Operate

How an undo gets triggered

The ways an undo actually starts: suggested by the AI, requested by you, or run across all your stores.

An undo always follows the same confirmed steps once it starts. What changes is what sets it in motion.

The AI suggests it

While looking into a problem, the AI may find that a specific past change is the likely cause and suggest going back to before it. Like any suggestion, you confirm before anything runs. See How Reqursor fixes unexpected changes with AI.

You request it

Through the dashboard or in conversation ("take Store X back to yesterday"), Reqursor Platform turns your request into one exact version before showing you anything to confirm. See How "go back to yesterday" finds the right version.

It runs across all your stores

An undo that is suggested or requested for a group of stores doesn't hit every store at once. It follows your rollout policy for all your stores (for example, one store first, checked, then the rest in groups), so a bad version is caught early rather than applied to every store at the same time.

Reqursor's own team, not just you

In rare support situations, the Reqursor Platform team can also start an undo directly: always through the same permission checks and the same record of who did what as any other trigger, never skipping them.

Why this matters for your agency

  • Whoever or whatever starts it, the guarantee is the same. Every undo (suggested by the AI, requested by you, or run across all your stores) goes through the same confirmation and the same steps.
  • Undoing across all your stores doesn't mean risk across all your stores. Your rollout policy catches a bad version before it reaches every store.
  • Nothing is undone without a record. Every trigger, including Reqursor's own, is permission-checked and recorded the same way.

Where to go next