| name | budget-creation-management |
| description | Creates and manages operating budgets — annual plans, quarterly refreshes, and rolling forecasts at department-level granularity, with variance tracking and budget-vs-actual analysis. Use when the user mentions "create a budget," "annual plan," "OpEx run-rate," "departmental budget," "budget vs actuals," "re-forecast," or asks about "allocating budget to departments." |
| version | 2.0.0 |
| author | Crewm8 |
| maintainer | Gokul (github.com/gokulb20) |
| license | MIT |
| homepage | https://crewm8.ai |
| tags | ["cfo","finance","budgeting","opex","variance-analysis","financial-planning"] |
| related_skills | ["revenue-forecasting","headcount-and-comp-planning","cash-forecasting","scenario-planning","forecast-accuracy-tracking"] |
| inputs_required | ["historical-actuals","headcount-plan","revenue-forecast","strategic-priorities","runway-target"] |
| deliverables | ["annual-budget","budget-vs-actual-tracker","variance-commentary","runway-projection"] |
| compatible_agents | ["hermes","claude-code","droid","cursor","windsurf","openclaw","openai","generic"] |
Budget Creation & Management
Build and maintain the company's operating budget — annual plans, quarterly refreshes, and rolling forecasts. Budget by department, track variances, and manage budget vs actuals. Goal: every dollar has an owner, a purpose, and a plan.
Purpose
Every startup dollar is scarce and every expense needs justification. This skill ensures budgets are built from scratch (zero-based), aligned with strategic priorities, and managed actively through variance tracking and re-forecasting. It solves the problem of budget drift — where last year's spend becomes next year's baseline without questioning whether it's still the right allocation.
When to Use
- "Create our annual budget"
- "Build a departmental budget"
- "What's our OpEx run-rate?"
- "Review budget vs actuals"
- "Re-forecast based on new information"
- "Allocate budget to departments"
Inputs Required
- Historical actuals — P&L by department (monthly, last 12–24 months).
- Headcount plan — current and projected from
headcount-and-comp-planning.
- Revenue forecast — from
revenue-forecasting (drives variable costs).
- Strategic priorities — what the company is trying to achieve (drives spend allocation).
- Runway target — months of cash to target (drives total spend envelope).
Quick Reference
| Concept | Description | Key Ratio |
|---|
| OpEx envelope | Total spend available after gross margin minus target burn | Revenue × Gross Margin % − Target Monthly Burn |
| Zero-based budgeting | Build each budget from scratch, not from last year's numbers | Don't use "last year + 20%" |
| Department benchmarks | Typical % of total OpEx by stage | R&D 30-60%, S&M 15-45%, G&A 10-25% |
| Variance threshold | Flag when actuals deviate from budget | > 10% or > $5,000 |
| Contingency | Holdback for unknowns | 10-20% of total budget |
Procedure
1. Set the Top-Down Envelope
Revenue: $X (from revenue-forecasting)
Gross Margin: $Y (revenue × margin %)
Target Burn (monthly): $Z (based on runway target)
Available OpEx: $Y - $Z
The total OpEx must fit within the available envelope. If it doesn't, either: grow revenue faster, improve margins, cut burn, or raise more money.
2. Build Department Budgets (Zero-Based Approach)
Don't just add 20% to last year. Build from scratch:
For each department:
- What are you trying to achieve? (OKRs, KPIs)
- What resources do you need to achieve it?
- What's the minimum viable budget? (must-do)
- What would an extra 50% get you? (growth lever)
- What would you cut if cash got tight? (triage list)
Work with the department head:
- Review last quarter's actuals.
- Map headcount to hiring plan.
- Project non-headcount spend (use run-rate for recurring, specific for one-time).
- Flag any step-function changes (new office, major conference, tool migration).
Department Budget Template
| Line Item | Jan | Feb | Mar | Q1 Total | Q2 | ... | FY Total |
|---|
| Headcount | | | | | | | |
| Salaries + Benefits | | | | | | | |
| Contractors | | | | | | | |
| Recruiting | | | | | | | |
| Travel & Entertainment | | | | | | | |
| Non-Headcount | | | | | | | |
| Software & Subscriptions | | | | | | | |
| Marketing Spend | | | | | | | |
| Events & Conferences | | | | | | | |
| Professional Services | | | | | | | |
| Office & Facilities | | | | | | | |
| Equipment | | | | | | | |
| Training & Development | | | | | | | |
| Other | | | | | | | |
| Total Department Spend | | | | | | | |
Department Benchmarks (% of Total OpEx)
Typical SaaS startup allocation:
| Department | Early (pre-PMF) | Growth (post-Series A) | Scale (Series B+) |
|---|
| R&D / Engineering | 40-60% | 35-45% | 30-40% |
| Sales & Marketing | 15-25% | 30-40% | 35-45% |
| G&A | 15-25% | 10-20% | 10-15% |
3. Consolidate & Stress Test
- Does the total budget produce the right runway?
