BytePlus DataFinder is a one-stop user analysis and operation platform that provides enterprises with digital consumer behavior analysis insights, optimizes digital touchpoint user experience, supports refined user operation, discovers key growth points of business, and improves enterprise efficiency.

1. Comprehensive data collection, make data collection at your fingertips

   A variety of event tracking forms collect data from various platforms such as APP, website, Wechat Mini Program, etc. Aggregate Inter-App communications, cross-touchpoint, and user life cycle data through API to achieve comprehensive insights into users.

2. Build an indicator system so that business effects can be measured

   Link different businesses with a unified data indicator system, conduct in-depth drill-down analysis of key node data, and better use data to measure business effects and gain insight into business problems.

3. Multi-perspective analysis, mining growth points from data

   Aggregate a variety of advanced analytical models, flexibly combine various metrics, and gain insight into users' behavior throughout the lifecycle with real-time data, thus discovering the growth potential behind the metrics.

4. Business growth attribution, overall increase in ROI

   What channel has the best delivery effect? Which resource level has contributed the most to the increase in operating metrics? Use the attribution model to identify the true cause of growth and solidify it into a proven business paradigm.

5. Assist business operations, refine clusters and operate accurately

   Accurately understand users, and circle users with common characteristics according to various demographic/behavioral/business characteristics, and recommend personalized operation activities through Push, pop-up windows, text messages, emails, and other channels to achieve hierarchical operation.

Functional advantages: simple and easy to use, powerful #

• The built-in product provides more than ten simple and easy-to-use visual analysis models, allowing business personnel to handle common analysis scenarios by themselves.
• The product supports multi-dimensional cross-segmentation, flexible drill-down and microscopic scrutiny, and can meet the high-level analysis needs of analysts.
• Good openness, support original data source export, statistical data export, OLAP engine open access, Kafka subscription, query API and other ways for data secondary application.

Technical advantages: massive data real-time architecture, high-performance query #

• 100 billion level data : The system architecture has been verified by the internal large data volume of bytes, which can easily support the 100 billion level data volume to meet the rapid growth needs of the business.
• Real-time data : The data processing full link adopts a true real-time stream processing architecture, and it only takes about 15s for the data to be reported, processed and available (friends are small batch processing architectures, and data processing takes about 32s). Real-time data can generate real-time insights into the business, enabling teams to seize fleeting growth decision opportunities.
• Efficient query : Based on Bytehouse query engine, it realizes second-level response of data dashboard, complex query and other scenarios, and instantly responds to business needs (the average query performance reaches more than 3 times that of friends).
• Storage optimization : The data storage optimization scheme that has been deposited within bytes for many years can achieve a compression ratio of up to 10:1 in the case of large data volume, which can effectively save storage costs.
• Elastic expansion : Growth analysis application layer services are all hosted by Kubernetes, which can quickly expand according to business growth and ensure that online business is not affected. (Friends must carry out the overall expansion operation of the product, and the time period is longer.)
• Self-developed + open source components : The core OLAP engine is based on the most advanced open source engine ClickHouse for secondary development and optimization of nearly 10,000 lines of code, all code is self-controllable, and its optimization point is to combine the advantages of ClickHouse itself, optimize hundreds of UDF query functions for UBA analysis models, and form a solidified module. (Friendship is based on the commercial CDH version, and the core code is subject to the commercial version restrictions.)