Why this matters
Marketing plans rarely go exactly as expected. Scenario forecasting helps you prepare for upside and downside so you can set targets, allocate budget, and communicate risks clearly. As a Marketing Analyst, you will:
- Plan campaign budgets with a realistic range for revenue and leads.
- Align teams on what “good” and “bad” months look like.
- Pre-define triggers for action (e.g., pause ads if worst-case trends appear).
- Report to stakeholders with a clear, credible range instead of single-point guesses.
Who this is for
Junior–mid Marketing Analysts, Growth Marketers, and PMMs who need to turn assumptions into a forecast range for channels, launches, or quarterly plans.
Prerequisites
- Basic spreadsheet skills (formulas, percentages, references).
- Comfort with core funnel metrics: sessions/visits, conversion rate (CR), average order value (AOV), cost per acquisition (CPA), return on ad spend (ROAS).
- Very light statistics: weighted average, percentage points vs percent change.
Concept explained simply
Scenario forecasting builds three views of the future:
- Best case: Optimistic but plausible if several things go right.
- Base case: Most likely given current information and typical seasonality.
- Worst case: Conservative view if key risks happen.
You create the three cases by changing the drivers (traffic, conversion, AOV, costs) using assumptions for each case. Then you compute outputs (orders, revenue, CAC, ROAS) for each case.
Mental model
Think of your forecast like stacking blocks:
- Block 1: Volume (sessions, impressions).
- Block 2: Quality/efficiency (CR, CTR, CPC, CPA).
- Block 3: Value (AOV, LTV, margin).
Best/base/worst simply use different-sized blocks. Don’t guess outputs directly—adjust the blocks (drivers) and let the math roll up.
Mini refresher: percent vs percentage points
Percent change changes the level: 2.5% CR to 2.8% CR is a +12% change because (2.8-2.5)/2.5 = 12%. Percentage points change is the direct difference: +0.3 pp. In scenarios, specify CR changes in percentage points to avoid confusion.
Build scenarios step-by-step
Gather last comparable period or a rolling average. Include seasonality factors if known (e.g., 0.95 for a slow month).
Common drivers: sessions (organic/paid), CR by channel, AOV, price/promo, budget/CPC, supply/inventory constraints, site speed, and funnel changes.
Base: expected trend + seasonality. Best: optimistic, still realistic. Worst: conservative impact from 1–2 key risks. Use percentage points for CR and percent changes for volume/value.
Orders = Sessions × CR. Revenue = Orders × AOV. Costs = Spend. ROAS = Revenue / Spend. CAC = Spend / Orders.
Define what you’ll do if actuals track closer to worst (e.g., if CR below X for 2 weeks, cut spend by Y%).
Present a clear range with the drivers and assumptions. Add a brief note on risks and mitigations.
Worked examples
Example 1: Monthly ecommerce revenue
- Baseline sessions: 100,000 (unseasonalized)
- Seasonality factor next month: 0.95
- Base CR: 2.5% (0.025), Base AOV: $60
- Best: sessions +15%, CR +0.3 pp, AOV +5%
- Worst: sessions -20%, CR -0.5 pp, AOV -5%
Base sessions = 100,000 × 0.95 = 95,000. Base revenue = 95,000 × 0.025 × 60 = $142,500.
Best revenue = 95,000 × 1.15 × 0.028 × 63 = $192,717.
Worst revenue = 95,000 × 0.80 × 0.020 × 57 = $86,640.
Example 2: Lead gen with budget risk
- Leads = Clicks × CVR-to-lead
- Clicks = Spend / CPC
- Base: Spend $40,000, CPC $2.00, CVR 8%
- Best: CPC -10%, CVR +1 pp
- Worst: Spend -25% (budget cut), CPC +10%, CVR -1 pp
Base: Clicks = 40,000 / 2.00 = 20,000; Leads = 20,000 × 0.08 = 1,600.
Best: CPC = $1.80; Clicks ≈ 22,222; CVR = 9%; Leads ≈ 2,000.
