Skip to main content

Create a Joiner configuration

A Joiner configuration defines how two datasets should be matched and what the merged output should contain.

The most important part of a Joiner setup is not the button sequence. It is deciding what relationship the two files are supposed to have.

Before you begin​

Before creating the configuration, write down:

  1. The main dataset - Which file drives the output?
  2. The secondary dataset - Which file provides additional fields or comparison records?
  3. The match keys - Which Fields should connect the two Files?
  4. The output Fields - Which Fields should appear in the result?
  5. The unmatched behavior - What should happen when a record does not match?

If any of these are unclear, pause and confirm the business rule first.

Build the configuration in this order:

  1. define the purpose of the join
  2. select the two datasets
  3. identify the key Fields
  4. choose the Fields to include in the output
  5. decide how to handle unmatched Records
  6. test with known matched and unmatched examples
  7. save with a clear name

Step 1: Define the business goal​

State the goal in plain language.

Examples:

  • Add product categories to the monthly sales file.
  • Attach account owner details to open customer cases.
  • Match vendor payments against the approved vendor list.
  • Identify transactions that do not have a known customer account.

A vague goal like "merge these Files" usually hides important choices about keys, Fields, and unmatched Records.

Step 2: Choose the datasets and their roles​

Decide which dataset is the primary input and which dataset is the secondary lookup or comparison input.

The primary dataset is usually the File whose Records you want to preserve, such as:

  • transactions
  • cases
  • orders
  • employees
  • accounts

The secondary dataset usually provides context, such as:

  • account details
  • product attributes
  • approved reference values
  • location or owner metadata

Step 3: Select reliable match keys​

Choose the Fields that should identify the same business entity in both Files.

Common match keys include:

  • customer ID
  • account number
  • product code
  • employee ID
  • vendor ID
  • location code

Use the fewest keys that correctly express the relationship. If one key is not enough, use a clear combination of fields.

Step 4: Choose output fields intentionally​

Include the Fields needed for the next step, but avoid carrying unnecessary Fields into the result.

Ask:

  • Which primary Fields must remain?
  • Which secondary Fields add useful context?
  • Are any similarly named Fields likely to confuse reviewers?
  • Should temporary helper Fields be excluded from the final output?

Step 5: Plan for unmatched Records​

Unmatched Records are often the most important part of a join review.

Before saving, decide:

  • whether unmatched primary Records should remain in the output
  • whether unmatched secondary Records should be reviewed
  • which Fields will help explain why a match failed
  • whether match issues should be fixed by cleaning, deduplicating, or correcting source data

Step 6: Test with representative examples​

Use a sample that includes:

  • Records that should match cleanly
  • Records that should not match
  • keys with blanks or unusual formatting
  • cases where multiple Records might share the same key

Testing should prove both the successful matches and the exception handling.

Naming guidance​

Name the configuration by business purpose.

Better examples:

  • Add Product Category to Monthly Sales
  • Attach Account Owners to Open Cases
  • Match Payments to Approved Vendors

Weaker examples:

  • Join 1
  • Lookup Config
  • Merge Test

Save checklist​

Before saving, confirm:

  • the two dataset roles are clear
  • the match keys are correct
  • the output Fields are intentional
  • unmatched Records have a review path
  • the setup has been tested with realistic examples

Next step​

After the Configuration is ready, continue to Run Joiner.