Preview before write Diffs, policy posture, and rollback readiness appear before the connector executes.
Simulation or sandbox first The product starts with a deterministic story before it asks for production trust.
Proof stays attached Approval, audit, artifacts, and recovery context stay on the same governed run.
Connect your sandbox

Repeat the sample in an environment that actually resembles production.

The buyer and the technical evaluator should both be able to see the same story in a realistic non-production environment. That means free developer org where available, vendor trial where necessary, and customer-owned sandbox where proof only matters with real permissions and data shape.

Two explicit entry points

Open simulation workspace when you want the first story without credentials.

Connect your own sandbox when you need the workflow to mirror your real environment.

Do not blur the two paths together in the docs or the app shell.

What the sandbox must prove

The canonical sample scenario still makes sense with your actual object model, permissions, and reviewer path.

Preview, approval, output evidence, rollback context, and audit history remain legible without founder explanation.

Reset or fixture refresh is possible so repeated demos do not drift into confusion.

How to choose the right sandbox lane

Use official free developer environments for Salesforce CRM, ServiceNow, and HubSpot when possible.

Use vendor trials for Zendesk, Freshdesk, and Intercom, but be honest that they may differ from paid production behavior.

Use customer-owned sandboxes for ERP, finance, and HCM workflows where the proof depends on realistic controls and reference data.

A strong first connector story

One connector only.

One narrow workflow only.

One realistic sample record or ticket family.

One clear reviewer group for risky writes.

One obvious go/no-go decision for moving toward production.

Start with Salesforce if you need the clearest story

Salesforce remains the clearest initial buyer and operator story because the sample workspace, canonical scenario, and first live developer-org workflow all line up.