Invoice processing · Production

Lido automates vendor invoice processing at Soldier Field ASM Global

The problem

Soldier Field ASM Global handled 300–400 vendor invoices every week through a fully manual process—extracting data from PDFs, entering it into their ERP, and verifying accuracy—consuming approximately 20 hours per week and remaining error-prone due to vendor-specific invoice format variation.

First attempt

GPT models, Power Automate, and AI agents were all evaluated but none were customizable enough to handle vendor-specific invoice formats, requiring significant manual intervention despite the automation attempt.

Workflow diagram · grounded in source
1
Vendor invoices arrive post-event
trigger
“processing of invoices for numerous vendors after each event”
2
Lido extracts and categorizes PDF data
ai_action
“Working closely with Lido's team, Rohan refined the automation process to extract, categorize, and process invoice data efficiently”
3
Data entered into ERP system
integration
“entering data into their ERP system”
Reported outcome

Implementing Lido led to a drastic improvement in efficiency and accuracy at Soldier Field ASM Global; the specific outcome metrics are not available in the captured page text as the results section is truncated.

Reported metrics
Weekly invoice volume300 to 400 vendor invoices every week
weekly time spent on invoice processing before Lidoapproximately 20 hours per week
Efficiency and accuracy improvementdrastic improvement in efficiency and accuracy
Reported stack
LidoGPT modelsPower AutomateERP system
Source
https://trylido.com/customers/soldier-field
Read source ↗

Frequently asked questions

What did this team achieve with this AI workflow?

Implementing Lido led to a drastic improvement in efficiency and accuracy at Soldier Field ASM Global; the specific outcome metrics are not available in the captured page text as the results section is truncated.

What tools did this team use?

Lido, GPT models, Power Automate, ERP system.

What results were reported?

Weekly invoice volume: 300 to 400 vendor invoices every week; weekly time spent on invoice processing before Lido: approximately 20 hours per week; Efficiency and accuracy improvement: drastic improvement in efficiency and accuracy (source-reported, not independently verified).

What failed first in this deployment?

GPT models, Power Automate, and AI agents were all evaluated but none were customizable enough to handle vendor-specific invoice formats, requiring significant manual intervention despite the automation attempt.

How is this invoice processing AI workflow structured?

Vendor invoices arrive post-event → Lido extracts and categorizes PDF data → Data entered into ERP system.