Mastering User Segmentation: Models, Strategies, and Design Tactics

Why Do We Need User Segmentation

With the decline of population dividends and the rise of both public and private traffic competition, businesses must extract and retain users efficiently; segmentation enables precise targeting, resource optimization, and maximized user value at minimal cost.

How to Perform User Segmentation

1. General User Segmentation Models

Several widely applicable models provide rigorous logic for dividing users.

RFM Model

R (Recency) – time since last purchase; F (Frequency) – purchase count in a period; M (Monetary) – amount spent in a period. By splitting each dimension into levels, up to eight basic user types emerge, which can be further refined for granular targeting.

AIPL Model

Awareness – new users; Interest – users who browse or add to cart but haven’t purchased; Purchase – users who have placed orders; Loyalty – repeat purchasers with high repurchase rates. The model reflects a user’s growth path from awareness to loyalty, guiding tailored interventions at each stage.

AARRR Model

Acquisition – attracting users; Activation – encouraging participation; Retention – reducing churn; Revenue – driving conversions; Referral – promoting sharing and repeat purchases. It maps the entire product lifecycle or a specific campaign funnel, helping identify growth opportunities.

2. Demographic Tag Segmentation

Beyond transaction‑based models, users can be clustered by attributes such as gender, age, region (basic attributes) and category preference, spending amount, purchase motivation (shopping attributes). This yields concrete personas like "Quality Guru" or "Fashion Diva," enabling highly personalized content.

3. Targeted Segmentation

For smaller or special‑purpose projects, segmentation can be aligned directly with business goals—for example, splitting users into new vs. existing for acquisition campaigns, or into tier‑1 vs. tier‑2 city users for market penetration.

User Segmentation Design Strategies

1. Traffic Allocation

After defining segments, backend algorithms identify users and tag them, while the frontend displays customized layouts. Edge cases such as anonymous users require default presentations. Segmentation should start at the traffic entry point, assigning distinct creative assets and allocating channel resources based on segment priority.

2. Product‑User Matching

Standard activity pages show identical content to all visitors; segmentation enables a thousand‑faces approach. By mapping each segment to specific content, navigation flows, and module styles, the experience becomes highly relevant. For a beverage promotion, the 4A model (Stranger, Interested, Drinker, Enthusiast) guides content selection, flow ordering, and visual styling.

Conclusion

User segmentation is a cornerstone of refined operations; as shopping experiences become more diverse, platforms increasingly rely on both algorithmic and operational segmentation to subtly influence behavior. Ongoing experimentation and knowledge accumulation will continue to drive more sophisticated and innovative segmentation practices.

References

1. RFM Model – Baidu Baike

2. AARRR Model – Baidu Baike

3. User Segmentation Models – UISDC

4. How to Implement User Segmentation for Fine‑Grained Operations?

5. Building Custom User Segmentation for Your Product