Budget
How much should a local business spend on marketing?
Every answer you will find to this question is a percentage of revenue, and almost none of them can tell you where the percentage came from. This page gives you a method you can check, and is honest about the limits of the one real survey we can point you to.
How much should a local business spend on marketing?
There is no credible universal percentage for a local business, and the rules of thumb in circulation are mostly repeated between marketing blogs rather than measured. The method that does hold up is working backwards: decide how many new customers you need, divide by the rate at which enquiries become customers to get an enquiry target, then multiply by what an enquiry costs in your category to get a monthly figure you can test.
The CMO Survey put marketing at 9.0% of company revenue in its February 2026 report, but that sample is 308 mostly large US companies and it describes what they spend rather than prescribing what a clinic or a contractor should.
Why nobody can give you one number
A dental practice with two chairs, a roofing company covering four counties and a solicitor taking three cases a month have nothing in common on this question except the question.
What a business should spend depends on what a customer is worth, how many of them it can physically serve, how long the sales cycle runs, how much competition is bidding for the same attention, and whether it is defending a position or trying to take one. A single percentage cannot carry all of that, which is why every percentage you see is either wrong for you or right by coincidence.
Read the full breakdown: Why nobody can give you one numberHide the full breakdown: Why nobody can give you one number
There is also a stage problem. A business that has been visible for a decade and needs to maintain that position spends very differently from one opening a second location into a market that has never heard of it. Same revenue, same category, completely different correct answer.
So this page does two things. It gives you the arithmetic that works regardless of category, and it is straight with you about what the available published figures actually measure, which is not what most pages imply they measure.
If you want a shortcut: the number you can defend is the one you built from your own close rate and your own customer value. Everything else is a conversation starter.
About the rule of thumb everybody quotes
There is a percentage of revenue figure for marketing spend that appears in almost every article on this subject, usually attributed to a United States federal agency for small business. It is quoted with total confidence, in proposals, in blog posts and in podcasts.
We went looking for the page it comes from. There is not one that we could find. What there is instead is a long chain of marketing articles citing other marketing articles, with the government attribution passed along intact and the actual source missing at every step. It is one of the cleanest examples of a citation loop in this industry.
Read the full breakdown: About the rule of thumb everybody quotesHide the full breakdown: About the rule of thumb everybody quotes
So it is not on this page, and if somebody quotes it at you in a proposal, the reasonable response is to ask for the link. How they handle that question will tell you more about them than the number would have.
This is not pedantry. A percentage of revenue rule is doing something very specific in a sales conversation: it converts your revenue into a spend figure without anybody having to justify what that spend will produce. The method further down this page does the opposite. It starts from what you need to happen and works back to what that costs, which means the number can be argued with.
The one figure we can actually source
The CMO Survey is run out of Duke University's Fuqua School of Business. Its February 2026 report surveyed 308 marketing leaders at US for-profit companies, 97% of them at vice president level or above, fielded between the 7th and 29th of January 2026.
SourceThe CMO Survey, February 2026 (Duke Fuqua, n=308 US marketing leaders)
Here is the caveat that matters more than the numbers. That sample is senior marketing leaders at mid-size and large US companies, which is nothing like a four-person clinic or a family roofing firm. It is a description of what those organisations spend, not a recommendation for anybody, and treating it as a target for a local business is exactly the mistake this page is trying to prevent. Use it as context for the conversation and nothing more.
The method that actually holds up
This takes about an hour with whoever answers the phone and whoever does the invoicing. It produces a number you can defend in a meeting, which no percentage rule can.
Start with how many new customers you need, not with a budget
Be specific and be realistic about capacity. Twelve new patients a month. Four commercial roof jobs a quarter. Six retained cases a year. If you cannot serve more than a certain number, that ceiling is the start of the calculation, because buying demand you cannot fulfil is the most expensive mistake on this page.
Work out what one of those customers is actually worth
Not the first invoice, the whole relationship. A patient who comes twice a year for a decade and refers a family member is worth a multiple of one appointment. A one-off emergency call may be worth exactly one emergency call. This single number decides whether an expensive lead is expensive or cheap, and most businesses have never written it down.
- Average value of the first job or appointment.
- How often they come back, and for how long, on average.
- Whether referrals from that customer are common enough to count, and if so, how conservatively you want to count them.
Find the rate at which enquiries become customers
Count enquiries for one month and count how many became paying customers. If nobody has ever measured it, measure it before spending anything, because this is the multiplier that turns a lead target into a budget and guessing it wrong distorts everything downstream. It is also usually the cheapest thing to improve.
