Why Alibaba Is Dismantling Its Middle Platform and What It Means for Innovation

According to insider information, Zhang Yong recently posted on Alibaba’s intranet expressing dissatisfaction with the current middle platform, stating that Alibaba’s business development is too slow and the middle platform needs to become thinner, more agile, and faster.

Thus, the debate concludes: Alibaba is completely dismantling its middle platform.

In 2015, Zhang Yong introduced the “big middle platform, small frontend” strategy. Five years later, he is now tearing down the platform he built. The article asks what this means, whether the strategy was wrong from the start, and proceeds to dissect the issue.

Alibaba’s “big middle platform, small frontend” has become an industry standard.

Zhang Yong became famous with the “all‑in mobile” battle, launching the Taobao app that helped Alibaba secure a leading position in the mobile internet era.

As Alibaba’s top leader, he also proposed the “big middle platform, small frontend” strategy. After more than five years, the middle platform seems to have become an industry norm, with larger companies building their own.

The original goal of the middle platform was to create a unified technical architecture, product support system, data‑sharing platform, security system, etc., to support diverse business forms.

Undeniably, Alibaba’s middle platform has strongly supported business growth over the past five years, contributing to increasingly stable systems during events like Double 11.

What problems can’t the middle platform solve?

After the “big middle platform, small frontend” strategy, Alibaba introduced the “Five New” strategy: new retail, new manufacturing, new finance, new technology, and new energy.

With Alibaba’s scale, the strategy moves beyond minor tweaks to address sustainable growth and second‑curve challenges, aiming for disruptive innovation.

Result: Alibaba’s “new manufacturing” is represented by Rhino Manufacturing; its CEO said they need an independent organization with its own business, technology, development, and product—similar to an independent company. Even with a strong middle platform, early‑stage businesses find middle‑platform resources inefficient.

Similarly, the Taobao Special Edition, seen as Alibaba’s answer to Pinduoduo, is also an independent organization from product to technology to operations, to pursue depth and speed of innovation.

Does the middle platform fail to support enterprise innovation?

The middle platform is not unsupportive of innovation; it actually incubated products like Hema Fresh, enabling rapid development of a full online‑offline new‑retail system.

Accurately, the middle platform is suitable for “combinatorial innovation” but not for “disruptive innovation”.

Combinatorial innovation combines existing capabilities, emphasizing standardization, which the middle platform excels at. For example, Hema Fresh reused middle‑platform services such as product, inventory, user, payment, AI, and security to create a new retail species.

Disruptive innovation breaks existing front, middle, and back ends, overturning current models—e.g., smart manufacturing or smartphones. You cannot create it on existing production lines.

Therefore, the middle platform does not support disruptive innovation; this limitation is determined by its DNA.

An Alibaba middle‑platform architect told the author that the current goal is to keep only the most abstract parts in the middle platform, making it thin, leveraging accumulated generic capabilities to improve efficiency, reduce manpower, and free people to work on front‑end personalization.

It’s not that the middle platform is ineffective; the scenario has changed. The middle platform remains excellent for improving organizational efficiency and enabling combinatorial innovation, still applicable for many large enterprises, but companies like BAT and TMD face more disruptive‑innovation challenges.

From Alibaba’s middle‑platform evolution, it will become thinner—a natural trend. This aligns with the author’s previous “middle‑platform fragmentation” concept: the platform will become more fragmented and lighter.

Thus, enterprise leaders should not blindly follow the trend of thinning their middle platforms. Digital transformation requires long‑termism; reckless jumps can be costly.

Reference: “Key Figures and Organizational Design in Alibaba E‑commerce”, LatePost.