Optimizing Clay Signals for Efficient Bulk Enrichment in Salesforce

I have around 500,000 Salesforce contacts in Clay, and I'm trying to build a Signals system using bulk enrichment. I want to make sure I structure this in the most credit-efficient way. I have a few questions: Q1: I have multiple Signals that need to run on different schedules (monthly, weekly, and daily). What’s the most efficient way to separate these so I don’t burn unnecessary credits? For example, I’m planning to split them into separate tables based on their run frequency so the monthly and weekly Signals don’t run daily. Related to this: If a row already has results (e.g., 300k rows in a daily Signal table) and I run the Signal again but the data hasn’t changed, will Clay still consume credits? Or can I configure the table so it only uses credits when something new or changed needs to be checked? Q2: When using bulk enrichment for Signals at this scale, how does Clay decide what to process? Does it repeatedly process the first 10k rows (if they’re constantly changing)? Or will it continuously run all 500k (considering for daily signals) rows every time the table refreshes? If yes is it possible for clay to do the same everyday?