How to build a marketing budget around the work you need
Scope a marketing project, uncover recurring costs, compare proposals, and work through an illustrative budget with a downloadable planning worksheet.
In this guide
A marketing budget is a set of choices about what the business needs next, what it can afford to maintain, and what it is prepared to learn. A percentage or a package price cannot make those choices for you.
To build a usable marketing budget, define the next capability your business needs, price the work and its ongoing costs over a named period, and check that your team has time to make it useful. Compare proposals against that same scope. Then release additional spending when you have evidence for the next decision.
This guide walks through that process for a website, brand project, campaign, or ongoing support. You’ll find a hypothetical budget, an apples-to-apples proposal comparison, and a downloadable worksheet. The examples illustrate planning decisions; they do not prescribe a spending benchmark or predict a return.
Diagnose the constraint before choosing the channel#
If inquiries are slow, it is tempting to buy more promotion. First ask whether the business has a visibility problem, an explanation problem, or a follow-up problem. The symptoms can look similar while requiring very different work.
| What you observe | Question to investigate | Possible scope to consider |
|---|---|---|
| Few relevant people reach the site | Where would the intended audience look for this offer? | Audience research, distribution, or search content |
| People arrive but ask basic questions the site should answer | Is the offer understandable and credible? | Service-page copy, proof, or page structure |
| Inquiries arrive but stall | Is the next step clear, timely, and appropriate? | Intake and follow-up processes |
| Customers want to buy in a way the site cannot support | What capability is missing? | Ecommerce or booking implementation |
| Every marketing task starts from scratch | What can be made repeatable? | Templates, shared assets, or an operating routine |
These are diagnostic prompts, not conclusions. A small sample can mislead. Talk to customers, inspect the current path, and record what you know versus what you are assuming.
For a business that has not launched, the immediate constraint may be uncertainty about the offer. Coaching or a focused planning engagement can help turn values and ambitions into decisions before you commission a large set of assets.
Define the result you can actually commission#
You can commission a working website, a set of product pages, a campaign, or a research resource. You cannot make a future sales result certain by putting it in a creative brief.
Write the project outcome in terms of capabilities and acceptance criteria. For example: “A visitor can understand our two services, see relevant work, and send an inquiry that reaches the right person.” Then list what must exist to make that true.
Our Colorado Sauce Company work included labels, a Shopify rebuild, product pages, email flows, and recipe content. Those deliverables supported a particular business transition. They are evidence of scope, not a universal package or a spending benchmark for another company.
A practical scope names what is included, what is excluded, who supplies the inputs, and how the finished work will be checked. That is the foundation for a budget you can compare.
Turn a vague deliverable into an acceptance check#
“Contact form” can mean a form that looks finished, a form connected to a service, or a fully tested inquiry path. Specify the version you are buying:
A visitor can submit their name, email, service interest, and message from a phone or desktop. Required fields have labels and clear error messages. A successful submission reaches the designated inbox and gives the visitor confirmation. The business owns the form-service account and receives instructions for changing the recipient.
That description makes implementation, testing, and handoff visible. It also creates a boundary: connecting the form to a separate CRM or building an automated follow-up sequence would be additional scope unless agreed otherwise.
Apply the same test to “brand guidelines,” “email setup,” or “SEO.” Name the artifact or capability, the inputs it needs, and the observable conditions for accepting it. A ranking or revenue target can guide the business objective, but it should not replace a definition of the work.
Account for four kinds of cost#
The quoted design fee is only one part of a project. Use four categories so the less visible work has somewhere to appear.
Creation#
This includes research, planning, writing, design, development, revisions, and launch preparation. Ask whether the estimate assumes that you will supply finished copy and images. If those inputs do not exist, they are work someone still needs to do.
Operation#
Hosting, domain renewal, email software, booking tools, payment-related services, and maintenance may continue after launch. List each expected service, its billing frequency, its owner, and the basis of the estimate. Confirm current pricing with the selected provider rather than copying an old proposal.
Distribution#
A finished asset still needs to reach people. Printing, postage, paid media, event materials, and content distribution can have separate costs. Keep media spend separate from the work of planning or managing the campaign so you know what each amount pays for.
Internal effort and uncertainty#
Someone in the business must supply information, make decisions, review drafts, and learn the system. Record that effort even if it does not create a new invoice. Identify unresolved dependencies—such as photography or a product-data cleanup—rather than hiding them under a vague “miscellaneous” line.
A working formula is:
Planned cost for the period = creation + operation during the period + distribution + an explicit allowance for identified uncertainty.
Record internal time alongside that total. Treat any reserve as a planning choice based on the uncertainties of your project, not as a universal percentage everyone should use.
