Why a budget that shows only cost is incomplete
A phased budget answers what marketing will spend, on which lines and in which months. The budget-building tutorial also asks for an expected outcome and a measure on every line, and for margin-aware targets, but it stops before pricing the outcomes as contribution or saying when that contribution should arrive. The question a finance partner asks, what did we get for it, is then left to be answered from actuals.
The retrospective half already exists. The marketing P&L explained builds marketing's own profit and loss view from the ledger after a period closes, the marketing P&L template lays out its lines, bases and variance notes and gives each line a budget column, without a method for filling it, and the comparison of ROI and ROAS bridges attributed revenue to a return on contribution with every cost in it. This tutorial builds the same view before the period starts: expected outcomes, margin, cost and timing. Because it uses the same bases, and the same lines where the actual view has them, a closed month can be set beside it with little translation, and that is the reason to build it.
It also gives the first disagreement the marketing finance pillar names between the two functions, which is timing, somewhere to be addressed: the planned view is where the spend, the expected return and the lag agreed with finance are written down together, so the lag is already on the page before the first review.
Three questions stay elsewhere. How large the budget should be belongs to the budgeting methods guide. Which attribution approach credits revenue to marketing belongs to the analytics root's attribution approach guide. The margin figures, the accounting treatment of revenue and the way the company forecast takes in marketing's plan belong to finance. This page uses all three as inputs and does not decide any of them.
Before you start
- A phased budget with outcome targets. The draft or approved budget, with each cost line phased by month, fixed, semi-fixed and variable lines marked, and an expected outcome and measure on each line.
- The agreed margin. The contribution marginRevenue left after every cost that varies with each sale. for each product line or segment, from finance, with what it includes, as the contribution margin guide sets out, and the date it was last refreshed. Without it the plan cannot be finished: a plan on revenue or gross margin repeats the mistake that guide describes.
- The bases of the actual view. The period, currency, accrual or cash basis, attribution basis and people-cost convention that marketing's own P&L view uses or will use. If there is no actual view yet, agreeing these is the first step below.
- Marginal estimates or cautious ranges. For the main paid lines, from step tests, past reallocations or spend-response curves; the marginal returns and forecasting guide covers how marginal returns are estimated.
- The lag between spend and sale. Read from earlier customers or leads grouped by the month of first contact. If no such grouping exists yet, say so, and treat the lag as an assumption to be marked as such.
- The company's revenue plan. Total planned revenue for the same periods, from finance, so the share attributed to marketing can be shown against it.
- A spreadsheet with months across. One column for each period and one block for each case.
The steps
Agree the bases with finance before any number goes in
Write the first row of the view before anything else. It names the period and currency, says whether the figures follow accruals or cash, states how revenue is attributed and sets the convention for people costs. The planned view must use the same ones as the actual view, because its test is that a closed month can be set beside it without adjustment. If the actual view does not exist yet, write the bases now, as the template's opening row asks, with the people-cost convention its non-media section asks for, and have the finance business partner confirm them. Then write who owns what in a line each: marketing owns the outcome forecast and its assumptions; finance owns the margin figures, the accounting treatment and the company forecast the plan feeds. The finance business partner guide shows where such a split is kept. Record the date it was agreed.
Copy the cost side from the phased budget
Take every cost line as the budget phasesSpreading each budget line across the months it will actually be spent in, rather than dividing the year evenly, so actuals can be read against plan. it: media by channel, and the non-media lines, each with its cost centre. Keep contingency as its own visible line. Add a memo line for anything marketing causes that finance books outside marketing's cost centres, shown once and labelled, and not counted again where the agreed margin already takes it off, as the template's memo line explains. Do not re-estimate anything here; if the budget changes at a reforecast, the planned view takes the new version. Carry over the budget's cost type for each line, whether fixed, semi-fixed or variable, because the lower case later moves variable lines first, then semi-fixed ones. A commitment that runs past the year contributes only its in-year part to this plan's cost; the remainder is flagged, as the guide to committing money beyond the budget year sets out.
