Designing Effective Live‑Stream Reward Systems: A Practical Guide

—"As a post‑90s, have you never tipped?" —"(￣▽￣")" Motivated by this conversation, I examined tipping features across major live‑stream and knowledge‑payment apps to outline a design framework for CCtalk's reward system.

Knowledge Payment

Knowledge payment has grown with the convenience of online payments, turning content into a commodity similar to e‑commerce. Platforms such as WeChat public accounts, Zhihu's paid articles, and Ximalaya subscriptions enable users—primarily post‑90s and middle‑class 80s—to directly support creators.

Definition of Tipping

Tipping is a voluntary monetary or virtual‑gift reward given by users to original online content (articles, images, videos, etc.). It is a non‑mandatory payment model first seen on platforms like Qidian and Weibo, later adopted by live‑stream services.

Value of Tipping

Tipping increases creators' income and platform revenue. Compared with likes, it reinforces user recognition of both content and creator, motivating higher‑quality output. It also helps identify highly engaged, loyal users.

Products Suitable for Tipping

Content Value

Only content that users deem valuable will receive tips, typical in premium communities, niche forums, or expert‑led groups (e.g., Zhihu, paid WeChat groups). Pure entertainment without perceived value rarely attracts tips.

Personal Brand Influence

Products that highlight individual creators—such as livestream hosts or influencers—encourage tipping as a direct expression of fan support for the personal brand.

Why Users Tip

Users voluntarily pay when they feel a social contract, perceive the creator’s livelihood depends on tips, or experience a “helping psychology.”

Building an invisible social contract that makes users feel the creator needs tips to survive.

Live‑show or game‑stream platforms subtly plant this idea, leading users to view tipping as essential support for the performer.

In contrast, article‑based platforms rely more on recognition and fan acquisition; tipping is less ingrained and may feel intrusive if introduced abruptly.

Users' “helping psychology” – supporting weaker or similar individuals for a sense of purpose.

Tipping Analysis

1. Tipping Targets and Forms

Knowledge‑type platforms (Zhihu Live, Qianliao, WeChat public accounts, Ximalaya, Hujiang Q&A) primarily use cash tips directed at the content creator or live host. Show‑type livestream platforms (Douyu, Yingke) favor virtual‑gift props, also purchased with cash.

2. Tipping Props

Props are split into free (earned via tasks) and paid (purchased with cash). Free props boost user stickiness and introduce tipping habits; paid props generate revenue and encourage higher‑quality streams.

Paid props can be low‑price “combo” items that encourage repeated gifting (e.g., chainable gifts) or high‑price items with distinct visual effects, both influencing overall platform turnover.

3. Post‑Gift Display

After a gift is sent, three display methods are common: a system‑message record in the chat, a dedicated area showing current gift information (sender, gift type, quantity), and an animated effect for the gift itself. Knowledge‑type platforms often only use the system‑message, while show‑type platforms combine all three to enhance visibility.

CCtalk Live Reward Design Case

CCtalk, a knowledge‑focused livestream platform, naturally supports tipping. The design decisions include:

Reward form: virtual props (emotional expression) combined with cash payment for the transaction.

Reward target: the content creator.

Reward props were defined as follows:

Free prop: "Flower" (earned via tasks).

Paid props: "Throat lozenge" (1 CNY), "Ice cream" (8 CNY), "Coffee" (38 CNY), "Golden microphone" (88 CNY), all purchased via transfer.

The MVP implementation limited each tip to a single prop per transaction, with the free flower remaining available. Future iterations may expand prop variety and interaction patterns.

Reward Display Design

Initially, only a chat‑area system message is used to keep development effort low while testing market response. Known issues include the message being buried among chat, lack of visual distinction between free and paid props, and rapid disappearance as chat volume grows.

Two display concepts were explored:

Option 1: A fixed area shows current gift information, emphasizing the sender; high‑value gifts receive additional animation.

Option 2: Low‑value gifts appear in three probabilistic zones (core screen, non‑core screen, microphone queue) with configurable ratios (2 % / 20 % / 78 %). High‑value gifts remain in a fixed queue.

These designs aim to balance minimal implementation cost with the ability to iterate based on user feedback.