Keep a separate payment schedule. A three-month software allowance is not necessarily the amount due in those three months: an annual plan may require the full year’s payment upfront. Likewise, a project fee may fall due at agreed milestones. For each item, distinguish the cost you allocate to the planning period from the cash actually due, and flag taxes, production charges, or other amounts that still need confirmation. A comparison that mixes monthly and annual figures can make the cheaper-looking option misleading.
Work through an illustrative budget#
The numbers below are invented to show the method. They are not Antilo prices, supplier quotes, or market averages. Imagine a service business planning a focused website release and one small promotional test over a three-month planning period.
| Item | Illustrative cost allowance | What must be confirmed |
|---|---|---|
| Planning and page structure | $700 | Research depth and decisions included |
| Copy for the initial pages | $1,100 | Page count, source material, and revision process |
| Design and implementation | $2,200 | Features, platform, and testing scope |
| Photography or other approved assets | $400 | Whether usable assets already exist |
| Software and hosting for the period | $180 | Actual products, billing terms, and ownership |
| Distribution test | $500 | Audience, destination, and what is being tested |
| Allowance for identified unknowns | $420 | What may change and who approves its use |
| Total | $5,500 | A planning example only |
Separately, imagine the owner expects to spend 18 hours gathering information, reviewing work, and learning the editing process. That is a capacity requirement, even if it is not paid to a supplier. If those hours are unavailable, the scope, timing, or outside support needs to change.
If the budget must come down#
Do not cut every line equally. Protect the work that makes the primary customer action possible. In this example, the business might postpone the distribution test until the inquiry path is verified, reduce the initial page count, or use existing approved photography if it genuinely serves the page.
Each change has a consequence. Fewer pages may leave some services for a later release. Reusing assets may limit the visual direction. Postponing promotion may mean relying on existing referral channels initially. Write down the tradeoff so “saving money” does not silently become expecting the same work for less.
If the budget can increase#
Return to the constraint rather than adding features by default. Better source material or deeper content work may matter more than another visual effect. A handoff session may be more useful than a dashboard the team cannot maintain.
The point of the example is the allocation conversation. Replace every amount with an estimate appropriate to your scope and confirm it with the people responsible for the work. The $420 allowance is not permission to add any feature that arises; record which uncertainty it covers and who can authorize spending it.
Compare proposals on the same basis#
A lower quote can be a better fit, a smaller scope, or an estimate built on different assumptions. You need to know which before comparing totals.
Send each provider the same brief and use a comparison sheet:
| Question | What a useful answer contains |
|---|---|
| What will be delivered? | Named pages, assets, functions, or activities |
| What must we supply? | Copy, images, data, access, approvals, and deadlines |
| What revisions are included? | A defined review process and treatment of scope changes |
| What costs recur? | Services, maintenance, ownership, and billing arrangements |
| How will we accept the work? | Checks for content, functionality, and handoff |
| What happens after launch? | Support boundaries and update responsibilities |
| What remains uncertain? | Dependencies, assumptions, and decisions still needed |
Beware of interpreting silence as inclusion. “Website development” does not tell you whether writing, redirects, form delivery testing, or training are covered. Ask for those details in plain language.
For a brand project, clarify the formats and uses of the assets. For print, distinguish design files from production and distribution. For email, separate building flows from ongoing copy, campaigns, software costs, and reporting. The distinctions matter more than the label on the package.
A worked comparison: the lower quote and the fuller scope#
Suppose two providers quote for the same hypothetical five-page service website. Provider A quotes $3,000 and expects approved copy and images. Provider B quotes $4,200 and includes copy development and a handoff session. Neither proposal includes hosting or the same $500 distribution test.
Every amount and time estimate in this comparison is invented. The figures are separate from the earlier budget and are not provider prices or benchmarks.
| Cost for the same initial release | Provider A | Provider B |
|---|---|---|
| Quoted project fee | $3,000 | $4,200 |
| Writing needed to meet the agreed scope | $1,100 | Included |
| Handoff session needed by the team | $200 | Included |
| Same hosting allowance for the period | $180 | $180 |
| Same distribution allowance | $500 | $500 |
| Comparable planned cost | $4,980 | $4,880 |
| Illustrative owner time, recorded separately | 18 hours | 10 hours |
Provider A’s initial quote is $1,200 lower, but its comparable planned cost is $100 higher once the missing work is included. That still does not make B the automatic winner. A may have a better fit for the project, and a business with finished copy may not need the writing allowance at all. In that case, A’s comparable cost would be $3,880. Capability, relevant work, working style, and the quality of the scope still matter.
Do not turn an unpriced omission into a convenient zero. Mark it unknown until clarified. If one provider has not specified redirects, accessibility checks, or account ownership, the comparison is incomplete even when the arithmetic is correct.
Control changes without making the project rigid#
A project often reveals something useful that was not in the initial brief. The aim is to make the decision explicit before the extra work starts.
