A marketing plan can ask for more work than its people, money or time can support. If the team says yes to every channel, its budget does not become more generous; the same capacity is spread more thinly. A marketing budget makes those choices visible. It describes what the planned work is expected to cost, when resources will be needed, who can approve them and how the plan will be reviewed against what actually happens.
You do not need to invent a percentage of revenue or copy another business's spending to begin. This chapter starts with an objective, names the work required to pursue it and asks whether the organisation can responsibly fund and deliver that work. You will make a blank, evidence-labelled first budget for one real offer. The diagrams and worksheet deliberately contain no fictional business, customer, price, result or benchmark.
Figure 1
Figure 1 in words: Situation evidence and Objectives give the work a purpose. Strategy and the 7Ps narrow the choices; Tactics and Action identify tasks, owners and costs. A budget checks whether money, time and people can support those tasks. Control compares the plan with actual evidence and may change the next plan. This is an original teaching synthesis of the planning links discussed in Chapters 12 and 13.
What does a marketing budget actually contain?
At its simplest, a budget is an approved plan for using resources over a stated period. It is not a guarantee that spending will cause a result. A useful marketing budget names the scope of work, the planned cost of each activity, the timing, the owner, the source of the estimate, the amount authorised to spend and the point at which a decision will be reviewed. OpenStax places action programmes, budgeting and controls together in a marketing plan: the planned work has costs, owners and dates, and those costs should be monitored (Gomez Albrecht, Green and Hoffman, 2023a).
Be careful about the word marketing. A business might call only its paid advertisements a marketing budget, while another includes research, content, tools and staff effort. State what is inside your particular budget and what is accounted for elsewhere. Otherwise two apparently comparable figures may cover different work. A campaign budget for one short period is also different from a whole-organisation annual marketing budget. In this chapter, you will build a small plan for one offer and one review period; the method can later be expanded.
There are related but distinct documents. The budget is the approved or proposed spending plan. A forecast estimates what is now likely to happen based on newer information. Actuals are costs and outcomes recorded after activity occurs. A cash-flow view asks when money must leave the business, which may differ from when a cost is recognised in accounting. You do not need to master accounting to keep these labels separate. For a broader explanation, Dhruvi Infinity's existing article on budgeting and forecasting explores those concepts in a business-finance setting. Its examples are separate from this chapter and are not used here.
First fix the scope and the decision
Write the offer, the audience or customer task, the period and the business owner at the top of your sheet. A plan for “all marketing” cannot be costed accurately if no one knows which offers or tasks it covers. Choose one outcome from Chapter 4 or 12 and state its measurement rule. The objective is the change sought, such as an appropriately defined customer action, not simply a desired number of posts. If no baseline has been observed, write baseline unknown and make a first measurement part of the plan.
State an affordability boundary before ranking activities. This is the amount the authorised decision-maker says can be committed to this plan without assuming unverified future revenue. It may be zero at the moment. The boundary is a constraint, not a recommended industry percentage. Record whether it is an approved ceiling, a tentative working limit or a request that still needs approval. The organisation's wider cash obligations and ability to pay belong with the appropriate finance owner; a marketing worksheet should not silently take authority over them.
Then list the non-money constraints: available staff hours, skills, production time, approval, technology, service readiness and time for review. Something can have no invoice yet still consume the one person who must also support customers. “Free” distribution is not free production. The Digital Marketing Institute's budgeting lesson notes that an objective-and-task estimate may reveal a plan that the business cannot afford; the useful response is to change the tasks, timing or goal openly (Digital Marketing Institute, n.d.).
Figure 2
Figure 2 in words: Money is the approved or proposed spending ceiling; People are the hours and skills available; Time covers sequence, dependencies and review period; Evidence supports estimates and the later decision. Each limit can make a proposed activity infeasible even when another limit looks generous. The boxes are prompts, not claimed amounts.
