Key Takeaways
- Automatic record creation – Every signal decision in the Decision Queue is permanently logged with EVA’s full analysis, your rationale, and a timestamp. No manual logging required.
- Outcome tracking built in – When you reverse a signal, EVA classifies the outcome after your review window closes: Met, Missed, or Unresolved. No follow-up needed from your team.
- Persistent issues never fall through – Missed reversals go back into the Decision Queue automatically, with a re-entry count so recurring signals are visible, not buried.
- Four tabs, one complete view – Waiting for Reversal, Reversal Met, Reversal Missed, and Accepted give every team full decision lifecycle visibility in a single place.
- Built for audit and compliance – Every record is immutable, tied to a person, a timestamp, and a rationale. The trail assembles itself.
The problem with cloud decisions isn’t making them. It’s proving they happened – and knowing if they worked.
Your cloud generates signals every day. Cost spikes, underutilised resources, security exposures, configuration drift. Your team reviews them, makes calls, and moves on.
But what comes after the decision? Who checks if that reversed signal was actually addressed? Who flags it when it wasn’t? And three months from now, when a compliance audit asks you to show your decision trail – where does that evidence live?
For most teams, the honest answer is: scattered across tickets, Slack threads, and spreadsheets – if it exists at all.
Decision Records on Cloudeva.ai changes that.
What Decision Records is
Decision Records is where every signal decision your team has ever made lives permanently.
Every time you accept or reverse a signal in the Decision Queue, Cloudeva.ai creates an immutable record automatically. That record contains EVA’s full analysis (what changed, the root cause, the recommended action), your rationale, the timestamp, and the person who decided. Nothing is retroactively edited. Nothing is deleted.
Think of it as your cloud’s decision memory. Not a log you have to fill in manually. A complete, auditable history that builds itself.
The part most teams don’t have: outcome tracking
Accepting a signal is a one-time action. Reversing one is a commitment – it means your team is going to address the issue within a defined review window.
The question is: does anyone follow up?
With Decision Records, EVA does. After the review period ends, EVA evaluates the outcome and classifies it automatically:
- Met – the fix was applied. Record updated. Done.
- Missed – the issue still exists. The signal goes back into the Decision Queue for another review cycle.
- Unresolved – the outcome couldn’t be determined with confidence.
No calendar reminders. No manual checks. No signals that slip through because someone assumed someone else was handling it.
Four tabs so you always know where things stand
Decision Records is organised into four views:
Waiting for Reversal
Decisions where the review window is still open. Your team committed to a fix; EVA is tracking it.
Reversal Met
Reversals that were successfully applied. Evidence that governance is working.
Reversal Missed
Reversals that weren’t carried out and have re-entered the queue. Each record shows a re-entry count – so you can see which signals have cycled back more than once, and treat them as persistent issues rather than one-time anomalies.
Accepted
All signals your team acknowledged and closed. The complete record of every cost signal, risk signal, and configuration change you made a conscious decision about.
What you see when you expand a record
Click any record and you get the complete picture: a visual timeline of every action taken on that signal, EVA’s original analysis, the recommended action that was surfaced, and the full outcome detail for reversals.
Not a summary. The actual decision – exactly as it happened, with the context that existed at the time.
Why this matters for different teams
- Cloud engineers – no manual follow-up needed. If the reversal wasn’t carried out, the signal comes back to you.
- FinOps teams –complete audit trail of every cost-impacting decision, with outcome evidence for reporting and compliance.
- Security teams – verification that flagged exposures were actually remediated, with automatic re-entry if they weren’t.
- Engineering managers – full decision lifecycle visibility – who decided, what EVA recommended, what happened after, and how many times a signal had to cycle back.
The difference in practice
Most governance tooling records that a decision was made. Decision Records records what was decided, who decided it, what context existed at the time, and whether the outcome was actually achieved.
That’s not a subtle difference. It’s the difference between an audit trail and evidence.
Traditional governance vs. Decision Records
| Traditional governance | Cloudeva.ai Decision Records |
| Decision history in spreadsheets or tickets | Permanent records with full signal context |
| “Did anyone actually fix that?” | EVA evaluates outcomes automatically |
| Manual follow-up on reversals | Missed reversals go back to the queue automatically |
| Decisions and outcomes in separate places | Everything in one view with four clear tabs |
| Audit trail requires manual assembly | Every action timestamped with who decided and why |
Access and scope
Decision Records is scoped to the accounts you have access to. Your view is always relevant to you – nothing from accounts outside your access scope appears in your records.
Available now
Decision Records is live on Cloudeva.ai. If you’re already using the Decision Queue, your records are already building. Head to Decision Records in your navigation to see them.
| Stop guessing. Start governing. Every cloud decision your team makes deserves a record – and every reversal deserves a follow-up that doesn’t depend on someone’s memory. Decision Records gives you both. Automatically. → Get started at zeo cost – cloudeva.ai |
Frequently Asked Questions
Do I need to do anything to start using Decision Records?
No. If you’re already using the Decision Queue, records are being created automatically. Every signal you accept or reverse is already logged – just navigate to Decision Records to see them.
Can I edit or delete a record after it’s been created?
No. All records are immutable. Once a decision is logged, the rationale, timestamp, and person who decided are permanently recorded. This is by design – the audit trail only holds value if it can’t be altered after the fact.
What happens when a reversal is missed?
EVA detects that the issue still exists after the review window closes, classifies the outcome as Missed, and automatically sends the signal back into the Decision Queue for another review cycle. The record retains its history, and the re-entry count increments – so you can see how many times a signal has cycled back.
What does “Unresolved” mean as an outcome?
It means EVA was unable to definitively confirm whether the fix was applied or not. This can happen when the signal involves a change that’s difficult to verify programmatically after the fact. The record stays in your history and the signal does not automatically re-enter the queue unless the status updates.
Who can see Decision Records?
Access is scoped to the accounts assigned to you. You only see records tied to accounts within your access level – there’s no cross-account visibility beyond your assigned scope.
Is the full EVA analysis preserved in each record?
Yes. When you expand any record, you see the complete EVA analysis as it existed at the time of the decision – what changed, the root cause assessment, and the recommended action. The record is a snapshot of the full signal context, not just the outcome.
Can engineering managers see decisions made by their team?
Yes. Decision Records gives managers full decision lifecycle visibility across accounts within their access scope – including who decided, what EVA recommended, and how the outcome resolved.
Does Decision Records cover both cost signals and risk signals?
Yes. Any signal you act on in the Decision Queue – whether it’s a cost signal, a risk signal, or a configuration change – generates a record. The audit trail is unified across signal types.