Keep a short change record with five entries: the request, why it matters, what it adds or replaces, its effect on price and timing, and who approves it. Distinguish a correction needed to meet an agreed acceptance check from a new requirement. “The form does not send messages” and “We now want a second form connected to our CRM” are different conversations.
If the new requirement is worthwhile, choose among three clear options: exchange it for lower-priority work, add budget and time, or put it into a later release. A revision allowance should not be the only place this decision lives; agree what revisions cover before the project begins.
Decide between a project and ongoing support#
A defined project suits work with a clear boundary: an initial website, a set of labels, or a campaign launch. Ongoing support suits recurring decisions and production, such as keeping content current, reviewing performance, or maintaining an email program.
A recurring arrangement still needs a scope. Specify priorities, capacity, the review rhythm, and how new requests affect existing work. Ask whether unused capacity carries over, what happens when a month’s work is blocked by missing approvals, and what you receive if the arrangement ends. “Unlimited marketing” is difficult to evaluate unless the practical limits are clear.
You can also combine the two: a focused first project followed by a smaller maintenance or improvement scope. Do not commit to a large publishing schedule before confirming who will provide the underlying information and approvals.
Make growth systems part of the budget#
Marketing creates work inside the business. An inquiry needs a response. A product update needs to appear in the store and any relevant communications. A finished article needs someone to check that its information remains accurate.
Choose a few processes you can sustain:
- A shared place for approved assets and current business information.
- A named person responsible for new inquiries.
- A repeatable way to review and approve content.
- A record of changes and the questions you want to investigate.
- A maintenance schedule tied to actual business changes.
A system does not have to be a complicated automation. A clear checklist and an owner can be an appropriate first version. Automating an undefined process may simply make its confusion happen faster.
Set review points before spending on expansion#
Choose what you will review when the first scope is in use. Separate completion from effectiveness: a form can function correctly while attracting the wrong inquiries, and an email can be delivered without answering the reader’s question.
A first review might ask whether the work was delivered and whether the team can use it. A later review can consider the quality of inquiries, customer questions, and other relevant observations. Choose the timing based on the business’s activity and decision cycle rather than promising a result by an arbitrary date.
Stage the work around decisions#
A release does not have to include every idea in the plan. For a hypothetical service launch, the sequence could look like this:
| Stage | What it funds | Condition for the next decision |
|---|---|---|
| Clarify | Offer, audience, source material, and scope | The team can explain who the service is for and what must be delivered |
| Build | Essential pages, evidence, and inquiry path | Agreed acceptance checks pass and a person owns follow-up |
| Test distribution | One defined way to reach the intended audience | The test has a destination, a spending limit, and a question it can help answer |
| Improve or expand | Work chosen from what the first release reveals | Evidence identifies a useful change and the team can maintain it |
For the distribution test, decide in advance what would make you stop, repair, or continue. If the destination breaks, pause the test. If inquiries are mostly outside your scope, investigate the message and audience before adding spend. If they fit but follow-up stalls, improve the response process. If there is too little activity to draw a conclusion, report that uncertainty; a quiet week is not proof that the channel cannot work.
Record where each inquiry came from when you can, whether it fits the service, and what happened next. Keep unknown sources marked unknown. Read those observations alongside the volume of relevant visits or other opportunities to respond. The marketing measurement guide explains why counts and denominators matter.
The budget then becomes a sequence of decisions supported by evidence, rather than a commitment to keep buying the same activity.
Build your own working budget#
Download the marketing scope and budget worksheet . Use it to record the outcome, cost categories, assumptions, internal effort, and decision points. Leave unknown numbers marked as unknown until you have a basis for them.
Before requesting a proposal, check that you can answer:
- What does the business need to do differently?
- What must be delivered to support that change?
- Which inputs already exist, and who supplies the rest?
- Which costs continue after the initial work?
- What can be deferred without breaking the primary customer path?
- What evidence will inform the next decision?
Questions about budgeting a first project#
Should every business spend the same percentage on marketing?#
A percentage alone cannot tell you what needs to be built or maintained. Use it, if relevant to your own planning, as a boundary to examine—not a substitute for a scope. A startup creating its first assets and a business maintaining an established program face different tasks.
Is doing the copy myself always cheaper?#
It may reduce outside writing costs, but it still takes time. Ask what “supplying copy” means: rough notes, a first draft, or approved text ready for design. If you can provide strong source material but need help shaping it, make that division explicit.
What if I don’t know what I need yet?#
Commission the decision before commissioning every possible deliverable. A focused planning conversation can identify the audience, the constraint, and the next useful scope. This project brief can help you prepare, even if several answers are still uncertain.
Antilo connects coaching, consulting, design, and implementation around the stage you are in. Tell us what you are trying to build or change , and we can work out a scope with clear responsibilities and room to learn.
