Outcome
By the end of this lesson, you will be able to build a top-down cost forecast from historical pattern plus leadership goals, allocate the total to BUs and teams using historical proportions, and recognize when top-down alone misses important team-specific dynamics.
| Prerequisites | M5.4, Budget governance · M5.3, Unit economics |
| Time | 9 minutes |
1. Concept
Top-down starts with one number for the whole company and divides it up.
You take where total spending has been heading, project that forward, and split it between business units and teams in the same proportions they used before.
It is quick, it needs nothing from anybody else, and it is the right method when you need a company-level number, or when teams are not yet in the habit of forecasting their own.
TOP-DOWN PROCESS: 1. Pull historical org-wide spend (12-24 months) 2. Apply a growth rate from leadership plan + extrapolation 3. Add planned major events - New product launches with quantified cost impact - M&A integrations - Migration projects with known cost shape 4. Add safety margin (typically 10-15%) 5. Allocate to BUs/teams using historical proportions 6. The result IS your committed forecast for the periodTop-down is one method. Bottom-up (L2) is the other. Hybrid (L3) reconciles the two. For most mature orgs, the answer is hybrid; understanding top-down on its own is the foundation.
Pros and cons
PROS CONS──────────────────────────────────────────────────────────────────Fast (2-4 hours total) Misses team-specific dynamicsLeadership-driven (uses growth plan) Less accurate for individual teamsEasy to update each period Top number is aggregate, not granularCaptures planned events Bottom-up has team-level rigorDefensible at the org level Hard to argue with team leads who see different patternsWhen top-down alone is enough
GOOD FIT Early-stage company without per-team data Org-level reporting only (board, investors) Stable cost trajectory; no major team-specific changes Quick directional forecasts for planning conversations Crawl- or Walk-stage maturity (per M5.1)When top-down misses
POOR FIT Major team-specific changes (one team scaling 3×) New product launches with different cost shape than historical Multi-quarter migrations spanning teams Acquisitions where historical trajectory doesn't predict future Architectural changes (e.g., moving from EC2 to Lambda)For these scenarios, augment with bottom-up (L2). The combination is hybrid (L3).
The growth-rate input
The growth rate is the most-leveraged input to a top-down forecast. Getting it right matters more than any other single decision.
SOURCES of growth rate: Leadership's stated revenue growth target (anchor) Historical CAGR (calibration check) Customer growth projections (if cost scales with customers) Workload expansion plans (engineering roadmap)
EXAMPLE input synthesis: Leadership target: 30% YoY revenue growth Historical CAGR: 25% Customer plan: 28% Workload plan: +15% from new launches
Synthesized growth rate: ~30% YoY (= 7% quarter-over-quarter) Confidence: medium (range 25-35% defensible)Safety margin sizing
The safety margin absorbs uncertainty. Too small and forecasts miss; too large and budgets become slack.
HORIZON TYPICAL MARGIN──────────────────────────────────────────────────────────────────1 month ahead 5-8%1 quarter ahead 10-12%6 months ahead 15-18%1 year ahead 20-25%The margin grows with horizon. A 12-month top-down with no margin is forecast theater; communicate the actual uncertainty.
Allocation to BUs/teams
EXAMPLE allocation, $1.55M Q4 forecast:
Historical proportions (last 4 quarters): engineering-platform: 35% engineering-product: 40% engineering-data: 20% shared-services: 5%
Allocated forecast: engineering-platform: $543K engineering-product: $620K engineering-data: $310K shared-services: $78K Sum: $1.55M ✓The allocation is a starting point. Teams that disagree (e.g., engineering-data has a planned migration not in the historical proportion) can adjust via bottom-up reconciliation.
Common mistakes
MISTAKE FIX──────────────────────────────────────────────────────────────────Use just last quarter as baseline Use rolling 4-12 quarters;(noisy) smooths outliers
Apply same growth rate to all teams Some teams scale faster; adjust per-team factors
No safety margin Always include; communicate the band
Ignore planned events Quantify launches, migrations, M&A explicitly
Forecast made then forgotten Re-forecast monthly; variance analysis at quarter-endTools for top-down
Pull the historical baseline from your billing data: 12+ months of monthly spend from Cost Explorer, Cost Management or the GCP billing export, or from ZopNight’s Cost Reports trend. Apply the growth rate and planned events in a planning sheet, then split the total per team by historical share (Cost Reports, Teams tab, gives the shares) or by custom weights. ZopNight’s charts project only to the end of the current month; the quarterly and annual forecast lives in your sheet. AWS Cost Explorer and Azure Cost Management also offer a built-in forecast you can use as a cross-check.
