How a run starts

A run is what actually produces your segments. Neuralift starts it for you once two things are in place: a filled-in definition and a prepared dataset. This page covers what needs to be true before a run starts, what happens while it works, and what you get back. For the full split of who does what, see Roles & permissions.

Before you run

You’ll typically confirm these together with Neuralift:

• The use case has a goal and business context in its definition.
• A desired segment size is chosen.
• The definition’s KPIs are added and prioritized.
• Data preparation has finished, so the data dictionary shows your columns and the prepared row count.

How a run starts

Once the definition and data are confirmed, Neuralift starts the run and the use case’s status moves from pending to running. You don’t need to keep the page open; status updates in real time, and the use case shows its current state on the dashboard.

Neuralift monitors every run. If a run fails, Neuralift investigates and re-runs it; there’s nothing you need to do. A use case can also be run again after it finishes, for example after a change to the definition.

Note. Runs work on the prepared dataset inside Neuralift. Nothing is written back to your warehouse by a run; delivering results is a separate, explicit step.

What you get back

When the run reaches done, the results are in: a segment landscape under the use case’s Discovery tab, along with ranked actions and their plans. At that point they’re visible only to Neuralift, who reviews them with your team. Once everything holds up, Neuralift publishes the run and the results open up to your whole organization.

See The run lifecycle for what each status means.