Most annual planning guides teach you the order of the phases and stop. You finish knowing that alignment comes before budgets and budgets come before pressure-testing, but with no idea when to start, how long anything takes, what to actually write down, or how to tell whether the plan you produced is any good. This one has the calendar, the worksheets, and the arithmetic.
It is written for companies between roughly five and a few hundred people — founder-led businesses, agencies and professional services firms, SaaS companies, and owner-operator businesses. If you have never run a formal planning process, start at the top. If you have, skip to the calendar and the worksheets.
Five principles that make the difference
Before any mechanics, the handful of things that separate a plan that changes behavior from a document nobody opens again.
- 1Debate it while it is being built. Commit once it is locked. A plan that is still being renegotiated in March is not a plan, it is a suggestion. The argument phase is real and necessary — and it has an end date.
- 2Define what you will not do. Guardrails are more useful than goals. A team that knows it cannot hire outside two cities, or cannot exceed a travel number, makes better decisions than a team given only a target.
- 3Make the targets stretch but reachable. A target nobody believes is a target nobody works toward. If your team privately thinks the number is fiction, you have not set a goal, you have removed one.
- 4Uncertainty is not an excuse. You will not know what next year holds. Name the specific uncertainties, attach a mitigation to each, and move — see the scenario section below, which turns this from a slogan into a mechanism.
- 5Plan for the company you will be, not the one you are. Budget the hire in the month you will actually make it, not January, and not never.
Who owns the budget
Budget ownership shifts as a company grows, and most planning friction comes from getting this boundary wrong — either an owner still approving every line at 60 people, or department leads handed targets they had no part in shaping.
The useful rule: the person closest to a metric sets the granular target for it. Executives own the company-level outcomes. Department leads own the drivers underneath. Staff own delivery.
| Level | Owns | Sets |
|---|---|---|
| Founder / CEO / Finance | Company outcomes | Revenue, margin, cash flow, financing needs, and the guardrails: hiring rules, travel and expense limits, headcount per group, bonus pools |
| Department leads | The drivers | Employee productivity assumptions, hiring needs and timing, departmental expense budgets, incremental asks with justification |
| Department staff | Delivery | Project staffing, individual and project-level expenses, the actual work against the metrics |
What planning looks like at your size
Do not run a 200-person planning process at 12 people. Each stage has a defining question, and the answer determines what is worth building.
- 1–5 people — "Is there demand for what we sell?"
- A pricing and revenue plan, a simple target, a hiring plan, and a view on whether you need outside capital. One spreadsheet. Annual planning at this size is a half-day with a co-founder, not a process.
- 6–50 people — "Will we be the one that wins?"
- Preliminary long-range planning, employee productivity assumptions, a growth strategy, KPI tracking against benchmarks, standardized pricing, and quarterly budget-vs-actuals. This is where the process below starts earning its keep.
- 50+ people — "Are the unit economics improving?"
- Monthly budget-vs-actuals, a formalized long-range plan, department-level planning with real bottoms-up models, acquisition planning if relevant, and profit-sharing or bonus programs tied to the plan.
The calendar
This is the part almost every planning guide omits. Below is a week-by-week schedule for a calendar-year company. If your fiscal year ends elsewhere, shift everything so that "final targets locked" lands about six weeks before year-end.
The total elapsed time is about sixteen weeks. That sounds like a lot until you have tried to compress it into three, at which point you discover that the pressure-testing round is where all the value is and it cannot be rushed.
| When | What happens | Who owns it | Output |
|---|---|---|---|
| Aug, weeks 1–2 | Refresh the long-range plan. Where do we want to be in three to five years, and what does next year have to deliver for that to stay possible? | Founder + exec team | A one-page multi-year view |
| Aug, weeks 3–4 | Close out the current year's read. Full-year forecast, what worked, what missed and why. No planning until this is honest. | Finance | Current-year landing estimate |
| Sep, week 1 | Executive offsite. Agree 3–5 themes and 2–5 company goals for next year. | Exec team | Themes and goals, written down |
| Sep, week 2 | Finance translates goals into fiscal targets: revenue, margin, cash, headcount envelope. | Finance | First-pass targets |
| Sep, weeks 3–4 | Set guardrails and hold back the executive contingency. Issue preliminary budgets to department leads, net of contingency. | Finance + exec | Budgets issued |
| Oct, weeks 1–3 | The solve. Department leads build bottoms-up models against their targets and identify risks, opportunities, and incremental asks. | Department leads | Bottoms-up models + asks |
| Oct, week 4 | Executive review, round one. Approve, reject, or send back each ask. | Exec team | Round-one decisions |
| Nov, weeks 1–2 | The solve, round two. Rework where round one said no. Expect one to two rounds total — if you are on round four, the original targets were wrong. | Department leads | Revised models |
| Nov, week 3 | Build the scenarios: base, upside, downside, with trigger points and pre-committed responses. | Finance | Three cases + trigger table |
| Nov, week 4 | Lock the plan. Final targets issued. The debate is over. | Exec team | Locked plan |
| Dec, weeks 1–2 | Cascade to staff. Individual goals, project staffing, and department-level communication of what was agreed. | Department leads | Individual goals set |
| Dec, weeks 3–4 | Load the plan into your accounting system as the budget, so January's budget-vs-actuals works on day one. | Finance | Budget loaded |
Building the revenue target
Revenue targets are the backbone of the plan and where most of the assumptions live. The objective is not to pick a number you like. It is to work out what would have to be true for that number to happen — and then decide whether you believe those things.
