celonis-ap-value-framing

Accounts Payable Value Framing Business Case

A Process Mining Portfolio Project | Celonis Methodology Applied to Real P2P Data

AP Value Tree


✅ Project Overview

This project applies Celonis’s 6-Step Value Framing Methodology to a real Purchase-to-Pay (P2P) event log dataset, producing an executive-ready business case for two high-priority Accounts Payable improvement opportunities:

  1. Payment Blocks & PO Blocks — manual intervention driving process delays
  2. Price Deviations (Change Price activity) — rework-heavy corrections adding cycle time

The analysis follows the Celonis Value Delivery Framework (Identify → Frame & Commit → Realize & Track) and is structured as a simulated consulting engagement — from raw data discovery through to executive sign-off.

Context: This project was built as a portfolio deliverable following the Celonis × AICTE Virtual Business Analyst Internship, during which I earned certifications in Process Mining Fundamentals, Business Value Delivery, and Celonis Foundations.


✅ Business Problem

Accounts Payable teams operating in SAP-based environments frequently face hidden process inefficiencies that are invisible in standard ERP reports:

Without process-level visibility, organizations cannot make data-backed decisions about where to invest in automation, process redesign, or vendor management.


✅ Repository Structure

ap-value-framing-business-case/
│
├── 📂 data/
│   ├── P2P-Cases.csv              
│   ├── P2P-Activities.csv         
│   └── data_dictionary.md         
│
├── 📂 analysis/
│   ├── p2p_data_profiling.py       
│   ├── deviation_analysis.py       
│   └── value_calculation.py        
│
├── 📂 presentation/
│   └── AP_Value_Framing_Business_Case.pptx   
│
├── 📂 assets/
│   ├── ap_value_tree.png           
│   ├── celonis_variant_explorer.png
│   ├── celonis_olap_table.png
│   ├── celonis_data_model.png
│   └── celonis_kpi_model.png
│
├── 📂 docs/
│   ├── methodology.md              
│   ├── assumptions_log.md          
│   └── kpi_definitions.md          
│
├── README.md
└── LICENSE

✅ Dataset Summary

Attribute Value
Source SAP-based P2P Training Dataset (Celonis Guided Learning)
Total Cases (PO Items) 812
Total Events 4,927
Unique Activities 19
Total Procurement Spend €207,096
Unique Vendors 20
Date Range 2016 – 2017
System SAP ERP (Tables: LFA1, EKKO, EKPO + Activity Log)

✅ Methodology: Celonis 6-Step Value Framing

This project implements all six steps as documented in Celonis’s Value Delivery Framework:

Step 1: Qualify    → Assess Strategic Fit, Business Impact & Feasibility
Step 2: Quantify   → Calculate Expected Monetary Value per opportunity
Step 3: Investigate → Root cause categorization (Process / Time / Attribute)
Step 4: Validate   → Align KPI definitions with business logic
Step 5: Prioritize → Impact-Effort Matrix placement
Step 6: Commit     → Executive sign-off documentation

✅ Key Findings

Process Health

| KPI | Value | |—–|——-| | Happy Path Cases | 537 (66.1%) | | Deviant Cases | 275 (33.9%) | | Total Deviation Events | 310 of 4,927 | | Automation Rate | 47.5% | | Manual Activity Rate | 52.5% | | Average Cycle Time | 29.6 days | | Maximum Cycle Time | 190 days |

Top Deviation Activities

| Activity | Event Count | Business Impact | |———-|————-|—————–| | Change Price | 110 | Price mismatches → manual rework | | Block Purchase Order Item | 38 | PO delays → extended cycle time | | Set Payment Block | 20 | Invoice holds → delayed clearance | | Change Currency | 20 | Currency mismatches → compliance risk | | Delete Purchase Order Item | 20 | Procurement waste |


✅ Business Case Summary

Opportunity 1: Resolve Payment & PO Blocks

Framing approach: Labor productivity savings from reducing manual block resolution

Parameter Value
Affected Events 58 (Set Payment Block + Block PO Item)
Effort per manual resolution 15 minutes
FTE Cost per minute $0.42 (based on $90K/year FTE)
Realization Potential 80%
Expected Monetary Value ~$293 annually (sample dataset)

Note: This dataset is a training simulation with 812 cases. In a real enterprise deployment, scaling this ratio to 279,000 cases (as seen in Celonis platform screenshots) would produce savings in the range of $1.5M–$2.2M p.a., consistent with Celonis’s published payment block benchmarks.

Opportunity 2: Eliminate Price Deviation Rework

Framing approach: Reduce manual effort from 110 Change Price correction events

Parameter Value
Affected Events 110
Effort per price correction 10 minutes
FTE Cost per minute $0.42
Realization Potential 70%
Expected Monetary Value ~$323 annually (sample dataset)

✅ Platform Screenshots

Celonis Variant Explorer — Process Flow Analysis

Variant Explorer

The most common variant (39% of 279k cases, 26-day cycle) follows the expected happy path: Process Start → Create PR → Create PO → Print & Send PO → Receive Goods → Scan Invoice → Book Invoice → Process End.


Celonis OLAP Table — PO Items by Vendor

OLAP Table

Vendor spend analysis reveals high concentration: Piccolo GmbH (€85.9M), C.E.B. BARCELONA (€259M), PAQ Deutschland GmbH (€9.6M) represent key strategic supplier relationships warranting close monitoring.


Celonis Data Model — SAP Table Relationships

Data Model

The P2P data model connects LFA1 (vendor master) → EKKO (PO header) → EKPO (PO items, case table) → _CEL_P2P_ACTIVITIES (activity log), joined on LIFNR/MANDT/EBELN keys.

### Analysis Charts ### Deviation Pareto Analysis

### Vendor Risk Matrix


✅ Tools & Technologies

Tool Purpose
Celonis EMS Process mining platform — variant analysis, KPI calculation, OLAP tables
Python (pandas) Data profiling, deviation classification, value calculation
PowerPoint Executive business case presentation
SAP ERP Source system (LFA1, EKKO, EKPO tables)

✅ How to Use This Repository

1. View the Business Case Presentation

Open presentation/AP_Value_Framing_Business_Case.pptx — a 10-slide executive deck following the Problem → Process → Analysis → Insights → Value → Recommendations storytelling arc.

2. Explore the Data

# Clone the repository
# Correct
git clone https://github.com/komalharshita/celonis-ap-value-framing.git
cd celonis-ap-value-framing

# Install dependencies
pip install -r requirements.txt

# Run data profiling
python analysis/p2p_data_profiling.py

# Run deviation analysis
python analysis/deviation_analysis.py

# View value calculations
python analysis/value_calculation.py

📧 Contact

Komal Harshita CSE student

LinkedIn GitHub


This project was developed as part of the Celonis × AICTE Virtual Business Analyst Internship (2026). All data used is from Celonis’s Guided Learning environment and is used for educational portfolio purposes.