Now that you've set up the files, datasets, match groups, and match group dataset filters, the next step is to create the rules. You can create rules by accessing the Match Rules form.
Use the Match Rules form to create match rules:
- Select Explorer in the navigation pane.
- Expand Fluence > Transaction Matching > Maintenance folders, and open the Match Rules form.
Note: Alternatively, open this form by selecting the Rules link from the Match Group form. The Match Group parameter is automatically passed, and the form opens with the corresponding rules for that match group.
To create match rules, fill out the fields in the form:
| Field | Description |
| Name | Name of your match rule. This field is required. |
| Description | Short description for your match rule. This field is optional. |
| Dataset1 | List of existing datasets, already associated with the match group. |
| Dataset2 | List of existing datasets, already associated with the match group. |
| Rule Type | The rule types are One to One and Many to Many. |
| Confidence | Confidence level shows how accurate the match is when it is generated by the rule:
|
| Sequence | The order in which matching rules are applied. The value must be numeric and distinctive to each match rule. |
| Active | Indicates if the rule is active or not. |
| Details | Link to open the Match Rule Details form. Use this link to set up matching rules and control how transactions on two datasets can be compared and matched. |
| Duplicate | Link to duplicate a match rule. Duplicating rules is useful when you need similar rules with small changes and when you need to change small details across many rules. |
- Select Add Row from the Edit menu to add a new row to the Active Match Rules list.
- Name your match rule.
- Add a short description.
- Choose a dataset for Dataset1 and Dataset2 from the predefined list of datasets.
Ensure Dataset1 and Dataset2 have different positions, with one left and the other right, unless it's an exact match. For example, you can create Rule 1 to match from left to right and Rule 2 to match from right to left. If the transactions match exactly, then match the transactions within the same dataset. For example, Rule 3 can match from left to left or from right to right. This flexibility is powerful and doesn't need extra match groups. View the position of the datasets in the Match Group Datasets form. - Define the rule type. There are two types of rules:
- One-to-one rules enforce a strict rule comparing one transaction from Dataset1 to another from Dataset2. For example, comparing a specific check number of a certain amount within a certain time.
- Many-to-many rules compare multiple transactions from Dataset1 to multiple transactions from Dataset2. This encompasses one-to-many, many-to-one, and many-to-many scenarios. For example, comparing multiple credit card transactions to a single deposit amount within a certain time.
- Set a confidence level for the rule.
Define the rule type before setting the rule's confidence level. Confidence levels can be adjusted to High, Medium, or Low based on business needs. One-to-one rules are often set to high confidence, but if there are variances, a lower confidence level may be appropriate. - Set a sequence value for the rule.
This is critical for running rules in a specific order. You might have multiple iterations of the same rule with slight variations, and they need to run in a precise sequence. Rules with lower values are applied first, followed by those with higher values. Sequence numbers such as 10, 20, 30, etc., are recommended so you can add new rules afterward without reordering everything. - Determine if the rule is active or not, for example, Y for active, and N for inactive.
Active rules are processed during matching, while inactive rules are moved to the inactive section and won’t be part of the matching process. You may set a rule to inactive after it has created matches (for example, 10 million matches), as you can’t modify the details of a rule once matches have been created. The rule itself serves as the evidentiary trail of the logic used to create matches. Even after deactivation, the historic logic remains available for review. - After creating the rules, select the Details link for a many-to-many rule. It has all fields, including Grouping which isn't present for a one-to-one rule.
If you attempt to process the matching rules without adding details, you will receive an error. - Set filters in the Filters field. These filters only apply after the match group filters.
Many-to-many rules require filters, groupings, and matching criteria. The filters are level 2, meaning they apply after the match group filter. For example, for a rule to check deposits, you might set a filter for the Bank dataset to include transactions where the Field is Type_Code_Lookup, Operation is contains, and Value is CheckDeposit. Positive and negative tolerances can be set if it's a date or amount field. You can add additional filters, such as filtering by currency, transaction type, or other fields. - After filtering, define the Grouping.
For example, group all Bank transactions by Trans_Date and all General Ledger transactions by Transaction_ID. You must group both datasets, and the groupings should be consistent on both sides. In some cases, you can group multiple transactions on one side and only one transaction on the other. - Once the grouping is set, define the Matching Criteria.
You must compare amounts, dates, or other relevant fields. First, select the dataset and field (for example, Amount), then choose an operator (for example, =). You can compare absolute values (for example, credits and debits) and apply positive or negative tolerances, such as ±$10. You can also compare Date fields by selecting operators like <= or >=. This allows for flexible date comparisons between datasets. - After setting up the rule details, select Save Data.
Note: You can modify the parameters and add additional match rule details as needed. If you receive an error when processing matching rules, it may be due to missing date fields in the grouping section. Ensure that both Date fields from the two datasets are included in the grouping section. Keep in mind that groupings must be consistent on both sides to avoid errors during processing.
After saving the changes, you must run the matching rules for them to take effect:
- Select Process Matching Rules, then
- Create a Calculation task in an existing workflow or a new one, save, and run the process.
Refer to add match rule details section for more information on Filters, Grouping, and Matching Criteria.
