within its immediate context toward identifying the common decision logic underneath them.

The Commitment Decision Framework is Zhang’s attempt to formalize that logic. It addresses when commercial and AI-initiated actions should proceed, the constraints that govern them, and the traceable decision record surrounding those actions. Zhang’s broader aim is to make decision structures that once lived inside individual teams or platforms easier to express, examine, and eventually reuse across different commerce settings.

That work has begun attracting discussion outside the companies where Zhang first developed her thinking. Major Matters published a three-part independent editorial series examining the Commitment Decision Framework and its application to agentic commerce. Zhang has also published practitioner articles through the Institute for Supply Management on executable governance, cross-border operations, and autonomous AI agents. Her work has additionally included a public NIST submission and direct technical correspondence with NIST Information Technology Laboratory personnel around agent traceability and auditability.

The external discussion marks a change in the setting for Zhang’s work. Problems she first encountered inside individual companies are now being examined through professional publications and standards-adjacent technical exchanges. Zhang’s current challenge is translating operating knowledge developed inside individual organizations into decision infrastructure that can be used across different commerce environments.

MKT//ENTRY is the first working application of that effort. Zhang is developing the AI-assisted decision product to help cross-border merchants evaluate whether a product should enter a new market, what pricing or operating model could support that move, and what would need to change when the current setup is not commercially viable. A market can look attractive when demand is considered on its own. The picture can change once tariffs, advertising costs, returns, fulfillment requirements, and local compliance obligations are considered together. MKT//ENTRY is intended to surface those questions before inventory, logistics resources, and marketing commitments are already in motion.

“Once resources are committed, a weak decision becomes much more expensive to unwind,” Zhang said. “The earlier you can surface the constraint, the more options the merchant still has.” MKT//ENTRY is designed to bring that evaluation forward, while a merchant can still reconsider the market, pricing approach, or operating model before committing scarce resources.

The product also reflects Zhang’s interest in making integrated commercial decision-making more accessible to smaller merchants. Large platforms can devote substantial internal resources to market analysis, compliance, fulfillment, transaction infrastructure, and operations. Smaller businesses often make many of the same consequential decisions with fewer people and less specialized support. Zhang’s stated goal is to make more of that integrated decision capability available before merchants commit capital and operating resources to expansion.

That objective connects directly to the framework’s portability. Zhang is not simply trying to reproduce the internal systems of a large commerce platform for a smaller company. Her focus is on identifying the decision structure that can travel across settings, while allowing the relevant commercial and operational constraints to change with the merchant, market, or transaction.

“The framework is really about taking something that used to live inside one operating context and asking what part of it can become reusable,” Zhang said. “The conditions may change from one business to another, but you can still create a clearer structure for how the decision is made.” That distinction is central to her move from company-specific implementation toward reusable decision infrastructure.

Longer term, Zhang sees the work extending beyond MKT//ENTRY. Her aim is to develop a reusable decision layer that could eventually sit inside commerce platforms, merchant systems, planning and fulfillment software, and AI-agent workflows. The focus would remain on the recurring problem she identified across years of operating experience: turning fragmented commercial conditions into a decision structure that can be applied across different environments.

For Zhang, the evolution of the work is as important as the framework itself. She began by solving problems inside individual products, then recognized that the same kinds of decisions were appearing across different parts of global commerce. Now she is trying to turn that accumulated operating knowledge into something more portable. The question guiding the next stage is how far that decision structure can travel beyond the individual organizations where Zhang first encountered the problems it was designed to address.