Cost the work, not just the advertisement
Begin with the actions in the SOSTAC® plan. For each, ask what has to be produced, checked, distributed, delivered and maintained. An article can require research, writing, editing, images, accessibility checks, publication and later updates. A paid placement can require a landing page and service capacity as well as media spend. A practical audit from Chapter 13 may show that an offer or sign-up step needs work before any new promotion is appropriate. These are possible cost categories, not instructions to undertake every task.
Separate one-off setup from ongoing work. A first design or measurement setup may happen once; hosting, content maintenance, placement, customer support or licence charges may recur. Separate fixed commitments, which do not vary much with the number of activities in the period, from variable costs, which may rise as activity grows. A quote can cover a defined piece of work; an ongoing contract may have minimum terms or a notice period. Record the real terms rather than treating the first month's price as the entire cost.
Money is only one column. Estimate hours for internal tasks and name the person or role expected to do them. Hours are not automatically an extra cash payment, but they are capacity used. Avoid adding the same staff time twice if salary is already included elsewhere in the organisation's approved budget. State whether the marketing sheet shows an internal time estimate, a cash expense, or both. OpenStax's marketing-plan explanation puts the cost next to what is done, when and by whom (Gomez Albrecht, Green and Hoffman, 2023a).
Figure 3
Figure 3 in words: Setup cost needs a quote or estimate and a due date. Ongoing cost needs a rate, billing period and duration. Variable cost needs a rule for what changes it and a spending ceiling. Internal effort needs a role, hours and capacity check. List any existing commitment separately so it is not accidentally counted or ignored twice.
When you do not know a price, do not insert a plausible-looking number. Mark it quote required and assign someone to get the quote. When a service's price changes with volume, record the unit, the expected range if supported, the maximum you are willing and authorised to commit, and how overrun will be prevented. A working estimate can be useful, but its source and confidence should remain visible. “Estimated” and “approved” are different statuses.
The small arithmetic you need
You can build the first total without a complicated financial model:
Planned cash cost for the period = sum of one-off cash items due in the period + sum of recurring cash items due in the period + planned variable spend.
Keep a separate line for internal hours = sum of estimated hours across named tasks. The owner checks those hours against the time actually available in the same period. If a cost is paid later, mark its payment date; if a subscription continues beyond the plan period, record the commitment that remains. Use consistent currency, dates and tax treatment on the sheet, and ask the finance owner how figures should be recorded. The formula does not forecast sales and does not turn attention or clicks into money.
Figure 4
Figure 4 in words: Add one-off, recurring and variable cash items for the selected period; check whether the resulting plan is within the authorised cap. In a separate calculation, total the hours and compare them with available capacity. Note when cash is due and which commitments continue. All value fields are blank.
Choose a method: objective and task, then an affordability check
One practical method is objective and task: define the desired outcome, choose the work that plausibly supports it, estimate each task's cost and sum the requirements. OpenStax describes this for communications budgeting, alongside top-down and percentage-of-sales methods (Gomez Albrecht, Green and Hoffman, 2023b). The Digital Marketing Institute also explains the appeal and the uncertainty of connecting objectives to tasks and costs (Digital Marketing Institute, n.d.).
The word plausibly matters. An objective does not tell you exactly how many sales an article or advertisement will cause. Work backwards to specify the tasks and assumptions, but label any expected response as a forecast, not a guaranteed conversion. With no history, a first task may be measurement or customer research. You can plan a smaller test rather than pretend to know the cost of achieving a precise growth target.
An affordability limit can coexist with this method. First calculate what the desired plan would require. Then compare it with authorised money and capacity. If the need is larger, you have choices: narrow the audience or offer, simplify the activity, lengthen the period, gather evidence first, seek a properly supported increase in resources, or revise the objective. Do not quietly keep the original target while deleting tasks needed to support it. This is the difference between planning within a constraint and making the numbers appear to fit on paper.
