Work2024蓝湖网络Expense & travel

Expense & travel assistant: travel cost -23%

Trip request, booking, expense and reconciliation handed to an AI assistant — zero receipt pasting for employees.

Internet费控商旅
Expense & travel assistant: travel cost -23%
Challenge

Expense & travel

Internet · 蓝湖网络 · 2024

Blue Lake Internet has 5,000+ employees, averaging 12K business trips per month. The integration of 4 OA systems, 3 travel platforms and 2 payment rails is reconciled manually, with average reimbursement cycles of 14 days.

Month-end close takes 15 days, overspend happens too often, and employees complain that pasting receipts is worse than writing code. We needed an AI assistant that fully automates request → booking → expense → reconciliation.

Path to production

From connect to ship, in four steps

01

Connect

In 2 weeks we adapt 4 OAs, 3 travel platforms and 2 payment rails through API / MCP, and inventory every governable expense rule.

02

Govern

In 3 weeks we ship the budget gate, expense rule engine, auto-approval flow, and sensitive-field masking / anomaly detection.

03

Deploy

In 3 weeks the Lark / WeCom entry is live; we pilot in 3 departments first, syncing employee + finance feedback biweekly.

04

Scale

In 8 weeks we roll out to the whole company, retune monthly, and compress month-end close from 15 days to 2.

Full project record
01

AI handles trip request, booking, expense and reconciliation — month-end close from 15d to 2d.Full case study delivered and reviewed by the Ouryun team.

Measurable outcomes

Numbers that prove value

0d
Month-end close (from 15 days)
0%
Travel cost reduction
0
Monthly auto-processed trips
0%
Employee reimbursement NPS
Expense & travel

Bring this capability into your business

Key capabilitiesAgent workflowAI Gateway unified channelsBudget gate + rule engineModel platform
OURYUN · Connect · Govern · Deploy · Ship ·