A sound 2027 marketing budget for a small business starts with a revenue target and works backwards: how many customers you need, how many leads that takes, what each lead can cost, and only then how to split money between paid ads, SEO and AI-search content, and the fixed costs of staying visible. As a starting frame, many small businesses do well with about 60 to 70 percent in channels that produce demand quickly (usually paid media), 20 to 30 percent in compounding channels (content, SEO, GEO) and 10 percent in testing. These are our own planning ratios, not market averages, and they shift with sector and cash position.
October to December is when budgets for the next year are drawn up, and when most are drawn up badly, by taking last year’s number and adding ten percent. Below is a method that takes one afternoon, produces numbers you can defend, and leaves room to change course in the first quarter.
Step 1: Start from revenue, not from marketing
Write down the revenue you want from new customers next year, separate from repeat business. Then work backwards using your own numbers:
- Average first-order value or deal size.
- Close rate from qualified lead to customer.
- Share of leads that are qualified.
As a simple illustration (made-up figures): to win €120,000 of new business at an average deal of €4,000 you need 30 customers. At a 25 percent close rate that means 120 qualified leads. If 60 percent of enquiries are qualified, you need about 200 enquiries a year, roughly 17 a month. If a customer’s gross margin is €1,600, you can spend at most that on acquiring them and still break even on the first sale; a sensible ceiling is a third to a half of margin, so €530 to €800 per customer or about €130 to €200 per qualified lead. Now you know the price your marketing must beat.
Step 2: Know what each channel can realistically deliver
Take actual numbers from this year’s accounts: cost per lead by channel, lead quality, and time to result. If you have no data, use ranges from your own tests and flag them as assumptions. Rough characteristics of the main channels:
| Channel | Speed | Cost behaviour | Role in the plan |
|---|---|---|---|
| Search ads (Google, Microsoft) | Days | Pay per click, scales with spend | Capture existing demand |
| Social ads (Meta, LinkedIn, TikTok) | Days to weeks | Pay for reach, creative-dependent | Create and nurture demand |
| SEO and blog content | 3 to 9 months | Fixed monthly cost, compounding | Lower long-run cost per lead |
| GEO (AI-search visibility) | 3 to 9 months | Mostly content and structure work | Be quoted when buyers ask assistants |
| Email and CRM | Immediate to weeks | Low marginal cost | Convert and retain what you already have |
For a worked paid-media split at one budget size, see how to structure a €10k monthly budget across channels, and for first-time advertisers, where to put your first €5,000.
Step 3: Split the budget by time horizon
The most useful split is not by channel but by how quickly money returns:
- Harvest (about 50 to 60 percent). Channels that bring leads this month: search ads, retargeting, email to existing contacts. These pay the bills.
- Build (about 25 to 35 percent). Channels that lower next year’s cost: content, SEO, GEO work, reviews, case studies. Cut this and you will feel it in the second half of the year.
- Test (about 10 percent). New channels, new offers, new creative. Treat it as tuition: you expect most tests not to work and a few to change the plan.
If cash is tight, shift towards harvest, but do not take build to zero. If you already have strong organic visibility and plenty of demand, tilt towards build and test. Our overview of what AI-search work costs may help you size the build line.
Step 4: Add the costs people forget
- Creative production. Ads need fresh creative several times a year; budget for photography, video and design, not just media.
- Tools and tracking. Analytics, call tracking, consent management, email platform, CRM, reporting.
- Landing page and site improvements. A modest ring-fenced amount for conversion work often beats the same money spent on more traffic.
- Management fees. Freelancer or agency costs, or the real value of your own time.
- A contingency of 5 to 10 percent. Platforms change, a competitor moves, a good test needs extra fuel.
Step 5: Phase the year, don’t spread it evenly
Few businesses have flat demand. Map your seasonality and weight spend accordingly. A rough pattern for many consumer and e-commerce businesses is a heavy Q4, a quieter Q1 where you invest in build work, and a second push in spring or early autumn. B2B services often peak in late January to March and September to November, with summer and December quiet. Hold back around 15 percent of the annual media budget in reserve, to deploy where results are strongest at the end of each quarter. Compare with this year’s calendar, and read our Q4 budget note to avoid repeating this year’s late scramble.
Step 6: Set review points and kill criteria
The difference between a budget and a wish is that a budget says when you will change it. Write these down now:
- Monthly: cost per qualified lead by channel against your ceiling; pipeline and revenue from new customers.
- Quarterly: move up to 15 percent of the budget from the weakest channel to the strongest, based on cost per customer, not cost per click.
- Kill criteria for tests: “If a test spends €1,500 without reaching X, we stop.” Naming the number in advance prevents slow bleeding.
- Content review at six months: compounding channels need patience, but not blind faith. Check whether rankings, citations and enquiries are moving, even slowly.
What if you cannot afford all three buckets?
Most small businesses cannot, and pretending otherwise leads to spreading money too thinly. A workable approach is to sequence rather than split. Spend the first quarter mostly on harvest while fixing the leaks (tracking, landing page, follow-up speed), because any later money goes further once those are sound. Start build at a modest level in the first quarter, even if it is a single good post a week or one new case study a month, so that results begin appearing by autumn. Add test money only after you can measure cost per qualified lead reliably. Do not run more than two tests at once on a small budget: you will not get a readable answer from either.
If the total is very small, say under €1,000 a month, concentrate on one paid channel that matches how your customers already look for you, plus email to your existing contacts, and put your own time into one piece of useful content per month. Narrow and consistent beats broad and thin.
Three common budgeting mistakes
- Percentage-of-revenue budgeting without a growth goal. “We spend 5 percent of revenue” repeats last year’s result. If you want growth, the budget has to follow the target, not the past.
- Funding only what is easy to measure. Paid ads give tidy numbers; content and reputation work pays back later and less visibly. Starving the second to feed the first makes the first more expensive within a year.
- Treating the budget as fixed. Plan quarterly checkpoints and keep a reserve. The businesses that adapt fastest usually do best, not the ones with the cleverest spreadsheet.
A one-page template to finish this week
Write one page with these lines: revenue target from new customers; customers needed; qualified leads needed; maximum cost per qualified lead; budget by harvest, build and test with euro amounts; quarterly phasing; reserve; review dates; and the two or three assumptions most likely to be wrong. Share it with whoever manages your marketing, and with your accountant. If an assumption turns out wrong in February, you will see it quickly and know exactly which line to change. That visibility, more than any particular split, is what makes a budget useful.