Douyu Live's Digitalization Journey: Data Platform Challenges, Practices, and Future Outlook

As big‑data technologies become widespread, enterprises are forming data‑middle‑platform teams to build comprehensive data and tool ecosystems; this article examines how Douyu Live, a leading game‑streaming platform, views and constructs such a system.

Data Application Challenges – Douyu distinguishes between data infrastructure (collection, cleaning, storage, computation) managed by a big‑data team and data application (query platforms, reporting, business‑specific tools) delivered by the data middle platform. The platform’s evolution follows four phases: (1) extensive monitoring dashboards and email subscriptions to cultivate daily data habits; (2) targeted analyses to answer specific business questions; (3) consolidation of insights into reusable business rules; (4) strategic use of data to guide overall operations.

Key Problems – Decision‑making efficiency is low because massive, diverse data requests rely on analysts for extraction and interpretation, and analysts’ value is obscured by repetitive SQL work. Additional issues include lack of unified data standards, highly variable requirements, and low data‑output efficiency.

Digitalization Path – Initially, simple Excel spreadsheets sufficed; as data volume grew, Douyu introduced self‑service query tools, then a custom reporting platform, and finally a self‑service analytics platform (观远). Product selection considered cost, visual appeal, ease of use for analysts and business users, mobile accessibility, integration capability, and maintainability.

Current Results – The new data portal, built on 观远, offers a unified view across platform, region, streamer, product, and user dimensions, improving analyst productivity and enabling business users to explore data independently. This has shifted analysts from routine data extraction to strategic insight generation.

Future Outlook – Short‑term goals focus on deeper integration of data with business processes, automating routine decisions, and using data to support creator‑growth strategies. The ultimate aim is to make every employee a data analyst, reducing reliance on manual data work and fostering data‑driven decision making across the organization.

Q&A Highlights – After adopting 观远, business users can directly query and analyze data, accelerating problem diagnosis; 360 analysis serves as a downstream platform that codifies successful data‑driven solutions for broader consumption.