The cloud bill jumped sixty percent this month. How do you find out why?

Start from the billing data grouped by service, then by account, region and tag, and compare against the previous month to isolate exactly which line item moved. Once you have the service, check whether it is more usage or a price and configuration change, such as a new instance family, cross zone data transfer, or storage that never expires. Then fix the cause and add a budget alert with anomaly detection so it is caught earlier next time.

I would treat it as a debugging exercise and go after the delta rather than the total. Group the bill by service and diff against last month, and usually one or two line items account for almost all of it. Then narrow within that: which account, which region, which tag. If tagging is weak this is where it hurts, and that becomes a follow up in its own right. Once I know the line item, I work out whether we are using more of something or paying more per unit, because the fixes are completely different. The recurring culprits I have seen are cross zone or egress data transfer after somebody moved a chatty service, log and metric ingestion where a debug level got left on, storage with no lifecycle policy quietly accumulating for years, and a non production environment that scaled up for a load test and never scaled back. Then prevention: budget alerts with anomaly detection per account, so a sixty percent jump pages someone within a couple of days rather than arriving as a surprise in the invoice.