Every revenue model reduces to volume times price. What changes by business type is what "volume" is made of.
Revenue = Billable headcount × Utilization × Billable hours available × Effective rate
Utilization is the share of available hours that are billable. Effective rate is what you actually collect per hour after discounts and write-offs — not your rack rate.
Ending ARR = Starting ARR + New + Expansion − Contraction − Churn
Build each of the four movements separately. A plan that models net growth as a single percentage is not a plan, it is a wish with a decimal point.
Revenue = Traffic × Conversion rate × Average order value × Purchase frequency
Worked example: doubling billable revenue
Say a consulting firm did $4.2M last year and wants $8.4M. Rather than writing "2x" on a slide, work the chain backwards.
- Target revenue: $8,400,000
- Billable hours available per consultant per year: 2,080 − 240 (holiday, PTO, sick) = 1,840
- Target utilization: 72% — up from 68% last year, and that increase is itself an assumption you must defend
- Effective rate: $185/hour, after an assumed 8% average discount off a $200 rack rate
- Revenue per fully-ramped consultant:
1,840 × 0.72 × $185 =$245,088 - Consultants required at full ramp:
$8,400,000 ÷ $245,088 =34.3 → 35 people - Currently on staff: 19. Gap: 16 hires.
Now the part that kills naive plans: a hire made in September does not deliver a year of revenue. Apply a ramp.
Effective annual capacity = Months productive in year ÷ 12 × Full-year capacity
Months productive = 12 − start month − ramp months. A consultant starting in month 4 with a 2-month ramp delivers 6/12 of a year, not 9/12.
Run that across a realistic hiring schedule and 16 hires spread through the year typically deliver somewhere around half of their full-year capacity. Which means hitting $8.4M needs either earlier hires, higher utilization, a rate increase, or a smaller target. That is the conversation the plan exists to force, and it is the one you never have if you write "2x" and move on.
Worksheet
Worksheet 1 — Revenue build
Fill this in for each revenue stream separately, then sum. One combined build hides the stream that is actually failing.
- Revenue stream namee.g. retained consulting, project work, licensing
- Prior year actual$
- Target for next year$
- Volume driverheadcount / customers / traffic
- Volume assumptionthe number, and why you believe it
- Price or rate assumptioneffective, not list
- Utilization or conversion assumption% — and last year's actual for comparison
- Calculated revenuevolume × rate × utilization
- Gap to target$ and %
- What has to be true for this to happenthe three biggest assumptions
- What would make it failname the risk
Building the headcount plan
People are the largest line in almost every small company, and headcount planning deserves more attention than any other part of the budget. Three archetypes drive it, and each is a different conversation.
- Service delivery
- Driven by the revenue target and productivity assumptions. This is arithmetic — see the revenue build above.
- Administration
- Driven by complexity, not revenue. Billing volume, entity count, compliance surface, internal management load. Grows in steps, not smoothly.
- Sales and marketing
- Driven by your go-to-market strategy and pipeline coverage. The hardest to get right, because the payback lags the cost by two to four quarters.
Loaded cost = Base salary × (1 + payroll tax rate + benefits rate) + equipment + software seats + variable comp
A practical rule of thumb: multiply base by 1.25–1.35 for a US employee. Sales roles with commission run higher.
The three cost levers
- Hiring pace. Every month a budgeted role stays open is a month of salary saved — and possibly a month of revenue lost. Budget start months honestly rather than assuming January.
- Attrition. You will lose people. A plan assuming zero attrition is wrong. Assume a rate based on your own history and decide in advance which departures you backfill.
- Location. Where you hire changes cost materially. If this is a lever you are willing to pull, make it an explicit guardrail rather than a case-by-case negotiation.
Worksheet
Worksheet 2 — Headcount plan
One row per planned hire — including backfills. Roles without a start month always end up costed as if they started in January.