Figure 5
Figure 5 in words: Define the outcome and its evidence; name tasks and estimate both money and hours; compare the requirements with authorised limits. When they do not fit, change scope, timing, tasks or the requested resources, then revisit the objective. A cost estimate alone never proves that the outcome will occur.
Other methods may appear in a business discussion. Percentage of sales starts from a chosen share of sales; it is easy to calculate but does not by itself show what tasks are required or what future revenue is safe to assume. Top-down starts with an allocation set by management; it provides a hard limit but may not reflect the work needed. Competitive parity tries to match a rival's spend, although their goals, mix, finances and true spending may not be known. The methods can help frame a decision, but none replaces costing your actual plan (Gomez Albrecht, Green and Hoffman, 2023b; Digital Marketing Institute, n.d.).
There is no universal number in this chapter for “what a small business should spend”. A percentage seen online is not permission to spend, and another company's results do not predict yours. This is especially important when money or staff time is tight. The first sensible allocation might be to make an offer usable and measurable, but the choice must come from your real Situation evidence and capacity rather than a rule invented here.
Prioritise the work when everything sounds useful
Return to the customer task and the promise checked with the 7Ps and 4Cs. Put proposed work into three decision groups: essential for a truthful and usable offer, a bounded activity worth testing, and work to defer. This is an editorial tool for a small plan, not a universal accounting classification. A necessary service fix may deserve priority over promotion, even though it produces fewer visible marketing posts.
Ask five questions of each item. Which objective or customer problem does it address? What evidence supports it? What is the full resource requirement: cash, hours, skill and maintenance? What needs to be ready first? What would you learn or protect by doing it? If the answers are unknown, the next funded action may be a check rather than a full launch. Keep “unknown” on the sheet until someone investigates it.
Figure 6
Figure 6 in words: Essential work keeps the offer and claims accurate and usable; a bounded test addresses a specific uncertainty within an approved limit; deferred work has a reason to wait and a condition for reconsideration. Assign an owner to each decision. These categories do not assert that any particular tactic belongs in one of them.
Avoid a fake precision score. Multiplying unverified “impact” and “confidence” ratings can create a convincing rank from guesses. If a team does use a scoring method, disclose the definitions and evidence; do not let the score overrule a required deliverability or customer protection check. Plain reasons are often better for a first budget: this task is needed to keep a promise; that task tests an uncertainty; the other has no owner or cannot be maintained yet.
Make a bounded test genuinely bounded
A test is an activity designed to learn a defined thing before a larger commitment. Write the question, audience or context, maximum authorised spend, staff time, duration, measure and decision point in advance. A test can still fail to answer its question when too few events occur or measurement is weak. In that case the honest conclusion is inconclusive, not “success” or “failure” by appearance.
Some spending is difficult to stop instantly: contracts, minimum terms, external production and approved commitments may continue after a campaign is paused. Put these in the test's full cost. A daily media setting does not necessarily equal the all-in project cap when design, tools and support are outside it. If a platform's billing or budget behaviour matters, confirm its current official documentation for the exact account and campaign before spending; this article does not assume that every platform's daily limit works in the same way.
Timing: the schedule and the cash are different questions
Draw a simple sequence. Research and terms must be checked before a claim is published. A landing page may need to work before a paid campaign points to it. A service team may need time to handle enquiries. Review needs to happen after enough meaningful observations can exist. If you schedule all costs in the first week but the available cash arrives later, the plan is not feasible simply because its total looks affordable.
Use four dates where relevant: when work starts, when payment is due, when the customer-facing activity is ready, and when the result will be reviewed. A quote can expire; a supplier can require an upfront payment; a recurring charge can continue after a test. Record these actual terms in your budget. Do not assume that forecast revenue from the activity can pay an earlier obligation. Ask the finance owner to check affordability and payment timing against the wider business plan.
Figure 7
Figure 7 in words: Each activity may have a preparation period, a payment due point, a customer-facing launch and a later review. The dates are blank. Check dependencies and the available cash before authorising a commitment, then allow enough time for the stated outcome to be observed.
