Cycle Time is a measure of how efficiently we are able to release value to customers. In combination with Deployment Frequency & Change Failure Rate, this is the best metric to understand how well you're able to deliver value to customers.
That said, Haystack always uses Aggregate Time for all it's calculations. In simpler terms, Haystack does not exclude any particular date ranges like holidays, PTO, weekends, etc.
While initially this may feel like your Cycle Time is artificially increased, we think about Cycle Times from the perspective of the customer. In this case, the customer is still waiting during the weekends and the increased Cycle Time continues to be a great signal on where bottlenecks might lie.
Example: Even if PRs are counted over weekends, it's beneficial to be alerted to these long standing reviews since weekends tend to introduce additional context switching, overhead, and PRs are prone to being lost/forgotten.
Pro-Tip: Aggregate numbers are less important than overall trends. By using the Average Cycle Time across larger time frames (e.g. 6 months), you'll get a high-level view into improvement trends.