This IT project is part of our digitalization and optimization of our customers’ IT landscape. Through targeted measures, we promote technological progress, optimize cross-system processes and create a sustainable basis for future developments. Our IT reference projects serve as a basis for orientation. They support the reusability of tried and tested concepts as part of project implementation.
This is one of 4822 IT projects that we have successfully completed with our customers.How can we support you?
AI-based account assignment suggestions for incoming invoices
Project duration: 4 months
Brief description
The goal of the project is to conduct a feasibility study for the development of a proprietary AI-based solution for the automated generation of account assignment suggestions for incoming invoices. The client benefits from a significant reduction in manual verification tasks and consistent, traceable account assignments. PTA identifies relevant data sources, prepares training data, and extracts accounting-relevant information from OCR texts. Using feature engineering, the selection of suitable machine learning (ML) algorithms, and hyperparameter tuning, PTA develops robust models for the automated generation of accounting proposals.
Supplement
PTA provides training data via SQL and extracts not only invoice metadata relevant for account assignment but also textual information using regular expressions and AI-powered text analysis (Snowflake Cortex AI). To this end, the invoice data-including OCR text-is transferred to Snowflake via CSV or text import. Data modeling and structured feature engineering are then performed based on this. Subsequently, PTA designs, trains, and evaluates various machine learning models using suitable Python libraries such as Scikit-Learn, XGBoost, CatBoost, and LightGBM, and optimizes them through hyperparameter tuning. Based on the presented results and the finding that the in-house development delivers account assignment suggestions of equivalent quality and, in some cases, even better ones, a cost estimate is finally provided for the technical implementation and integration of the solution into the existing ecosystem of OCR software and document management systems.
Subject description
The product automates the posting of incoming invoices by generating suggestions for the general ledger account, cost center, and tax code. It helps departments and invoice reviewers minimize repetitive manual tasks and make posting decisions that are consistent, fast, and traceable. Ultimately, this enables invoices to be reviewed, approved, and posted more quickly.
IT project data
- Industries Logistics
- Specialist tasks Business Analytics, Data Cleansing, Data Mining, machine learning, make-or-buy analysis, Predictive Analytics
- DP tasks IT consulting
- System environments Artificial Intelligence (AI), Business Intelligence, Cloud Computing
- Procedure model customer standard
- Programming languages Python, SQL
- Databases Snowflake
- Standard software and BI Tableau Desktop
Project period01.11.2025 - 28.02.2026
Customers who trust us
Have we sparked your interest ?
