Within a highly scaled platform like JD, data, efficiency, and supply chain capabilities form the core strengths of developing own-brand products. However, in food—a category highly dependent on intuitive judgment and long-term accumulation—does systematization and scale introduce new constraints? This session will draw from JD’s own-brand experience to discuss which capabilities become harder to maintain as the platform scales, and how these challenges influence long-term decisions in food selection and innovation.

Key discussion points:

Amid massive amounts of data, which signals are actually unreliable? Among metrics like search, clicks, conversion, and repurchase, which are short-term noise and not suitable to directly guide new products?

When data consistently points toward “faster, cheaper, more homogenous,” how does the platform preserve long-term product direction?

In “slow-variable” categories like food, should data lead or lag decision-making? Which food trends inherently show delayed signals? How does JD capture shifts in sentiment, health, and culture ahead of the data curve? Is data more suited for discovery or validation in food selection?

As platform scale increases, which capabilities paradoxically become more difficult? Food innovation often requires small-batch, multi-round experimentation with unstable outcomes, whereas JD’s system is designed for predictable quality and large-scale replication. Where does JD insist on maintaining scale, and where does it intentionally sacrifice scale to preserve diversity?