Data Tracking (埋点) Application Scenarios, Workflow, and the Seven‑Word Guideline

Introduction – Data tracking (埋点) records user actions (both passive and active) to support business and product needs, such as counting banner clicks during a promotion or measuring exposure of recommended items. The collected events are reported via SDKs, aggregated, and analyzed to guide product optimization and operations.

Application Scenarios

Understanding overall metrics: PV, UV, exposure counts, user numbers, repurchase rates, etc.

User behavior analysis: usage habits, decision paths, heat‑map distribution.

Product trend monitoring: daily traffic, lifecycle stages, and data changes before/after major campaigns.

Feedback for product iteration: conversion funnels based on browsing, clicking, favoriting, adding to cart, commenting, sharing, and related conversion rates.

Backend Data Tracking Classification

Exposure tracking – records when a page region is viewed (e.g., app home page, WeChat Moments ad slot, TikTok splash screen). One view per user counts as one exposure; repeated exposure is counted only if the region changes.

Click tracking – records any user click (e.g., cart button, image, banner). Clicks are active behaviors and are logged separately from exposure.

Page‑event tracking – records various page‑level attributes such as URL, user ID, device info, source/target pages, and product details.

Workflow of Data Tracking

The complete workflow includes requirement collection, review, implementation, testing, deployment, and post‑mortem. Key participants are business owners, data product managers, advertising product line, page product line, front‑end developers, testers, and data engineers.

1. Requirement Submission – Operations or product managers propose tracking points (e.g., app splash, homepage banner) and specify target terminals (App, H5, WAP, Mini‑Program, etc.).

2. Requirement Clarification & Solution Design – Define tracking position, parameters (page code, module code, product code, etc.), terminal type, and template name.

3. Requirement Review – Review documents with product managers, front‑end, back‑end, data collection, and data warehouse teams.

4. Development – Front‑end developers implement exposure, click, parameter injection, and asynchronous requests.

5. Integration Testing – Verify correctness of exposure and click events and completeness of parameters in a test environment.

6. Deployment – After testing, deploy on the agreed date and validate through order attribution or similar checks.

7. Post‑mortem – Update verification results, list released tracking items, and summarize issues for future improvements.

8. Data Statistics & User Behavior Analysis – Some companies build a tracking platform to aggregate daily/weekly/monthly metrics, compare trends across campaigns, and derive insights such as user retention or conversion rates.

Seven‑Word Guideline (七字诀)

位 (Position) – Define the exact location to embed the tracking point.

埋 (Embedding) – Align with tracking standards and let front‑end implement.

时 (Timing) – Schedule integration testing and release dates.

测 (Testing) – Conduct tests during integration and after deployment.

传 (Transmission) – Ensure parameters are correctly sent to data collection and warehouse.

表 (Table) – Store the data in real‑time or offline Hive tables.

统 (Statistics) – Perform statistical analysis after successful validation.

For more practical cases, templates, and testing samples, refer to the book “Data Product Manager: Solutions and Case Studies”.