# Duplicate Business Partner Detection Report (Name, Tax Number, Address)

> A custom ALV report that finds cases where the same customer or vendor was created under more than one business partner number, using the tax number, a name comparison that normalizes Turkish characters and legal form suffixes, the address and the phone number. Likely duplicates are grouped with a match score and reason, so the master data team can quickly decide which record to keep.

- Type: Report, Custom development, ALV report
- Module: CA
- Estimated effort: M (5–7 working days)
- Last updated: 2026-11-17
- URL: https://nekuhub.com/en/reports/duplicate-business-partner-detection-report/


## Why is the standard not enough?

The duplicate check in transaction BP only runs when a new record is created and needs the address search service to be set up separately; it does not scan thousands of existing records and list the duplicates. Spelling differences such as "Atlas Ambalaj A.Ş." and "ATLAS AMBALAJ SANAYİ TİC. AŞ" stay separate records in the standard search. The custom report scans the whole master data, compares normalized names and returns duplicate groups with a score.

### Closest standard alternatives

- BP: Can check for duplicates when a new record is created (needs the address search service); it does not scan existing records in bulk.

## Tables used

| Table | Role | Access | Join | System |
|---|---|---|---|---|
| BUT000 | Business partner general data: type, grouping, name, creation (driving table) | Read | - | ECC + S/4 |
| DFKKBPTAXNUM | Tax numbers; the strongest match criterion | Read | DFKKBPTAXNUM-PARTNER = BUT000-PARTNER | ECC + S/4 |
| BUT020 | Address number of the business partner | Read | BUT020-PARTNER = BUT000-PARTNER | ECC + S/4 |
| ADRC | Postal code, city, street and phone | Read | ADRC-ADDRNUMBER = BUT020-ADDRNUMBER | ECC + S/4 |
| BUT100 | Business partner roles (customer, vendor) | Read | BUT100-PARTNER = BUT000-PARTNER | ECC + S/4 |

## Selection screen parameters

| Parameter | Description | Type | Reference field | Required |
|---|---|---|---|---|
| S_PARTNR | Business partner | Range / multiple | BUT000-PARTNER | No |
| S_BUGRP | Grouping | Range / multiple | BUT000-BU_GROUP | No |
| S_TYPE | Partner category (1 person, 2 organization) | Range / multiple | BUT000-TYPE | No |
| S_CRDAT | Created on | Range / multiple | BUT000-CRDAT | No |
| S_CNTRY | Country | Range / multiple | ADRC-COUNTRY | No |
| S_RLTYP | Role | Range / multiple | BUT100-RLTYP | No |
| P_SCORE | Minimum match score | Single value | - | Yes |
| P_TAXON | Tax number matches only | Checkbox | - | No |

## Output columns

| Column | Source |
|---|---|
| Duplicate group | Calculated: Group number given to records that match each other |
| Business partner | BUT000-PARTNER |
| Name | BUT000-NAME_ORG1 |
| Grouping | BUT000-BU_GROUP |
| Tax number | DFKKBPTAXNUM-TAXNUM |
| City | ADRC-CITY1 |
| Postal code | ADRC-POST_CODE1 |
| Street | ADRC-STREET |
| Phone | ADRC-TEL_NUMBER |
| Roles | Calculated: List of BUT100-RLTYP values |
| Created on | BUT000-CRDAT |
| Created by | BUT000-CRUSR |
| Match score | Calculated: Tax no 100, name 60, similar name 40, address 30, phone 20 (best pair) |
| Match reason | Calculated: Criteria that make up the score |

## How does the report work?

1. Business partners in the selection are read from BUT000 (type, grouping, creation date); records flagged for archiving (XDELE) are skipped.
2. Name normalization: NAME_ORG1 (for persons NAME_FIRST + NAME_LAST) is converted to upper case, Turkish characters are folded (İ→I, Ş→S, Ğ→G, Ü→U, Ö→O, Ç→C), legal form suffixes (A.Ş., AŞ, LTD. ŞTİ., SAN. TİC. and so on), punctuation and spaces are removed.
3. Tax numbers are read from DFKKBPTAXNUM; the same tax type and number is a strong match.
4. The standard address is read from ADRC through BUT020: postal code, city, normalized street and phone.
5. Score: same tax number 100, same normalized name 60, name similarity above the threshold 40, same postal code and street 30, same phone 20. Pairs above the score threshold are merged into duplicate groups.
6. Roles (customer, vendor and so on) are added from BUT100, so it can be checked which record has open items or documents. The result is listed per group in ALV.

## Authorization objects

- B_BUPA_GRP: Display authorization by business partner grouping
- B_BUPA_RLT: Display authorization by business partner role

## Frequently asked questions

### Does the report merge duplicate records?

No, it only lists them. The master data team decides which record to keep; the other one is blocked or flagged for archiving. For records with open items, accounting has to clean up first.

### Does this report work on ECC too?

Yes, on ECC systems that use business partners it works with the same tables. For ECC systems that still keep customer and vendor master data in KNA1/LFA1, the same logic can be adapted to those tables; this is often requested as a clean-up before an S/4HANA conversion.

### How is name similarity measured?

The edit distance (Levenshtein) between the normalized names is calculated and converted into a percentage similarity based on length. The default threshold is 85% and can be changed on the selection screen.

---
Let us build this report for your system: https://nekuhub.com/en/request-a-quote/?report=ca-duplicate-business-partner-detection
