Standardizing Registration & Login Prompt Styles for Better UX

Why Prompt Design Matters

Prompts act as a guide for users when they start a service, similar to a key for registration and a door for login. If the experience of “unlocking” and “entering” is disrupted, users may abandon the product.

Problems Identified

1. Misuse of Prompt Styles – Every error on the registration page is shown as a dialog box. Dialogs are blocking feedback meant for high‑risk or severe errors; overusing them interrupts the flow and discourages users.

2. Inaccurate Meaning – Some prompts only point out the error without guiding the user. For example, after multiple captcha failures the message “Too many captcha errors, please try later” does not clearly explain the waiting time, leading to confusion.

3. Inconsistent Styles for Same Type – Similar input errors are sometimes shown as red‑text inline hints and other times as toast notifications, affecting perceived importance and product uniformity.

Deep Dive: Front‑end vs. Back‑end Prompts

Prompts can be divided into two categories:

Front‑end prompts : Validate input format or other checks that do not require server data. Errors are displayed directly in the UI.

Back‑end prompts : Require server verification (e.g., password mismatch). The server returns a code, which the front‑end maps to a specific message and style.

This separation often leads to each side creating its own style, causing inconsistency.

Creating a Unified Prompt Style Guide

All registration and login prompt texts were collected. For each message, a scenario description and trigger timing were added, allowing classification into types (input, verification, system). This mapping determines the appropriate style—for example, the message “This account is locked, please self‑unlock” is a back‑end verification alert and should use a dialog.

Optimization Plan

The guide categorizes prompts into three groups: input errors, clause verification/system feedback, and operation indications. Each group includes style specifications and example cases.

An online self‑check list was built, listing each prompt issue, responsible owner, and progress, which reduced inefficient communication and duplicated work.

Before & After Comparison

Before optimization, most error prompts were dialogs, interrupting the flow. After applying the guide, prompts were changed to inline red‑text beneath input fields, lowering interruption while still informing the user.

Conclusion

Prompt style optimization should be evaluated on three dimensions: uniformity , accuracy , and effectiveness . Prompts must align with the product’s tone, be defined from real usage scenarios, and guide users efficiently. A well‑defined specification improves design quality, team collaboration, and overall user experience.