- Role title
- Department
- Typenew / backfill / conversion
- Planned start month
- Ramp time to full productivitymonths
- Base salary$
- Loaded costbase × 1.25–1.35
- Cost in plan yearloaded × months employed ÷ 12
- What breaks if we do not hire thisthe honest answer
- Prioritymust-have / should-have / if we beat plan
Building expense budgets
Non-people expenses are smaller but they are where discipline is visible. Split them into three buckets and budget each separately, because they behave differently.
- Internal operations — software, IT, rent, team meals, company events, insurance, professional fees.
- Sales development — conferences, sponsorships, sales meals, advertising, content.
- Client-related — travel, reimbursable and non-reimbursable client costs, project materials.
Budget these as guardrails, not entitlements. A department given a number and the freedom to spend it as they see fit will make better decisions than one that has to seek approval per line — provided the number is genuinely a limit.
Worksheet
Worksheet 3 — Expense budget by department
One sheet per department. The last two columns are what make it a plan rather than a wish list.
- Department
- Expense category
- Prior year actual$
- Next year request$
- Change vs prior year$ and %
- Driver behind the changeheadcount / volume / price / new initiative
- What we would cut first if we had toname it now
- Recurring or one-time
The executive contingency
The single best tactical idea in corporate planning, and one most small companies have never heard of. Before you issue budgets, strip out a reserve and hold it centrally. Departments receive targets net of that reserve and never see it.
It does two jobs at once. It stops leaders sandbagging — if everyone assumes there is slack in their number, everyone pads, and you end up negotiating against padding rather than reality. And it gives you dry powder to release when the pressure-testing round surfaces a genuinely underfunded team or a real new opportunity.
Contingency = 5% to 10% of planned operating expenses
A widely used range rather than a measured benchmark. Toward 5% for a predictable business with stable revenue; toward 10% when you are entering a new market, changing pricing, or carrying real uncertainty in the revenue plan. It does not have to be spread evenly across departments.
If nobody needs it, you do not spend it — it absorbs overspend elsewhere, cushions a revenue miss, or simply becomes profit. There is no version of this where holding it back hurts you.
The solve: pressure-testing the plan
Top-down targets meet bottoms-up reality here, and this phase is the reason the process is worth running at all. Department leads take their allocated budget and build a model showing how they would actually deliver against it — then report back what is achievable, what is not, and what they would need.
The loop: targets issued → owners build bottoms-up models → they surface risks, opportunities, and asks → executive review approves or rejects → revised targets issued. Expect one to two rounds. If you are on round three, the original targets were not defensible and you should say so out loud rather than grinding the team through another cycle.
Worksheet
Worksheet 4 — Department solve submission
The form each department lead returns. Standardizing it is what makes the executive review round take two hours instead of two days.
- Department
- Target as issuedrevenue and/or expense
- Bottoms-up resultwhat the model actually produces
- Gap$ and %
- Top three assumptions in the model
- Biggest risk to hitting itand the mitigation
- Biggest opportunity not in the plan
- Incremental ask$ and what it buys
- What we deliver with the ask vs without itthe trade, quantified
Scenario planning
Most planning guides say "flag uncertainties and provide mitigation plans" and leave it there. Here is how to actually do it.
Build three cases off the same cost structure, varying only the revenue assumptions you identified as least certain. Then — and this is the part that matters — decide now what you will do in each case, and what specific signal triggers it.
| Case | Revenue assumption | Trigger to declare it | Pre-committed response |
|---|---|---|---|
| Downside | ~80% of plan | Q1 revenue more than 10% below plan, or two consecutive months below | Pause all should-have hires; cut discretionary marketing; hold contingency; re-forecast cash |
| Base | Plan | On or near plan through Q1 | Execute as written; release contingency only against approved asks |
| Upside | ~115% of plan | Q1 more than 10% above plan and pipeline supports it | Pull forward the contingent hires; release contingency into capacity ahead of demand |
Writing the response down in November is worth more than the forecast itself. In April, when you are 12% behind, the argument about whether to slow hiring is emotional, political, and slow. If the answer was agreed in advance, it is just a decision you already made.
Worksheet
Worksheet 5 — Scenario and trigger plan
- Scenario namedownside / base / upside
- Revenue assumption$ and % of plan
- Which assumption is being flexedbe specific
- Trigger — the observable signala number and a date, not a feeling
- Who declares it
- Actions we commit to nowlist them
- Resulting operating income$
- Resulting cash position at year end$ and months of runway
Checking your plan against benchmarks
Before you lock anything, sanity-check the shape of the plan. These are widely circulated rules of thumb, not laws — the point is that if you are far outside a range you should know why, and be able to say it in a sentence.