- Are department allocations in line with benchmarks?
- Is headcount growth sustainable? (Can recruiting actually hire this fast?)
- Are there any "cliffs" (large one-time expenses in a single month)?
4. Budget Approval & Lock
Present the consolidated budget to the CEO/board. Once approved, lock it as the plan of record.
5. Ongoing Management
Monthly:
- Budget vs actual by department.
- Flag variances > 10% or > $5,000.
- Each department head explains material variances.
Quarterly:
- Re-forecast based on actuals and new information.
- Adjust the remaining months, don't re-litigate the past.
Output Format
- Annual budget (monthly × department)
- Budget vs actual tracker
- Variance commentary by department
- Runway projection based on budget
Done Criteria
The skill is complete when:
- The top-down OpEx envelope is set and aligned with runway targets.
- Each department budget is built zero-based with department head input.
- The consolidated budget is stress-tested against hiring velocity and cash constraints.
- Variance thresholds are defined and tracking mechanisms are in place.
- The budget is approved and locked as the plan of record.
Pitfalls
- "Last year + 20%" incremental budgeting — copying the prior period and scaling by a flat percentage destroys accountability and misses structural shifts in the business. Always zero-base.
- Treating all costs as fixed — modeling cloud hosting, payment processing, and variable marketing as fixed line items leads to gross margin surprises when revenue scales.
- Building budgets without department head input — a budget imposed by finance that the VP of Sales never reviewed won't be owned, won't be accurate, and won't be respected.
- Ignoring hiring velocity constraints — budgeting for 5 new hires per month when recruiting's actual throughput is 2 creates a phantom underspend. Cap hiring projections at realistic throughput.
- No contingency line — every startup budget needs a 10-20% holdback for unknowns. A budget that allocates every dollar to the penny is a budget that will be broken by the first surprise.
Heuristics
- Build for the most likely case, have a contingency for the worst case: 80% confidence budget + a 10-20% contingency line.
- Hiring is almost always slower than planned: a hiring plan that says "hire 3 engineers per month" will probably do 1.5. Model this.
- Non-headcount expenses are stickier than you think: cutting SaaS tools or travel budgets takes 2-3 months to show up. Plan accordingly.
- Marketing spend is the most flexible lever: it's easy to turn on/off. Model it as a variable, not fixed, line item.
- Budget owners must buy in: if the VP of Sales didn't help build the sales budget, they won't feel accountable to it.
Edge Cases
- Hypergrowth: if growing > 10% MoM, monthly budgets can't keep up. Use rolling 3-month forecasts.
- New departments: when adding a function (e.g., first sales hire), don't just add salary — add tools, travel, events, onboarding costs.
- Office moves / expansions: often the largest single line item. Model as a project budget separate from the recurring OpEx.
- Currency exposure: if paying employees/vendors in multiple currencies, build in an FX buffer (3-5%).
Verification
Can you answer "What's our total OpEx spend this quarter vs budget?" for each department? Can you explain every variance > 10%? Are the assumptions in the budget still valid given what you know today? If not, the budget is already stale.
Example
-
User: "Create our annual operating budget for next fiscal year, starting with the revenue forecast, then build departmental budgets zero-based."
Expected behavior: You load the revenue forecast, set the top-down OpEx envelope, then walk through each department (R&D, S&M, G&A) building from headcount outward. Outputs a monthly × department budget with benchmarks and a runway projection.
-
User: "Review Q2 budget vs actuals and tell me why we're off."
Expected behavior: You compare budget vs actual line items by department, flag material variances (>10% or >$5k), surface commentary from each department head, and highlight patterns — e.g., hiring running at 60% of plan, marketing overspend from an unplanned conference.
-
User: "Re-forecast the rest of the year based on what actually happened in Q1."
Expected behavior: You take Q1 actuals as the new baseline, adjust remaining months' projections (don't re-litigate Q1), update runway math, and flag any assumptions that need revising — e.g., slower hiring pace, higher AWS costs.
Linked Skills
- Revenue projections →
revenue-forecasting
- Headcount & comp assumptions →
headcount-and-comp-planning
- Cash impact →
cash-forecasting
- Scenario modeling →
scenario-planning
- Track budget accuracy →
forecast-accuracy-tracking