Control operations keeps ZeroTB's team carrying checks and repairs after the Pilot ends.

Works with the stack you already run
ZeroTB has undergone an independent audit and operates this control lifecycle in its own compliance program. Its report is available under NDA.
ZeroTB keeps checking and fixing controls all year
Control operations combines ZeroTB's software and team to check and repair controls after audits for US software, AI, fintech and healthtech companies of 25 to 250 people.
Same owners, same permissions. The owners and access agreed in the Pilot carry into continuing work
Pilot inventory reused. We start with the commitments and control conditions established during the Pilot
One record. ZeroTB keeps commitments, checks, repairs and audit evidence together
Old evidence costs audit findings and repeat investigations
Accounts retain access, repository rules change and device protections fall behind while earlier evidence describes the old condition.
A control can fail long before the next audit reveals it. The gap returns as a finding that your team investigates again.
Systems and permissions change
Earlier evidence stays on file
The gap reaches audit
Our team fixes each failing control and rechecks it
A repair stays open until ZeroTB checks that the control works. Mean Time to Verified Restoration measures how long reaching that result takes.
Agree the requirement
We extract each commitment from your report with its scope, owner and check.
Check the condition
Daily checks compare systems with commitments, attaching source clauses and owners to gaps.
Prepare the repair
We investigate and prepare a runbook, pull request, API action or guardrail.
Verify and record
ZeroTB rechecks the system and signs a Verified Control Record.
We cover accounts, releases, cloud and recurring work
ZeroTB handles technical controls and scheduled activities, including training and vendor reviews that need a person to complete them.
We check accounts, access permissions and device protections against your policies.
We check branch rules, required reviews and CI gates on repositories that ship code.
We compare infrastructure settings with policy and prepare repairs as infrastructure-as-code pull requests.
We schedule training, vendor reviews, incident exercises and policy reviews, assign owners and escalate overdue work.
You decide exceptions and see which controls work
Confirm commitments
You confirm that each extracted requirement matches what your company committed to do.
Approve material changes
Pull requests follow your normal review and merge process.
Decide exceptions
Your authorized people decide and own legitimate exceptions.
Commitment inventory
Each commitment retains its source, scope, owner and expected evidence.
Current control states
Checks show operating or failing controls, with unknown, stale and disconnected observations kept explicit.
Verified Control Record
Auditors read each signed restoration record in ZeroTB. You retain portable evidence.
Example of removing lingering access
The contractor in this example has left, but the account still works. The record shows its removal and the check that confirms access ended.

Closed task. The departure task is complete, but its status cannot establish the account's condition.
Remaining access. The account still reaches company systems after the contractor has left.
Confirmed removal. The record shows approved removal and a later check confirming that access ended.
Talk to our team
Talk with us about which controls need continuing attention and where investigations or repairs keep passing between teams.
FAQ
How does Control operations follow the Pilot?
Control operations starts from the Pilot's commitments and control conditions. The agreed owners and permissions continue as our team carries new gaps through repair.
Can a repair involve work a person performs?
A repair can use a runbook for a person to follow. Other forms include reviewed pull requests, approved API actions and preventive guardrails.
Must we grant write access to every connected system?
ZeroTB starts read-only. Write access is scoped to classes of action and remains revocable. Direct API actions require opt-in permission.
How is the time to restore a control measured?
Mean Time to Verified Restoration runs from drift detection until ZeroTB verifies the restored control, including any work remaining after a ticket is marked done.
Can one repair support SOC 2, ISO 27001 and HIPAA?
One repair can support SOC 2, ISO 27001 and HIPAA when their requirements share the restored control. The record retains the requirements it supports.
Does Control operations replace periodic control testing?
Control operations adds observations between periodic tests, following each commitment's scope and cadence. Continuous checks do not replace the wider testing plan.
Start with the Pilot
The eight-week Pilot maps up to 25 commitments and takes two real control gaps through verified restoration, with portable evidence.