How Data Thinking Transforms Product Design: From Insight to Action

In the data era, designers are expected to use data not just as a bonus but as a core skill to discover problems, set quantifiable goals, and produce assessable solutions.

What – Identify Key Issues Through Data

By examining three seemingly unrelated data points—page view rate, Q&A module click‑through, and average chat messages—we can conclude that users cannot find the information they need on the page.

Why – Analyze Underlying Causes

To avoid one‑sided thinking, we apply the five product‑design elements across different layers:

Presentation layer : Content format (text, image, video, tags, cards) may be insufficiently noticeable.

Framework layer : Poor information hierarchy and placement reduce exposure.

Structure layer : Required user actions interrupt information flow.

Scope layer : Missing content that users actually need.

Strategic layer : Issues related to product positioning beyond the information itself.

How – Design Solutions Based on Causes

Using user data (online communication, Q&A queries) together with research methods (surveys, interviews), we pinpoint the problematic layer and apply targeted improvements:

Presentation: Strengthen and make content formats more eye‑catching.

Framework: Adjust content priority and placement to increase exposure.

Structure: Surface information earlier, optimise interaction flows.

Scope: Add the content users are looking for.

Strategy: Refine the business model based on research insights.

Core Thinking Skills for Data‑Driven Design

Goal Thinking

Clear objectives keep design work on track. The OSM model helps translate business goals into design strategies and validation methods, linking company objectives, tactics, and data collection.

Metric Thinking

The AARRR funnel (Acquisition, Activation, Retention, Revenue, Referral) provides a structured way to measure user‑growth and business performance. Key metrics include DNU, CAC, ROI, DAU/MAU, DAOT, ARPU, ARPPU, PUR, LTV, and K‑factor.

Comparison Thinking

When projects lack explicit goals, we uncover problems and opportunities by comparing data. Five comparative methods—Similarity, Difference, Commonality, Covariance, and Exclusion—help infer causal relationships from data patterns.

Structure Thinking

To ensure solutions comprehensively address identified problems, we use top‑down decomposition or bottom‑up aggregation. The three‑step process includes collecting information, categorising projects, and summarising each category with a concise title.