Worst: Spend = $30,000; CPC = $2.20; Clicks ≈ 13,636; CVR = 7%; Leads ≈ 954.
Example 3: Subscription trial-to-paid
- Trials = Visits × Signup rate
- Paid subs = Trials × Trial-to-paid CR
- Base: Visits 200,000; Signup 6%; Trial-to-paid 30%
- Best: Visits +10%, Signup +0.5 pp, Trial-to-paid +3 pp
- Worst: Visits -15%, Signup -1 pp, Trial-to-paid -5 pp
Base: Trials = 12,000; Paid = 3,600.
Best: Visits 220,000; Signup 6.5% → Trials 14,300; Trial-to-paid 33% → Paid 4,719.
Worst: Visits 170,000; Signup 5% → Trials 8,500; Trial-to-paid 25% → Paid 2,125.
Scenario setup checklist
- Baseline and seasonality defined.
- Drivers separated by channel or funnel stage.
- Assumptions stated in percent or percentage points (clearly labeled).
- All outputs derived from drivers (no hard-typed outcomes).
- Worst case includes at least one real risk (budget, supply, traffic).
- Actions/triggers defined for downside.
Common mistakes and how to self-check
- Mistake: Changing outputs directly (e.g., “revenue +10%”). Fix: Change sessions/CR/AOV and recompute revenue.
- Mistake: Confusing percent and percentage points for CR. Fix: Note CR changes as pp; convert to percent only when needed.
- Mistake: Single uncertainty source. Fix: Stress at least 2 meaningful drivers in best/worst.
- Mistake: Copying base multipliers across channels. Fix: Use channel-specific assumptions when data suggests differences.
- Mistake: Forgetting seasonality. Fix: Apply factors before scenario adjustments.
- Mistake: Linear add when multiplicative is needed. Fix: Apply sequentially: sessions → CR → AOV.
Self-check mini task
Pick one driver you did not change in your worst case. Would a realistic risk affect it? If yes, add a conservative adjustment and see how the range changes.
Hands-on exercises
Do these in a spreadsheet. They mirror the exercise cards below.
Exercise 1: Ecom monthly scenarios
Given: Baseline sessions 100,000 (unseasonalized). Seasonality 0.95. Base CR 2.5%, AOV $60. Best: sessions +15%, CR +0.3 pp, AOV +5%. Worst: sessions -20%, CR -0.5 pp, AOV -5%.
- Compute base, best, worst revenue.
- Compute best and worst deltas vs base (absolute and %).
Open the exercise card below to check the full solution.
Exercise 2: Probability-weighted revenue
Using your results from Exercise 1, assign probabilities: Best 20%, Base 60%, Worst 20%.
- Calculate expected revenue (weighted average).
- State the scenario range (worst to best).
Open the exercise card below to check the full solution.
Practical projects
- Quarterly campaign plan: Build best/base/worst for paid search, paid social, email. Include triggers for reallocating 20% of budget if worst trends persist for 2 consecutive weeks.
- New product launch: Create scenarios with traffic uncertainty and supply constraints. Share a one-slide summary with assumptions and mitigations.
- Subscription KPI pack: Best/base/worst for trials, trial-to-paid, churn. Add a simple probability-weighted MRR for internal planning.
Learning path
- Forecasting basics refresher (drivers and funnel math).
- Scenario modeling with three cases and clear assumptions.
- Channel-level scenarios (paid vs organic differences).
- Probability-weighted planning and trigger thresholds.
- Communicating ranges and actions to stakeholders.
Next steps
- Turn one of your current plans into best/base/worst with drivers, not outputs.
- Share with a stakeholder and ask, “What risk feels underrepresented?” Update worst case accordingly.
- Track actuals weekly against the three paths; note which path actuals follow and why.
Mini challenge: One-slide scenario story
Create a single slide that shows: 1) Base forecast; 2) Range bar from worst to best; 3) Top 3 drivers changed; 4) One trigger and mitigation. Keep it to 90 seconds of talk time.
Quick Test note: The quick test is available to everyone; only logged-in users get saved progress.