Staff time has a timing problem too. A person may have enough hours across a month but not on the one day an approval or live response is required. A contractor may be available for production but not for later changes. Review the peak load and the continuing load. Chapter 12's Action stage should name a real owner and a handoff, not just a role on a spreadsheet.
Measure spending and results without pretending to know causation
At review, keep four columns distinct: approved budget, actual cost, updated forecast for remaining work, and a note explaining any difference. A variance is the difference between a planned and actual figure under the same scope and period. An underspend is not automatically success: work may have been delayed. An overspend is not automatically waste: a quote may have changed or an approved new task may have been added. Investigate the reason and record who approved a change.
Use measures that fit the objective. The spending line says what was paid. An activity measure says what was produced or distributed. An intermediate response might record a relevant action such as a qualified enquiry. A customer or business outcome says whether the desired task or result happened. A guardrail checks that the work did not create a serious service or clarity problem. Define the source, period and counting rule. OpenStax positions control as examining performance and making adjustments when the plan differs from expectations (Gomez Albrecht, Green and Hoffman, 2023a).
Figure 8
Figure 8 in words: Approved budget is the authorised plan; actual is recorded spend; updated forecast estimates remaining expected cost; the explanation records what changed, why and who decided. Compare like periods and like scope. The table contains no actual amounts.
Do not calculate a confident return on investment (ROI) simply by dividing all revenue after a campaign by its advertising bill. Revenue is not necessarily incremental profit caused by the activity, and the bill may omit production and service cost. To make a defensible financial claim, you would need an appropriate definition of gain, a relevant cost base, a time period and a way to compare what likely would have happened otherwise. Attribution is especially hard when multiple channels and customer touchpoints contribute. CIM cautions that ROI is useful in some contexts but too narrow as the only measure of marketing effectiveness (CIM, 2025).
This does not mean “ignore numbers”. It means match the strength of the conclusion to the evidence. You can say that a defined cost was spent and a defined outcome was recorded. You can identify a plausible contribution and an alternative explanation. You should not promise that a click caused a sale, or that every sale in a review window came from one tactic, without an appropriate design and data to support the claim.
Figure 9
Figure 9 in words: The budget records resources used; activity records what the team did; response records a relevant customer action; outcome records the chosen customer or business result; guardrails check quality and delivery. The ladder does not assert that each lower step caused the higher one. Keep sources and limitations attached to the review.
When should the plan change?
Set the review rhythm before work begins. Check a broken purchase route promptly; a longer-term outcome may need more time. A single day of noisy data is not a fair test of a longer plan. Define a decision rule in ordinary language: if the cost exceeds the authorised boundary, pause new commitments and seek a decision; if the customer route fails, address the experience before expanding reach; if the response is unclear, examine evidence before reallocating.
The decision need not be “increase spend” or “stop everything”. You can continue within the cap, correct a page, change an audience choice, investigate a measurement issue, or defer a tactic. Record the date, actual evidence, alternative explanations, owner and next review. A budget is useful because it makes a later change explicit, not because it forbids change. Dhruvi Infinity's broader budgeting and forecasting guide can help when you want to understand the distinction in the wider business finances.
Try it: a first budget for one real offer
Allow about 65 minutes for a first pass. The worksheet is a planning record, not an authorisation to purchase anything. Use actual quotes, internal capacity and the approved limit when available; write unknown where they are not. Keep longer evidence notes outside the one-page summary.
Scope and authority (7 minutes). Name the offer, one customer task, the period, the owner and whether the limit is proposed or approved. If no amount is authorised, mark it accordingly.
Objective and gap (8 minutes). Name the outcome and its source or missing baseline. Check the most important offer or delivery gap from Chapter 13.
Tasks and dependencies (12 minutes). List only work needed for the chosen direction. Give each task an owner, prerequisite, start and review point.
