Anaplan does have some limitations in regards to data validation. Specifically, it cannot prevent a missing value. In addition, it can't prevent many incorrect values.
This said, there are some things we can do to flag bad (or missing) data, as well as prevent that data from being shared in downstream models or applications.
First, we can use Conditional Formatting to highlight where values are either invalid or missing. In most cases, the conditional formatting rule is tied to a 2nd (often hidden) line item that contains a formula that evaulates to some specific numbers that help control the related conditional formatting colors.
Second, where data is being exported to another model (or application), we can apply filters that remove the bad data from the export, so that only validated data gets exported.
It goes without saying that we can create (what I refer to as) Audit Dashboards that allow us to quickly see our key data filtered to only show invalid (or missing) values. This way, we can fix things without having to search too long.