Choose the unit of outcome for each line
Give each line the outcome that converts to money with the least guessing. For an online sale that is orders; as general practice, where repeat purchase matters, new customers; for lead generation, the expected contribution of a lead at its pipeline stage, taken from the lead valuation guide, whose values are internal planning figures agreed with finance and never added to revenue; for a service business, booked jobs, as the service business unit economics guide sets them out. Lines whose return cannot be read month by month, such as brand work, are not forced into a contribution figure. Give them their own lines with the evidence and horizon agreed in advance, as the pillar advises, and see the brand against performance guide on measuring brand honestly for the evidence; as this page's own method, include their cost in the total and leave their contribution blank with the reason written beside it.
Forecast what each period's spend should buy, at marginal returns
For every paid line, use the marginal estimate for the step of spend the budget plans (the marginal return on ad spendExtra revenue returned by the last increment of ad spend. where the line has one), not the line's average, because in most channels the last share of a larger budget buys least. Give each line a range, a practice the guide to reading small numbers recommends where volumes are thin, carried as a low, a base and a high figure, and phase the result from the business's own seasonal history, which the annual planning guide describes reading. A new channel is an experiment: enter its cost, leave its forecast contribution out and, as general practice, write down what it should teach. Record the outcomes as earned by each period's spend. When they land is a separate question, and the sixth step answers it.
Price the outcomes on the agreed margin
Turn outcomes into revenue at the value of each after returns and discounts, on the attribution basis written in the first step, and show attributed revenue as its own line. Then apply the contribution margin for that product line or segment to reach contribution, not a gross margin that stops at the cost of goods; the template's line labelled gross margin is this same after-variable-costs margin. Where products differ, use a margin for each product line or segment rather than a single blended rate, since cost and return rate differ between products. As general practice, make sure returns and discounts come off once, either in the order value or in the margin, and write down which; the example takes them off in the order value. The chain from spend to outcome to revenue to contribution is one branch of a KPI tree; drawing it once shows finance how each channel figure reaches profit. As general practice, if a test has shown that the attribution basis over-credits or under-credits a line, write the adjustment as a stated assumption instead of quietly changing the basis; the incremental against platform ROAS comparison explains why the two differ. For lead lines, use the stage values from the third step as expected contribution and keep them out of the revenue line.
Place the contribution where the lag puts it
Place each period's expected outcomes on the business's own lag curve, its conversion lagThe time between an ad interaction and the conversion it leads to, which leaves the most recent reporting days incomplete., which the diagnosis of a missed marketing forecast shows how to read from cohortsA group of customers who share a starting event and period, tracked together over time. of customers or leads grouped by the month of first contact, using the cohort table the cohort analysis tutorial builds. One ad platform's own attribution settings are a reminder that response goes on after the spend. The diagnosis stops at spreading outcomes forward; this tutorial's own method adds that the same lag works in both directions, so the plan shows three lines separately. The first is the part of each period's contribution that lands within the plan, in the same period or a later one. The second is contribution carried in from spend made before the plan began, which the same lag produces in the opening months; shown separately, it stops the early periods looking better than the new spend deserves. The third is the part of this plan's spend that will land after the plan ends, so the year's spend is not judged on only part of what it earns. Mark the curve as measured or assumed and give its source.
Add cost and contribution and read the position by period
For each period, subtract total marketing cost from contribution to reach contribution after marketing and, as general practice, keep a running total beside it. If the template's new-customer view applies, show it too. Then read the position plainly: in which period the running total turns positive, which periods are negative by design and why. A plan whose opening periods spend before they earn should say so, and finance should have agreed that reading before the year starts, not met it in the first quarter's report. As general practice, rebuild one period by hand from its outcomes, the value of each and the margin, so that an error in the chain shows before finance finds it. The campaign break-even calculator, given the agreed contribution margin in the field it labels gross margin, shows the orders a period's cost must earn at the agreed order value, a quick check against that period's landed orders.
Show the gap to the company's revenue plan
Set attributed revenue beside the company's planned revenue for the same periods and show the remainder as its own unattributed line, not hidden, as the P&L guide recommends for the actual view. Both totals belong on the page. Where attributed lines add up to more than the company plan could carry, check, as general practice, for overlap between channels before questioning the plan, and use the bridge in the ROI comparison to reconcile the two totals. A plan that quietly assumes marketing produces more revenue than the company expects to make will be noticed by finance.