Runway (months) = Cash on hand ÷ Average monthly net burn
Net burn, not gross — cash out minus cash in. If your plan ends the year with under six months of runway and no financing event, that is not a plan, it is a countdown.
Revenue growth rate (%) + Profit margin (%) ≥ 40
A shorthand for whether growth justifies the cost of getting it. Below 40 is common and survivable; it is a discussion prompt, not a verdict.
Burn multiple = Net cash burned ÷ Net new ARR added
How many dollars you burn to add a dollar of recurring revenue. Under 1.5 is generally regarded as strong, 1.5–2 as reasonable, above 3 as a signal that growth is being bought rather than earned.
CAC payback (months) = Sales & marketing spend ÷ (New customers × Monthly gross profit per customer)
Commonly targeted under 12 months for small-business customers and under 18–24 for enterprise. Longer paybacks are viable but they consume cash first and return it much later.
Revenue per employee = Annual revenue ÷ Average full-time headcount
The fastest way to see whether a headcount plan is realistic. Compare against your own prior years first — the trend matters more than the absolute number, and the absolute number varies enormously by business model.
| Metric | Services / agency | Subscription software | E-commerce |
|---|---|---|---|
| Gross margin | 30–55% | 70–85% | 25–50% |
| Payroll as % of revenue | 45–65% | 35–55% | 15–30% |
| Utilization (billable staff) | 65–80% | n/a | n/a |
| Months of cash reserve to hold | 3–6 | 6–18 | 3–6 |
Living within the plan
A locked plan is not a static one. The discipline that makes annual planning worth the effort is a monthly cycle where the plan stays fixed and your forecast moves.
That requires three columns, and understanding the difference between them is the whole idea.
- Plan
- Locked in November. Never changes all year. This is what you said you would do.
- Outlook
- Your current best forecast for the full year, revised every month as actuals land. This is what you now think will happen.
- Actuals
- What happened. Comes from your accounting system, which is why the books being current is a precondition for any of this working.
Compare all three pairings. Actuals vs Outlook tells you whether you can forecast. Actuals vs Plan tells you whether you are on track. Outlook vs Plan tells you what the year is now going to look like — and it is the one most companies never look at, despite it being the one that should change decisions.
The monthly cycle
- 1
Evaluate results
Close the month, actualize the budget with what happened, and diagnose where you over- and under-performed against plan. This requires the books to be closed on time — a variance report on incomplete books is worse than none, because it looks authoritative.
- 2
Identify implications
Given year-to-date, what does the shape of the rest of the year look like? A 5% Q1 revenue miss is not a 5% annual miss if it compounds through a sales cycle — work it through rather than annualizing the gap.
- 3
Revise business plans
Decide what changes: what gets cut, what gets accelerated, whether a trigger has fired. Update the Outlook. Leave the Plan alone.
Worksheet
Worksheet 6 — Monthly variance review
Run this on the same day every month, within a week of close. The cadence matters more than the sophistication.
- Month
- Revenue: plan / outlook / actual$
- Gross profit: plan / outlook / actual$
- Operating expenses: plan / outlook / actual$
- Operating income: plan / outlook / actual$
- Cash at month end$ and months of runway
- Three largest variances vs outlook$ and why
- Is this timing or is it real?the question that matters most
- Implication for the full yearrevised outlook
- Decisions made this monthand who owns each
- Any trigger fired?from Worksheet 5
Worksheet
Worksheet 7 — Annual planning readiness checklist
Run through this before you lock. If you cannot tick an item, you are not ready to lock — you are ready to argue about it, which is better done now.
- Prior-year results are closed and honest
- 3–5 themes and 2–5 company goals agreed by the exec team
- Revenue built bottoms-up per stream, with named assumptions
- Headcount plan has a start month and loaded cost for every role
- Every hire tagged must-have / should-have / contingent
- Expense budgets issued as guardrails, net of contingency
- Contingency sized and held centrally
- Every department completed a bottoms-up solve
- Every incremental ask has a stated trade
- Three scenarios built with triggers and pre-committed responses
- Plan checked against benchmarks; outliers explained
- Year-end cash and runway modeled in all three cases
- Plan loaded into the accounting system as the budget
- Monthly variance review scheduled for the whole year
The precondition nobody mentions
Every worksheet above depends on one thing: knowing what actually happened. Prior-year actuals by department, revenue by stream, payroll as a share of revenue, real cash position. If your books are three months behind, you are not planning — you are guessing with a spreadsheet, and the plan will be wrong in ways you cannot see.
The monthly cycle has the same dependency. A variance review needs a closed month, within a few days of month end, every month. That is the operational thing most small companies cannot reliably do, and it is why plans quietly stop being referenced by March.