Share this on:
What You'll Learn
Implementing a match-merge process is a part of master data management. This process involves taking data not only from different source systems but also from within the same source. The process helps find possible duplicates, or identical matches (and merge as required) and create a golden copy of the record.
Two Match Process Strategies in Informatica SaaS MDM
- Exact Match: A deterministic match compares records with identical values in the match columns.
- Fuzzy Match: A probabilistic match that includes variations in data patterns such as misspellings, transpositions, the combining or splitting of words, omissions, truncation, and phonetic variations.
Based on the Outcome of the Match and the Merge Configuration, the Records are Either:
- Automatically Merged: Use the automated merge strategy for declarative rules that are based on unique identifier fields, such as social security number or passport number.
- Skip: The records in the pairs are not matched against each other if their match scores fall within the range set for the merge threshold for skipping merge.
- Queued for Manual Merge: Use the manual merge strategy for declarative rules that are based on fields, such as the address field. You need to manually review the records to confirm whether these are duplicate records.
- Threshold Based: Available only when you select the fuzzy match strategy. Use the threshold-based merge strategy for any field. Based on the match score of the record pairs, the match and merge process determine whether the result is automated, manual, or skipped merge.
Survivorship Configuration
Define attribute/field specific source system ranking:
- Ranking Rule: The source ranking rule will act in case of the multiple records from different source system.
- Recency: Multiple records from the same source system the value from the source with the newer LUD will survive.
A record coming from Rank1 Source is NULL, whereas there is a valid record coming from Rank2 Source. Rank2 source record will survive though less trusted.
Validation Process
Points to check before validating the merge and survivorship process:
- Access to IICS
- Declarative Match-Merge Rules and Survivorship Order of Precedence should be defined in Business 360 Console
- A job to perform the match and merge process defined in Business 360 Console
- Data should be created/loaded in Customer 360 Console
How to Validate Merged Records and Survivorship Order of Precedence in Informatica MDM
Below are some of the basic scenarios based on survivorship rules:
- Records from two different sources are present in C360.
- Records from the same sources are present in C360.
- Records from 2 different sources with rank1 has null value for 1 attribute in C360.
Note: There will be multiple declarative rules based on the business requirement.
Sample of an Auto-Merged Record Validation
Consider scenario 2 for the sample below:
Sample Match Rule: Exact Customer First Name & DOB.
Sources: Customer from same source system.
Expected Result: Records with same Customer First name & DOB should get merged &survived based on the survivorship order.
Step 1
For validating the above rule, create sample records/load source data that satisfies exact match condition.
We have created sample records with the same customer first name and date of birth with the same source system.
Customer 1:
Customer 2:
Both are from the same source system: Informatica Customer 360
Step 2: Run the merge job
Step 3: Validate if the above customers got merged
After running the match and merge job, navigate to Customer 360 and check for the merged record in the MDM.
Both the customers got merged:
Step 4: Verify the survived record
Both customers got merged and Last name, Gender, and Marital status got survived from the latest record (SUMAN NIRU2) since both customers are from the same source system.
The same steps can be followed to validate any merge rules.
About LumenData
LumenData is a leading provider of Enterprise Data Management, Cloud, and Analytics solutions and helps businesses handle data silos, discover their potential, and prepare for end-to-end digital transformation. Founded in 2008, the company is headquartered in Santa Clara, California, with locations in India.
With 150+ Technical and Functional Consultants, LumenData forms strong client partnerships to drive high-quality outcomes. Their work across multiple industries and with prestigious clients like Versant Health, Boston Consulting Group, FDA, Department of Labor, Kroger, Nissan, Autodesk, Bayer, Bausch & Lomb, Citibank, Credit Suisse, Cummins, Gilead, HP, Nintendo, PC Connection, Starbucks, University of Colorado, Weight Watchers, KAO, HealthEdge, Amylyx, Brinks, Xylem, Clara Analytics, and Royal Caribbean Group, speaks to their capabilities.
For media inquiries, please contact: marketing@lumendata.com.
Authors
Sr. QA Lead
QA Lead
