Scenarios & what-ifs

The Scenarios screen is a model's what-if workbench. The first scenario is Base — it captures the original solve and is read-only. Every other scenario branches from it with its own solve and its own KPI deltas.

The KPI strip across the top keeps score: Base, Best scenario, and Scenarios solved (n of total), with deltas shown as +$amount (+pct%).

Creating a what-if

Plain English is enough. The scenario editor's What if… field carries its own example: "What if flour costs jump 30% and we lose 2 hours of oven time?" — RIA translates the sentence into the right model adjustments. Under Advanced you can state explicit overrides yourself. Run scenario solves it.

Chat works too: on the Scenarios screen, "each message becomes a new what-if scenario."

RIA also proposes Starter what-ifs — questions worth asking about your specific model, each with the reasoning behind it. Click one to run it.

The Outside context card lists external market factors RIA is watching for your business — each one a click from becoming a scenario.

Comparing

Select 2–5 scenarios and click Compare — they open side by side in a workspace (objective ladder, KPI deltas, per-period comparisons). Pick top 3 by objective grabs the leaders in one click.

With many scenarios, Browse all opens the index: search by name, prompt, or lever; sort by objective, date, name, or status; smart lists for Pinned, Top performers, Recent, Stale, Infeasible, Solved, Failed.

Staleness

When the model is re-authored, existing results are flagged stale · solved rev N — they came from an older structure. Each keeps its prior result so you can see the delta after re-solving; Regenerate all re-runs them against the current model.

Diagnosing

A scenario with no feasible answer gets a Diagnose button — a focused diagnosis of what's binding, the shortfall, and the prescribed fix, in business terms.

Verifying

Verify builds an evidence record for a solved scenario: the Result, the Inputs it used, the Checks run against it, and the Trail — with Export for taking the evidence outside eode. Results start as Estimated; re-verifying re-loads the model, re-solves, and pulls a deterministic per-component breakdown, upgrading the scenario to Verified.

Going deeper

Open in workspace turns any solved scenario into an analysis canvas — see Workspaces. The Workflows card on the same screen lists the model's workflow templates; Run with [scenario] starts a workflow run with the scenario's results feeding the run's widgets.