Marketing Technology Architecture and Challenges at Taobao

Every year the Double‑11 shopping festival attracts billions of consumers and millions of merchants. Taobao's marketing technology team has accumulated extensive capabilities to support large‑scale promotional activities.

The marketing business is divided into ToB (organization and supply) and ToC (presentation and conversion). ToB involves platform‑level merchant onboarding, product selection and rule definition. ToC focuses on delivering offers through venues, channels, live streams and other marketing scenes.

Typical technical challenges include:

Handling millions of merchants and tens of millions of product submissions with fast rule validation.

Reducing the cost of merchant asset creation.

Ensuring data consistency for tens of millions of items across many downstream systems.

Providing clear discount information to users during checkout.

Maintaining price‑as‑seen consistency across all pages.

Fast, accurate price‑adjustment refunds for post‑event price protection.

Supporting ultra‑high‑concurrency lottery draws (millions of QPS).

Optimizing page rendering for low‑end devices.

Building reusable 3D/XR interactive scenes.

To address these, a platform‑centric architecture was built, separating domains such as招商 (merchant acquisition), 投放 (ad placement), 权益 (benefits) and 搭建 (scene construction). Standardized domain models enable 80‑90% feature reuse and rapid extension.

High‑availability solutions include multi‑level rate limiting, fallback mechanisms, and progressive bucket inventory for concurrent discount deduction, ensuring service continuity even under massive traffic spikes.

Data‑driven and AI techniques are applied for algorithm‑assisted operation (e.g., automated merchant selection, schedule optimization) and for full‑process automation. Low‑code tools like imgcook generate front‑end code from design drafts, and a JVM‑level lightweight container isolates core platform functions from plug‑in extensions, reducing deployment time by over tenfold.

AI‑generated image cropping assists merchants in producing multi‑size assets, and AIGC‑driven interactive experiences are explored for future campaigns.

The article concludes with a call for engineers to join the team, providing a contact email for applications.