/dm:what-if

Purpose

Quick scenario comparison tool. Test 2-4 marketing scenarios against each other — different budget allocations, channel mixes, or strategic approaches — and see projected outcomes side-by-side. This is the lighter, faster alternative to full Monte Carlo simulation (/dm:simulate). Where simulate runs thousands of iterations with full probability distributions, what-if uses point estimates with simple variance bands to give directional answers in minutes. Use it for rapid decision-making when you need a quick read on "should we do A or B?" without the statistical depth of a full simulation — team meetings, Slack discussions, quick planning calls, or narrowing down options before running a deeper analysis.

Input Required

The user must provide (or will be prompted for):

• Scenarios to compare: 2-4 named scenarios, each with channel-level budget allocations and expected ROI per channel. Examples: "Scenario A: Heavy paid — $50K Google Ads, $30K Meta, $10K email" vs "Scenario B: Content-led — $20K Google Ads, $15K Meta, $40K content, $15K SEO." Each scenario needs a descriptive name and channel budget breakdown. If the user provides only high-level descriptions ("more on paid, less on organic"), ask for specific dollar allocations or percentage splits
• Current baseline: The existing budget allocation and recent performance as the reference point for comparison — what the brand is doing right now so each scenario shows a clear delta. If not provided, pull from brand context historical data
• Evaluation criteria (optional): What matters most for this decision — total revenue, ROI efficiency, risk level, speed to impact, or a weighted combination. Defaults to expected revenue if not specified
• Time horizon (optional): How far out to project — defaults to 3 months. Shorter horizons favor paid channels, longer horizons favor organic and content investments due to compounding effects

Process

1. Load brand context: Read ~/.claude-marketing/brands/_active-brand.json for the active slug, then load ~/.claude-marketing/brands/{slug}/profile.json. Pull historical channel performance, recent ROI data, and known benchmarks to calibrate scenario projections. Also check for guidelines at ~/.claude-marketing/brands/{slug}/guidelines/_manifest.json. Check for agency SOPs at ~/.claude-marketing/sops/. If no brand exists, ask: "Set up a brand first (/dm:brand-setup)?" — or proceed with industry defaults.
2. Define current baseline and alternative scenarios: Structure the current state as Scenario 0 (baseline) with actual recent performance data. Then define each user scenario with channel budgets and ROI assumptions — using brand historical data where available, industry benchmarks where not. Flag any assumptions that differ significantly from historical performance so the user can validate them.