How an Online Appointment Redesign Boosted Conversion for a Local Services Platform

Business Background

In the 58.com local services platform, the core capability is to solve information asymmetry between service providers and demanders, mainly through phone calls and messages. NPS monitoring showed that the offline service provision stage has the greatest impact on C‑end user experience, while the existing connection methods cannot capture merchants’ real fulfillment status.

New Connection Form

The platform cannot directly manage merchants like some competitors, so it introduced an online order connection: merchants provide quotes, users pay a 30 CNY reservation fee, and the remaining amount is paid after service completion. This aims to improve price transparency and service quality.

Design Directions

Data analysis identified two main problems: (1) conflict between the new online reservation and the existing “call” touchpoint, which dominates the visual hierarchy; (2) user friction in the reservation flow revealed by A/B testing.

The design directions are: (1) “Pre‑connection – Build Awareness” – create a dedicated page to introduce online reservation; (2) “In‑connection – Reduce Friction” – optimise the order page.

Design Solutions and Validation

Pre‑connection – Build Awareness

Replace the “call” label with “online reservation”, shift header information from post content to merchant details, and change the middle section from service info to product categories. Add merchant certification badges, service volume, store age, rating, and guarantee information to increase trust.

Show real‑time and historical online reservation data to reduce uncertainty. After launch, conversion rate (page view → paid order) increased significantly.

In‑connection – Reduce Friction

Analysis of A/B tests revealed two costly fields (“cleaning area” and “service personnel”). The solution was to remove them, which raised conversion dramatically.

Another test showed that adding a “select service” step lowered conversion due to long navigation and duplicate options. The UI was changed to a popup selector and merged similar items, which slightly reduced conversion but improved order completion and reduced cancellation rates.

Summary and Planning

The online reservation and two‑stage payment model is an early‑stage experiment. Continuous data‑driven iteration is required to balance speed and accuracy. Future work will optimise key nodes for users, merchants, and the platform to lift both conversion and order‑completion rates.