Skip to main content

Data Validation rules

Automated data validation at the metric level, enabling fund managers and company admins to set custom rules that flag anomalies based on year-on-year variance, metric comparisons, or defined thresholds.

F
Written by Femke Hummert
Updated today

What are data validation rules?

Data validation rules are automated checks that monitor the quality and consistency of your data. They are set by fund managers or company admins and automatically trigger (fail) when data crosses a defined threshold.

Instead of manually reviewing data for unusual changes or inconsistencies, these rules run continuously in the background and proactively flag potential issues. This helps ensure data accuracy, improve oversight, and create a clear audit trail of validation checks within the platform.

Please note: data validation rules only work for numerical metrics.

What rules you can set

There are three types of data validation rules available:

  • Year-on-Year Variance
    Triggers if a metric’s year-on-year change exceeds a specified percentage threshold.

  • Metric vs. Metric
    Triggers if one metric crosses a defined threshold in relation to another metric.

  • Metric vs. Threshold
    Triggers if a metric exceeds (or falls below) a custom value set by the user.

You can apply a rule to one or multiple metrics at once. When multiple metrics are included, the rule follows an “OR” logic. It will fail if any of the selected metrics breach the threshold.

How to set up data validation rules

  1. Log into your account.

  2. Navigate to the ‘Validation Rules’ tab in the left-hand menu under ‘Setup’

  3. Click “Add Rule.”

  4. Configure the rule by selecting:

  5. Rule type (Year-on-Year, Metric vs. Metric, or Metric vs. Threshold)

A) Rule name

B) Target metric(s)

C) Threshold value

6. Save the rule.

Once created, the rule runs automatically in the background and will trigger as soon as the defined threshold is crossed.

Users can also select any of the validation rules from our validation library:

These can be a good starting point which can be edited to match your validation rules.

Fund manager rules automatically flow down to portfolio companies.

Company admin rules apply only at the company level and do not flow back up to fund managers.

How to monitor your data validation

To monitor validation results:

  1. Go to the “Data Validation” tab under the Data section.

2. View the status of all rules, clearly marked as:

  • Passing

  • Failing

  • Incomplete Data (not enough data to assess)

  • Not Run (no data available yet)

  • Mixed (when multiple metrics are included and have different statuses)

  • Suppressed

You can click on the ‘view status key’ section to view the definitions in more detail

3. Expand any rule to see detailed information about why it is passing or failing.

Fund managers can view rule status by portfolio company.

Company admins see a breakdown by rule within their own company.

This visibility ensures issues are identified early and resolved before data submission, improving efficiency, oversight, and audit readiness.

Admins can also manually suppress a rule, which means it will no longer be evaluated for the current year.

Boundaries

Insufficient data

In situations where the platform only has partial data (e.g. for year-on-year variance, we don’t have all of this year’s data), the platform will say ‘insufficient data’, as it can’t run the rule until all the data is there to know whether it’s a pass or fail

Not run

In situations where the platform does not have any data (For example: user hasn’t input, or the collection hasn’t started) the platform says ‘not run’

Multiple metrics in a rule

When setting up the rule, it is possible to set the rule for multiple metrics at once.

This uses an ‘OR’ model, where the overall rule will fail if ANY of the metrics have increased or decreased by more than the threshold

If none of them have, then the overall rule will pass, unless any of the metrics (for that rule) are in a state of ‘insufficient data’ or ‘not run’, in which case the overall rule will show as having a “mixed” status

Did this answer your question?