Invoice processing · Production

In2 Project Management automates invoice validation for a water corporation using Nanonets

The problem

The client, a water corporation spending circa $10 million AUD per year on external maintenance, needed to audit invoice accuracy across 72 invoice formats and 3500 line items per month — a scale that exceeded what manual spreadsheet review could handle, leaving discrepancies undetected.

First attempt

The team's manual approach of collecting invoices and building spreadsheets could not scale to the volume and variety of invoice formats, leaving verification gaps and discrepancies unaddressed.

Workflow diagram · grounded in source
1
Invoice intake from multiple sources
Trigger
Invoices are received from SAP Ariba System, third-party supplier invoices, and Sharepoint.
▾ source quote
“receives invoices from SAP Ariba System, 3rd Party supplier invoices and Sharepoint”
2
Real-time invoice data extraction
Ai action
Nanonets Pre-trained Invoice Extractor captures invoice information in real-time.
▾ source quote
“Nanonets Pre-trained Invoice Extractor captured information in real-time”
3
Route to SQL database
Integration
Extracted information is routed to In2 Project Management's SQL Database.
▾ source quote
“routed this information to In2 Project Management's SQL Database”
4
Validation against spend management system
Validation
The extracted information is validated against the client's spend management system.
▾ source quote
“The extracted information was validated against client's spend management system”
5
Live ABN company validation
Validation
Supplier ABN numbers extracted by Nanonets are compared live using the tax office ABN lookup web services.
▾ source quote
“the supplier ABN number of the company extracted by Nanonets Pre-trained Invoice Extractor was compared live using the tax office ABN lookup web services”
6
Power BI visualisation output
Output
The updated database is plugged into Power BI to produce insightful visualisations.
▾ source quote
“The updated database was plugged into Power BI, which brought insightful visualisations alive”
Reported outcome

Nanonets enabled identification of a $30k discrepancy between SAP Ariba portal expenses and extracted invoice data, surfaced cost inconsistencies across suppliers for the same equipment, and validated supplier ABN numbers — described as fundamental in saving cost.

Reported metrics
discrepancy identified in SAP Ariba vs extracted invoice data$30k
Annual client maintenance spendcirca $10 million AUD / year
Cost saving impactvery fundamental in saving cost
Reported stack
NanonetsNanonets Pre-trained Invoice ExtractorABN lookup web servicesSAP AribaSharepointSQL DatabasePower BI
◆ Does this fit your context?

Compare to your context

Tell us your scale, team, and constraints. We'll show what changes at your size, what fails at your scale, and whether this case is a fit, needs adaptation, or won't scale to you. Free demo, no signup.

Compare to your context →
~30 seconds · free
Source
https://nanonets.com/customer-success-story/in2-project-management-automates-invoice-processing-with-nanonets
Read source ↗

Frequently asked questions

What did this team achieve with this AI workflow?

Nanonets enabled identification of a $30k discrepancy between SAP Ariba portal expenses and extracted invoice data, surfaced cost inconsistencies across suppliers for the same equipment, and validated supplier ABN num…

What tools did this team use?

Nanonets, Nanonets Pre-trained Invoice Extractor, ABN lookup web services, SAP Ariba, Sharepoint, SQL Database, Power BI.

What results were reported?

discrepancy identified in SAP Ariba vs extracted invoice data: $30k; Annual client maintenance spend: circa $10 million AUD / year; Cost saving impact: very fundamental in saving cost (source-reported, not independently verified).

What failed first in this deployment?

The team's manual approach of collecting invoices and building spreadsheets could not scale to the volume and variety of invoice formats, leaving verification gaps and discrepancies unaddressed.

How is this invoice processing AI workflow structured?

Invoice intake from multiple sources → Real-time invoice data extraction → Route to SQL database → Validation against spend management system → Live ABN company validation → Power BI visualisation output.

Related invoice processing cases

WHAT TO DO WITH THIS

Now compare it to your context

This case is one data point. Whether its pattern fits you depends on your volumes, your stack, and your exception load — that comparison is the step no case study can do for you.

Browse the Invoice processing cases →