# Simulate a policy before publishing > How to replay a draft policy against recent decisions with no model call, and how an explainer's suggestion can land in a draft without ever publishing it. - Canonical URL: https://lumtry.com/guides/simulate-a-policy-before-publishing - Last updated: 2026-09-26 Before you publish a new policy version, Lumtry can replay it against recent cases with no model call: the same draft, window and point in time always produce the same result. An explainer can propose one rule change from that replay, but it only ever lands as a pending suggestion in your draft, never a publish. ## What does a policy simulation actually run? The same deterministic policy evaluator and floor gate a live case runs, replayed against a window of recent decisions as of a point in time you choose. Nothing about a simulation calls a model or writes to a published version. ## Will a simulation give the same answer twice? Yes. The same draft, window and as-of point produce byte-identical results every time; that determinism is what a release gate checks before the platform ships. ## What does the explainer add on top of a simulation? The explainer reads a simulation's results and can draft one candidate rule change. Accepting it adds that rule to your draft, tagged as an AI suggestion; it never publishes on its own, and your workspace's active published version and its history stay untouched until you publish. ## Is simulating a policy available on every plan? Yes, simulation is free on every plan, with no model call involved. The explainer that proposes a rule needs the AI rule suggestions entitlement, included on Growth, Scale or a trial. ## Simulate a draft and review a suggestion in five steps 1. **Write or edit your draft policy**: Change the rules in a draft version; nothing you simulate touches the version your live cases run against. 2. **Choose a window and a point in time**: Pick the recent decisions to replay and the moment to evaluate them as of, so the simulation reflects a specific slice of your history. 3. **Run the simulation**: The draft replays through the same evaluator and floor gate a live case uses, with no model call, and returns the same result every time you run it unchanged. 4. **Read the explainer's suggestion, if one appears**: Where the AI rule suggestions entitlement is on, an explainer can propose one candidate rule from the replay, shown as a pending suggestion you can accept, edit or ignore. 5. **Publish only when you're ready**: Accepting a suggestion adds it to the draft; publishing is still a separate, explicit step that starts a new policy version. ## Frequently asked questions ### Does a simulation change what my live cases decide? No. It runs against your draft only; a published policy version is immutable once live, and simulating never edits or replaces it. ### Can I trust that a simulation didn't quietly call a model? Yes. Simulation is built from the pure policy evaluator and the pure floor gate alone; a release gate treats any model call during a simulation as a failure. ### Does accepting an explainer suggestion publish my policy? No. Accepting adds the suggested rule to your draft with an AI-suggestion tag. Your active published version and how many versions you've published stay exactly the same until you publish yourself. ### Who can see that a rule came from the explainer? Anyone reviewing the draft. An accepted suggestion carries its AI-suggestion provenance in the draft, so it is never indistinguishable from a rule you wrote by hand. ## Related - [Home](https://lumtry.com/index.md): Decide refunds by a versioned policy, send the exceptions to your team in Slack or the dashboard, and keep every step on the record. Start free. - [Guides](https://lumtry.com/guides.md): Answer-first guides to refunds, returns, chargebacks, approvals and audit trails for Shopify, WooCommerce and BigCommerce merchants. - [Slack approvals for Stripe](https://lumtry.com/guides/slack-refund-approvals-for-stripe.md): How to route Stripe refunds that need a person to Slack, approve them from the message, and keep the dashboard as the source of truth. - [Voice approvals](https://lumtry.com/guides/voice-refund-approvals.md): How outbound phone calls let an approver confirm a held refund when they are away from Slack and the dashboard, with the answer recorded. - [Return abuse screening](https://lumtry.com/guides/return-abuse-screening.md): How to score refund requests on claim history, missing returns and unusual values, and hold risky cases for a person without slowing honest customers. - [Chargeback evidence](https://lumtry.com/guides/chargeback-evidence-automation.md): How to assemble dispute evidence from orders, policy and the refund audit trail so every chargeback response starts from a complete file. - [Refund audit trail](https://lumtry.com/guides/refund-audit-trail.md): What a refund audit trail should record, why it must be append-only, and five steps to build one that holds up in disputes and audits. - [Refunds for AI agents](https://lumtry.com/guides/mcp-refunds-for-ai-agents.md): How AI assistants can look up refund cases and propose refunds over the Model Context Protocol while policy and people keep the final say. - [Retention offers](https://lumtry.com/guides/retention-offers-before-refund.md): When to offer store credit, an exchange or a partial refund before paying money back, and how a policy decides which offer fits a case. - [Marketplace refunds](https://lumtry.com/guides/multichannel-marketplace-refunds.md): How to bring store and marketplace refunds from Amazon and eBay into one queue, with a policy per channel and every decision in one audit trail. - [AI decisions and contests](https://lumtry.com/guides/ai-decision-record-and-contest.md): How Lumtry records every AI-assisted refund decision and lets a shopper contest it, with a person reviewing and deciding the outcome. - [Canadian refund rules](https://lumtry.com/guides/canadian-refund-rules-for-merchants.md): What Quebec and Ontario require for online sales, why a posted refund policy can create legal exposure, and how to set these rules as your own policy in Lumtry. - [Pricing](https://lumtry.com/pricing.md): Lumtry plans by case volume and seats, from Free to Enterprise, with add-ons and capped outcome fees. Start free or hear when billing opens. - [MCP server](https://lumtry.com/developers/mcp.md): Connect an AI assistant or your own agent to your refund workspace through the Model Context Protocol. Read cases and propose actions, never move money. - [Trust](https://lumtry.com/trust.md): How Lumtry secures merchant data, keeps AI inside a deterministic refund policy, lists every sub-processor and explains data flow and retention. Review it. - [llms.txt](https://lumtry.com/llms.txt)