Build the lower and higher cases and list what must be true
Take the lower and higher scenarios from the budget and run the same chain for each, valuing what each case removes or adds at the marginal return rather than the average. Put the higher case's extra spend where returns should fall away slowest and say so. The scenario planning framework pairs each case with the condition it assumes and the trigger agreed for it in advance. Then list what must be true for the base case to hold. Change one input at a time, as the sensitivity table in the growth model tutorial does, and note which input moves the result most. Mark each input measured or assumed and give its source, its date and an owner; review the assumed ones first. Kaushik's measurement model sets a target against every key measure (KPIA metric tied directly to an objective.); the plan's outcome figures are that set of targets, and the list gives the reasoning behind each.
Agree it, save it as the baseline and read each month against it
Walk through the view with the finance partner in the monthly review, with the papers sent in advance and the decision recorded. Finance owns the company forecast and decides how the plan feeds it; marketing supplies the ranges and the assumptions. Save a dated version and leave it unchanged. At each close, read the actual view against it. Split the varianceThe gap between the budget, or the latest forecast, and the actual for a line, explained by its cause and marked as timing or permanent. into volume, rate and timing before it is explained, as the diagnosis of a missed marketing forecast does, then write each variance note in the tracking guide's terms: the line, the amount, whether the cause is timing or permanent and what it does to the full year. Allow for the lag before calling an early shortfall permanent. At each reforecastA rebuild of the full-year view of the budget from actual spend and what marketing now knows, done at the points finance sets; not a fresh budget., rebuild the planned view beside the budget, because the reforecast tutorial already asks for a revised outcome forecast at marginal returns; keep the earlier versions so a later review can tell which assumption failed and change the method.
The planned P&L at a glance
Each row is a line of the planned view, with columns for each period and each case. The rows match the lines of the actual view where it has them, so a closed month can be read beside it.
| Line | Contents | Source |
|---|---|---|
| Period and basis | Period and currency, accruals or cash, how revenue is attributed, people-cost convention, date agreed | Agreed with finance |
| Marketing cost | Media by channel and non-media lines, with cost centres, and contingency on its own line | The phased budget |
| Memo costs | Marketing-driven costs that finance books outside marketing's cost centres | The finance partner and the coding map |
| Outcomes earned | What each period's spend should buy, as a low, base and high figure | Marginal estimates for each line |
| Attributed revenue | Outcomes at their value after returns and discounts, on the stated basis | Order values and the agreed attribution basis |
| Contribution | Attributed revenue at the agreed contribution margin, or expected lead contribution | The margin from finance; lead values |
| Carried in | Contribution landing in the opening periods from spend made before the plan | Earlier spend and the lag curve |
| Landing later | Contribution from this plan's spend that arrives after the plan ends | This plan's spend and the lag curve |
| Contribution after marketing | Contribution less total marketing cost, for each period and as a running total | Calculated |
| Gap to company plan | Attributed revenue beside planned company revenue, with the remainder shown | Finance's revenue plan |
| Cases | The same lines for the lower and higher scenarios | The budget's scenarios |
| Assumptions | Each input marked measured or assumed, with source, date and owner | The line owners and finance |
What the planned view can and cannot say
The planned view is a plan. It states what the outcomes should be worth if the assumptions hold and shows which assumptions matter most; it does not promise a result, and its ranges should stay ranges when it is presented. Attributed revenue on it is revenue on an agreed basis, not the company's revenue, and expected lead contribution is a planning value, not revenue. Brand lines carry cost without contribution until their agreed evidence is in. It does not decide the size of the budget, the attribution approach or the accounting treatment of anything. As general practice, where finance draws the lines differently, change the view to match finance's lines rather than keeping two versions of the plan.
The finance reporting hub shows where the planned and actual views sit with the rest of the reporting routines.
A fire pit seller plans a year of marketing contribution (illustrative)
Skenuth Fire Pits builds its planned marketing P&L
Skenuth Fire Pits, an invented seller of steel fire pits for home patios, sells through its own shop and by mail order, with a marketing team of two. Finance has agreed a contribution margin of 30% on fire pit orders after product cost, delivery and card fees, and an average order value of £900 after returns and discounts. The bases are agreed first: calendar quarters, accrual basis, attribution on the basis the team already reports, no people costs. The approved budget is £300,000 of media, phased £60,000, £120,000, £70,000 and £50,000 across the four quarters, and £52,000 of agency fees, photography and tools phased evenly at £13,000 a quarter, £352,000 in all, with no contingency held, the media lines marked variable and the other lines fixed.
