Starting a service business often means making several spending decisions before there is much history to guide you. A website, advertising and follow-up tools can all feel necessary, but buying everything at once does not guarantee a working customer journey.
I would build a marketing budget for a new service business around a focused test you can afford to learn from. Start with the service you can deliver, the customers you want and the cash available after operating needs. Then fund the assets and sales handling required to make that first test meaningful.
In this guide
Begin with capacity and cash flow
First, work out how much additional work the business can take on. A company with a few available appointments should not plan the same campaign as one seeking several long-term contracts. Capacity affects the offer and how much demand is useful.
Next, look at the contribution from a customer after direct delivery costs. Revenue alone can be misleading. A busy calendar may still leave little room to recover acquisition spending if the jobs involve expensive travel, labour or materials.
Consider payment timing too. A customer might agree quickly but pay later, while advertising and staff costs arrive now. Your test budget should not depend on immediate returns to cover essential operating expenses.
Set a limit you can sustain and a review date before launch. This gives you room to learn without turning each disappointing day into a crisis. It also helps you resist adding more channels simply because the first one has not yet produced enough evidence.
Separate launch costs from recurring costs
Launch costs may include a landing page, basic creative assets, analytics setup and CRM configuration. Recurring costs may include media spend, campaign management, software and sales time. A page built once should not be confused with a monthly media commitment.
Request a written breakdown so you can identify the minimum viable setup. You may already own suitable photography or a working website. Use those assets where they genuinely serve the customer journey. Do not spend the entire budget on a redesign if the immediate problem is missed enquiries.
Design one test you can interpret
Choose a priority service, a defined audience and a clear next action. Start with a channel that fits how customers buy. For urgent local searches, search advertising may be a candidate. A visual, considered service may benefit from a different approach. Compare Google Ads and Facebook Ads for service businesses before deciding.
A hypothetical £2,000 test might allocate £500 to setup, £1,000 to advertising, £300 to management and £200 to software and follow-up support. The split is a planning example; your existing assets and sales process will change the allocation. Your actual allocation should reflect the work already completed and the market you are testing.
Set learning goals before spending
Define what would make the test informative: suitable enquiries, contactability, quote requests and sales outcomes. Establish an acceptable acquisition cost based on your margin, while recognising that an early campaign may need improvement. Record the assumptions behind the target.
Schedule checkpoints for technical reliability and commercial results. Verify forms and routing immediately; evaluate conversion after enough opportunities have progressed. If results are weak, identify the failed step before increasing spend. See what a customer acquisition plan should include.
Increase the budget when the process supports it
Scale gradually when customer economics and delivery capacity justify more volume. Keep reserves for testing new messages and improving the page. Track cash collected alongside bookings so apparent demand does not hide cancellations or slow payment.
A manageable first test for a new business
Imagine a new local service company has a working website but no reliable way to record enquiries. The temptation may be to spend the entire budget on traffic. I would first set aside enough time and money to make the handover dependable.
Choose one service and explain it clearly on a relevant page. Add a form with useful questions, confirm that the answers are stored and assign an available person to respond. Decide how a quote and a won customer will be recorded.
Then run a bounded acquisition test. The aim is not to prove that every future campaign will succeed. It is to learn whether the offer reaches suitable people and whether the business can turn their interest into paid work.
At the review, distinguish technical problems from commercial questions. A broken form should be fixed immediately. An unclear offer may need better wording. A sales process that leaves enquiries unanswered needs attention before more traffic is bought.
Keep some flexibility in the budget for the improvement the evidence suggests. If everything is committed to media, you may have no capacity to fix the page or train the person handling calls.
Complete the customer acquisition plan form if you want help deciding what the first test should include. Tell us what already exists, which service you want to grow and what your team can realistically handle. That is a better starting point than copying another company’s budget.
Frequently asked questions
Should I spend a fixed percentage of revenue?
A percentage can be a planning reference, but a new business may have little revenue history. Cash capacity, customer economics and test requirements provide a more useful starting point.
Do I need paid ads immediately?
Not always. Referrals, partnerships and local visibility may be appropriate first steps. Choose activities you can deliver consistently and measure.
When should I raise the budget?
When suitable enquiries convert at a sustainable cost and your team can handle more. More spend should follow evidence, not frustration with slow progress.
Ready to work on this for your business?
If you would like me and the DIL team to help you put this into practice, fill out the customer acquisition plan form. Tell us where you are now and what you want to improve, so we can start with your situation.