AI agents that audit receipts, re-shop fares and pick the cheapest dates

The agent reads every document sent to Expense — photo, PDF or invoice — extracts amount, date, tax ID and category, then compares it with the company travel policy. Instead of sampling, AI reviews every receipt and marks each expense as in policy, out of policy or pending justification.

• Automatic OCR + AI reading of receipts and invoices in seconds
• Limit checks by category, city, job level and cost center
• Duplicate, altered receipt, mismatched amount and trip-date conflict detection
• Justification requested from the traveler before the manager approves
• Full audit trail of what the AI analysed, ready for finance and auditors

Airfares and hotel rates move constantly. Loupit's robot monitors issued bookings and, when the price drops for the same route, date and standard, it flags the opportunity and shows the net saving — change fees and fare difference included, so you only swap when it truly pays off.

• Continuous monitoring of the same itinerary after ticketing
• Net math: new fare + taxes + change fee versus what was already paid
• Alert to traveler and manager with one-click online rebooking
• Works with online rebooking, no ticket to the agency required
• Captured savings logged for the CFO's report

Before booking, the fare calendar displays daily prices and highlights the cheapest dates in the period. AI combines price history, lead time and demand patterns to suggest when to fly whenever the agenda allows one or two days of flexibility.

• Monthly price view with the lowest fare highlighted
• Suggestion to shift departure or return by 1–2 days, with estimated savings
• Ideal booking lead time for the company's busiest routes
• Policy-aware: only approvable options are suggested

Every client has its own profile: recurring routes, preferred suppliers, limits per job level, tolerance for connections, hotels already used. The agent learns that profile and ranks marketplace offers accordingly — without hiding other options or bending policy.

• Offer ranking tuned to history and company rules
• Pre-filled preferences, documents and cost center
• Flags choices that deviate from the profile's usual pattern
• Policy recommendations based on what the company actually spends