1. Introduction

Our client operates a fast-growing automotive marketplace, acting as the digital bridge between over 20 car dealers and a large network of buyers and partners. Years earlier, we built a custom conditions database and internal UI for managing vehicle pricing and condition data.

As the marketplace scaled, the challenge wasn’t the platform; it was the volume and inconsistency of dealer data. Dealers sent updated conditions in different formats, structures, and layouts every week. What used to work became a major operational burden. The client needed a modern, automated system that could transform these multi-format dealer files into clean, structured data with minimal human effort.

Goal:

Convert a fully manual conditions workflow into an automated, AI-assisted system that anyone on the team can operate.

2. The Problem We Solved

Before automation, the weekly workflow looked like this:

• Dozens of files arrived from multiple dealers

• Formats varied: Excel, CSV, PDF, email attachments, custom exports

• Employees reformatted, cleaned, and aligned each file manually

• Data had to be compared with existing values

• Updates were entered into the conditions system by hand

• Everything needed verification to avoid pricing errors

This resulted in common issues:

• Time-consuming – weekly updates required hours or days of repetitive work  

• Error-prone – manual entry caused wrong prices, missing values, mismatched vehicle models  

• Inconsistent – every dealer used a different structure  

• Not scalable – adding new dealers multiplied the workload

The client needed a reliable way to process different file structures automatically.

3. The Solution: Automated Conditions Management With AI

We built a fully automated system that integrates seamlessly into the existing conditions platform.

a) Uploading dealer files: any format, any structure

Users can now upload dealer files through:

• Telegram

• Email

• Direct upload

No preparation is required. The system automatically detects the file type, reads the structure, extracts the data, and prepares it for comparison.

b) AI-assisted analysis and comparison

After upload, the system prompts the user to confirm the dealer and brand. The AI engine then:

• Compares uploaded conditions with existing records

• Identifies new or modified values

• Highlights pricing and condition changes

• Organizes them into a clean, structured list

c) A dashboard for reviewing changes

Users receive a dashboard link where they can:

• Review suggested updates

• Compare old vs. new values

• Validate pricing changes

• Approve new vehicle listings

This replaces cluttered spreadsheets with a clear, review-focused interface.

d) One-click approval

Once the reviewer approves the updates:

• The system writes changes directly into the conditions database

• No manual entry is required

• No formatting issues occur

This dramatically reduces weekly workload and eliminates inconsistencies.

If a value needs adjusting, the reviewer can edit it directly in the dashboard before applying changes. Over time, these corrections help improve system accuracy.

4. Tech Stack

This automation project was delivered using:

• n8n

• ReactJS

• .NET