Skip to content
Least Privilege Principle

Least privilege projects stall because removing access carries more risk than granting it

Cloud least privilege projects stall at four points: nobody can show a removal is safe, usage data covers too short a window, clean-up happens one permission at a time, and the report ends in a ticket. Provider documentation read on 16 Nov 2026 describes a way past each of the four.

Open the calculator ยท The field guide

Granting is cheap and removal has an owner who fears the outage

AWS documents the pattern in its own IAM guidance: you might start with broad permissions while you explore what a workload needs, and as the use case matures you work to reduce them. The first half of that sentence happens on every project. The second half needs someone to decide that a removal is safe, and that person carries the outage if the decision is wrong.

That asymmetry is the first reason projects stall. A broad grant costs nothing to approve and a failed deploy is visible within minutes. A removal that breaks a quarterly job is visible three months later, with the engineer who approved it as the named cause. Teams respond by freezing the clean-up queue. The way past it is evidence per change: a record of what the change would have blocked, before it ships.

The usage window decides what looks unused

Usage-based tools compare what is granted with what was used in a window. Google documents that the IAM recommender compares a principal's permissions with those used in the last 90 days, and that by default it does not generate recommendations until it has 90 days of data. For project-level recommendations the minimum can be set to 30 or 60 days, and Google says accuracy might be affected. AWS lets policy generation read up to 90 days of CloudTrail events.

A role used once a year for failover, or a job that runs at quarter close, can sit outside any 90-day window. Removing its permissions because they look unused is the mistake that makes the next removal harder to approve. Treat the window as a stated limit of the data, and list the identities whose duties run on longer cycles before any removal. Privilege creep lists the other windows.

Permission-by-permission clean-up does not scale

Google's role recommendations each suggest that you remove or replace a role for a principal. That is the right unit for a review and the wrong unit for an estate of hundreds of accounts. If every finding is a ticket, the queue grows at the rate that engineers create roles, which is faster than anyone closes tickets.

A boundary is the alternative unit: one reviewed limit that acts on every identity beneath it, covered on what a boundary is. The Access Debt Calculator shows the size of the difference for your own counts, and its arithmetic is on the methodology page.

A finding ends the tool's work and starts the team's

AWS Access Analyzer recommends a refined policy for unused permissions and the user attaches it. Google states that the IAM recommender does not apply recommendations automatically, and that you apply or dismiss each one. Both are sound designs for a provider tool. A recommendation service recommends, and the remaining work is staffing: someone must read each result, test it and ship it.

Whoever owns the last step also owns the risk of the change, so the tool's own work is not the whole project. Name the owner of the last step before the first finding arrives, and put it in the written policy.

New access arrives faster than old access leaves

Even a clean estate decays. Workloads and identities are created and removed by pipelines at any hour, so a clean-up is a snapshot, and the next deploy can re-grant what was just removed. The durable answer is a check before production: a policy test in the pipeline, with exceptions handled in the team's own tools. AWS documents custom policy checks in IAM Access Analyzer, which check a policy for new access against your security standards and can be run from the CLI or API as well as the console.

A project with no check before production is cleaning a floor while the tap runs. The first 90 days ends with such a check for that reason.

Count the stall by its stage. If findings exist and no change has been shipped, the stall is the first or second point and needs an owner and a longer window. If changes ship and then return, the stall is the fifth point and needs a check in the pipeline. If every change takes a week of meetings, the unit is wrong and a boundary is the better candidate. Each diagnosis points to a different next step, so naming the stage saves a quarter spent on the wrong one.

Sources

  1. AWS, security best practices in IAM, read 16 Nov 2026
  2. Google Cloud, role recommendations overview, read 16 Nov 2026
  3. AWS, generate policies based on access activity, read 16 Nov 2026
  4. AWS, IAM Access Analyzer findings, read 16 Nov 2026
  5. AWS, IAM Access Analyzer custom policy checks, read 16 Nov 2026

Last reviewed 16 Nov 2026.