For commitment portfolio planning (RIs, SPs), top-down forecasts feed commitment sizing (M5.7).
2. Demo
A clean top-down Q4 forecast:
ORG: B2B SaaS, $1.2M/mo current run rate (Q3 2026)
INPUTS: Q3 2026 actual: $1.2M/mo Growth rate: +8% (leadership plan, calibrated against 6% CAGR) Planned events: Black Friday traffic spike: +$50K (one-time over 2 weeks) New product launch (October): +$30K/mo ongoing Safety margin: 12% (quarterly horizon)
COMPUTATION: Base growth: $1.2M × 1.08 = $1.296M/mo + Black Friday (amortized): +$16.7K/mo for Q4 ($50K / 3) + Product launch: +$30K/mo ongoing Subtotal: $1.343M/mo Safety margin (12%): +$161K/mo
Q4 FORECAST: $1.504M/mo (committed) ($1.343M + $161K margin)
ALLOCATION: Historical proportions × $1.504M: eng-platform (35%): $526K/mo eng-product (40%): $602K/mo eng-data (20%): $301K/mo shared-services (5%): $75K/mo
Sum: $1.504M ✓
PRESENTED to leadership: "Q4 cost forecast: $1.343M/mo, range $1.18M - $1.50M (±12%) Drivers: 8% baseline growth, Black Friday, October launch. Recommend budgeting at the top of the range, $1.504M/mo (the committed figure above, margin included)."The forecast is built in ~2 hours of work and is defensible enough for a board slide.
3. Hands-on (5 min)
Build a top-down forecast for your next quarter:
CURRENT QUARTER ACTUAL (or YTD run rate): $__________ /mo
GROWTH RATE INPUTS: Leadership target: ____% YoY Historical CAGR: ____% Synthesized rate: ____% (per quarter or per year)
PLANNED MAJOR EVENTS (with dollar impact): __________ +$__________ __________ +$__________ __________ +$__________
SAFETY MARGIN: ____% (10-12% for quarter horizon)
TOP-DOWN FORECAST: $__________ /mo
ALLOCATION to teams: Team A (__%): $__________ Team B (__%): $__________ Team C (__%): $__________ Sum check: $__________
CONFIDENCE BAND: $______ to $______If you cannot fill in the growth-rate inputs, the forecast isn’t ready. Source the inputs from leadership before committing the number.
4. Knowledge check
Q1
Top-down forecasting is best for:
A. Org-level forecasting, stable trajectory, quick turnaround
B. Highly granular per-team forecasts
C. Per-resource detail
D. New workloads that have no cost history behind them
Show answer
Correct: A. Top-down is the right tool when you need a defensible org number in a few hours and the per-team detail can come from a separate bottom-up exercise.
Q2
A 12% safety margin in a top-down forecast accounts for:
A. Profit
B. Cloud rate hikes only
C. Forecasting uncertainty; major events you didn’t quantify, unexpected growth surprises, cost-rate changes, normal variability
D. Currency movement over the period, which can move the reported total by several percent in either direction indeed
Show answer
Correct: C. Without the margin, forecasts that come in over budget look like overspend; with margin, normal variability is absorbed and only true overruns surface.
Q3
Top-down combined with bottom-up (hybrid):
A. Worse than either alone
B. Only useful at the org level
C. Same as top-down
D. Better than either alone
Show answer
Correct: D. See L3 for the reconciliation process. Top-down provides org context and growth assumptions; bottom-up provides team-level rigor and planned-event detail; reconciliation surfaces discrepancies and improves both.
5. Apply
Build the top-down each quarter in your planning sheet, using 12+ months of history from your billing console or ZopNight’s Cost Reports. Calibrate growth rates against actuals each quarter: your synthesized rate should improve over time.
Related lessons
- L2: Bottom-up forecasting (next)
- L3: Hybrid and reconciliation
- L4: Forecast accuracy
- L5: Communicating uncertainty
- T5.M5.4: Budget governance