Marginal estimates from last year's step tests give the base orders each quarter's spend should earn, with low and high figures beside them in the sheet: 330, 560, 330 and 220, 1,440 in all. Cohorts of first contact from last year show that 70% of the orders a quarter's spend earns land in the same quarter and 30% in the next, a curve the team marks as measured from four cohorts. Last year's fourth-quarter spend earned 200 orders, so 60 land in this year's first quarter. Orders landing are therefore 291, 491, 399 and 253, 1,434 in all; another 66 orders from this year's fourth-quarter spend land next year, which is the landing-later line.
| Row label | Q1 | Q2 | Q3 | Q4 | Year |
|---|---|---|---|---|---|
| Marketing cost | £73,000 | £133,000 | £83,000 | £63,000 | £352,000 |
| Orders landing from the same quarter's spend | 231 | 392 | 231 | 154 | 1,008 |
| Orders landing from the previous quarter's spend | 99 | 168 | 99 | 366 | |
| Orders carried in from last year's fourth quarter | 60 | 60 | |||
| Attributed revenue at £900 | £261,900 | £441,900 | £359,100 | £227,700 | £1,290,600 |
| Contribution at 30% | £78,570 | £132,570 | £107,730 | £68,310 | £387,180 |
| Contribution after marketing | £5,570 | −£430 | £24,730 | £5,310 | £35,180 |
| Running total | £5,570 | £5,140 | £29,870 | £35,180 |
Read plainly, the running total is positive from the first quarter, and only the second quarter, which carries the heaviest cost, is slightly negative on its own. Against a company revenue plan of £1,900,000, attributed revenue of £1,290,600 leaves £609,400 shown as unattributed. Without the carried-in orders the first quarter would show a loss of about £10,600, which is the new spend's own result; the 66 orders landing next year are worth about £17,800 of contribution that the year's spend earns but this plan cannot claim. The lower case takes £40,000 from the second quarter's media. The step tests suggest the marginal order there is about £400 of media against £270 of contribution per order, so the case loses 100 orders, raises contribution after marketing by £13,000 to £48,180 and sends that marginal figure to the marketing head and the finance partner as an assumption to retest in April. The higher case adds £30,000 in the third quarter, where the team assumes returns fall away slowest, at an assumed £500 an order: 60 more orders, £16,200 of contribution and £13,800 less in the year.
Changing one input at a time shows that order value and margin move the year most: an order value 10% lower, or a margin of 27% instead of 30%, would each take about £38,700 off it, against about £15,100 for 10% fewer orders earned by the second quarter's spend, the quarter whose orders matter most, which is why order value and margin are written down with their sources. A slower lag, 60% in the same quarter and 40% in the next, moves about £6,200 from the second quarter to the third and takes only £540 off the year, because the extra orders carried in from last year, 80 against 60, nearly offset the extra orders pushed into next year, 88 against 66. Finance decides how the plan feeds the company forecast. At the close of the first quarter, 262 orders have landed against 291 with cost on plan; the team splits the gap by volume, rate and timing, finds that the cohorts still have orders to come, and leaves the plan unchanged until finance's next reforecast.
Common mistakes
- Building the plan on revenue or gross margin. Outcomes priced at either overstate what marketing leaves, which is the error the contribution margin guide exists to prevent; price them on the margin finance agreed.
- Using the average return for new spend. The extra money buys less than the line's average, so a plan built on averages is too high before the year starts; use the marginal estimate and a range.
- Booking each month's sales against that month's spend. Sales from spend keep arriving afterwards, so the early periods look worse and the later ones better than they are; place contribution on the observed lag.
- Leaving out carry-in and landing-later lines. Without them the opening periods are credited with work from before the plan and the closing spend is judged on only part of what it earns.
- Forcing brand work into a contribution figure. A short-horizon return invented for activity that works slowly cannot be checked; give it a line with its agreed evidence and horizon instead.
- Hiding the gap to company revenue. A plan whose attributed lines exceed what the company expects to make will be noticed by finance; show both totals and the remainder.
- Keeping assumptions in someone's head. An input not marked measured or assumed, with source, date and owner, cannot be reviewed when a month misses.
- Editing the baseline to match the result. Save a dated version, leave it alone and record every change at a reforecast with its reason, so a later review can learn from the miss.
- Keeping a second version of the plan. A planned view that does not reconcile with the lines finance uses may be set aside; align it with finance's lines and bases, and keep one current version.