Cash and hours (12 minutes). For each task, enter a verified quote or labelled estimate, payment timing, ongoing commitment and internal hours. Mark missing prices for quotation.
Compare with limits (8 minutes). Sum the plan period's cash items and hours separately. Check the proposed total against approved money and actual capacity. Do not use hoped-for future sales as approved cash.
Choose and defer (8 minutes). Identify essential work, a bounded test if appropriate, and tasks that must wait. Record the reason and the condition that would change each choice.
Review rules (10 minutes). Set actual-cost recording, outcome and guardrail definitions, a review date and who decides whether to continue, revise or stop a new commitment.
Figure 10
Figure 10 in words: The top records offer, objective, period, owner and authorised or proposed limit. Blank task rows ask for cash estimate and source, internal hours, payment date or continuing commitment, and decision status. The bottom separates total planned cash from available capacity and provides space for the priority, review evidence and decision owner. It contains no made-up amounts.
Read your sheet aloud as a decision: “For [real offer and task], we propose [named work] because [evidence]. It needs [verified or clearly estimated cash] and [available hours] over [period]. The authorised limit is [approved amount or still pending]. We will review [measure and guardrail] on [date].” If the brackets cannot be filled honestly, make collecting that information the next action. You do not need a perfect forecast to make a smaller, responsible choice.
Common mistakes to catch before spending
Starting with a channel split before the problem and objective are known. Calling unpaid staff effort free.Using an invoice price without setup, review or continuing costs.Assuming an advertisement's daily setting caps the total project.Counting forecast sales as current cash.Keeping the same outcome target after removing the tasks that could support it.Treating an underspend as proof of effectiveness.Calling observed revenue a return caused by one tactic. Each error can be corrected by adding a source, date, owner, scope and explicit decision.
The article and worksheet stand on their own. Public YouTube descriptions and metadata were inspected on 14 November 2026; neither selected video was played in full or transcript-reviewed. Check relevant content, spoken language, captions, UK availability and embedding before publication. Their demonstrations and dollar-based examples are not used in this article.
See a first budget explained:How to Create a Marketing Budget — Even If You've Never Made One Before!, The Seasoned Marketer (2025). The description covers a fixed amount, an end-of-month allocation and a percentage-of-revenue method, with sections through 06:27. It is an independent small-business explanation. Its suggested percentages or amounts should not be transplanted into your plan.
Explore a more detailed calculation:Calculate Digital Marketing Budget (For Beginners), Grace Leung (2023). Description chapter markers start with a target KPI at 00:49 and reach a media-budget step at 08:27, then measurement at 12:39. It is an independent practitioner walkthrough focused on a forecast for paid activity; cost-per-click and conversion assumptions must be rechecked against your own evidence. Neither video is an academic research paper. The academic reading above is OpenStax.
What to remember
A marketing budget connects an objective and its tasks to approved cash, actual staff capacity and timing.
Price the whole task, including setup, maintenance, review and continuing commitments. Mark unverified estimates clearly.
Objective and task gives a reasoned requirement; an affordability check may then force a smaller scope or revised objective.
Keep approved budget, actual cost and updated forecast distinct. Review the customer outcome and service guardrails, not just spending or clicks.
No universal spend percentage or automatic ROI claim can replace evidence about this offer and this organisation.
Further watching: The Seasoned Marketer (2025) ‘How to Create a Marketing Budget — Even If You've Never Made One Before!’, YouTube, 26 February. Available at: https://www.youtube.com/watch?v=Bl6UX7oNHEM (Accessed: 14 November 2026). Grace Leung (2023) ‘Calculate Digital Marketing Budget (For Beginners)’, YouTube, 6 December. Available at: https://www.youtube.com/watch?v=t2s5H98Pi7k (Accessed: 14 November 2026).
A beginner-friendly route through 26 short lessons on customers, offers, channels, content and measurement. Start anywhere, with no account or paid tools.
Learn the difference between features and benefits, write a promise customers understand, and check which kinds of online value you can honestly deliver.