See the remaining steps: The method that actually holds upHide the remaining steps: The method that actually holds up
Convert the customer target into an enquiry target
Divide the customers you need by that rate. Needing twelve customers at a one in four conversion rate means forty-eight enquiries. This is the moment most businesses discover their real problem is the conversion rate rather than the budget, because the enquiry target it produces is implausible.
Price the enquiry target and sanity check it against cash flow
Multiply the enquiry target by a realistic cost per enquiry for your category. That gives a monthly media figure, to which you add whatever the work of running it costs. Then ask the question that actually decides it: can this be sustained for long enough to learn anything, typically several months rather than several weeks? A budget that runs out before the data arrives buys nothing at all.
Run it once and the number will look uncomfortable in one direction or the other. That discomfort is the useful output. It either tells you the growth target is unfunded, or it tells you the conversion rate is where the work is.
What an enquiry tends to cost in paid search
You need a cost per enquiry to complete the arithmetic. LocalIQ's Search Advertising Benchmarks 2026 aggregate their own customers' campaigns across Google Ads and Microsoft Ads, which makes them a starting reference rather than a market average or a figure anybody should promise you.
SourceLocalIQ, Search Advertising Benchmarks 2026
These are costs per lead, not costs per customer. Do not quietly convert one into the other: a lead becomes a customer at whatever rate your business actually converts, and that rate is the number from step three. Note too that these describe paid search specifically. Enquiries arriving through organic search, a Google Business Profile or referral do not carry a media cost, which is the whole argument for building those channels alongside.
One booked job, several channels
Budgets go wrong when they are allocated as though each channel works alone. Most booked work involves several touches, and the last one gets the credit in almost every report.
This has a direct budget consequence. The channel that appears most efficient in reporting is usually the one nearest the decision, which is branded search, retargeting or a direct visit. Reallocating budget toward those and away from whatever created the awareness is one of the most common ways a working programme is slowly dismantled by its own dashboard.
The practical defence is to keep one total that nobody can flatter: enquiries and booked jobs for the whole business, month by month. Channel level numbers are for diagnosis. The total is for judgement.
A useful annual question when you set the split: if paid media stopped tomorrow, how many enquiries would still arrive next month? A business that cannot answer that has built everything on rented ground, and the budget should probably shift some weight toward the parts it owns.
How agencies charge, and where each model points the incentive
None of these models is dishonest. Each one simply points the incentive somewhere, and the useful thing is to know where before you sign rather than to discover it in month seven.
A percentage of what you spend on ads
The fee moves with the media budget. The conflict is plain: the agency earns more when you spend more, so recommending a larger budget always pays them and recommending a smaller one always costs them. In a month when the right advice is to pull back, that advice is expensive to give.
- Ask how the fee behaves if you cut the budget in half.
- Ask what work the percentage covers when spend rises but the workload does not.
A flat monthly fee
The fee is fixed regardless of media spend, so advice to reduce spend costs the agency nothing and the budget conversation stays honest. The conflict moves to effort: once the fee is set, the quiet incentive is to spend fewer hours on the account, which is how accounts drift into neglect without anybody deciding to neglect them.
- Ask what is included each month and what triggers an additional charge.
- Ask how you will see the work, not just the results.
A base fee plus a smaller share of spend
The hybrid softens both problems and inherits a little of each. Whether it behaves like the first model or the second depends entirely on which part dominates, so the question is not whether it is hybrid but what the ratio is.
- Ask which portion is larger at your current budget.
- Ask what happens to the ratio as spend grows.
Paying on results
This sounds like perfect alignment and is the hardest of the four to get right. It requires both sides to agree what a result is and to trust the same party to count them, and it pushes work toward whatever is easiest to attribute rather than whatever is most valuable. It also strains badly when leads arrive and the business cannot convert them, because the agency is owed for work the client considers worthless.
- Agree the definition of a qualified lead in writing before anything starts.
- Agree who counts, in which system, and what happens when the two disagree.
Who owns what, and what happens if you leave
This is the question most businesses want answered before any of the arithmetic, and it usually gets a vague reply, so here is the direct one. Everything should belong to you. The Google Ads account, the analytics property, the tag manager container, the domain, the website, the Google Business Profile, the ad creative and the content. The agency is granted access to those things and that access is removable.
Google Ads supports several access levels, including admin, standard, read only, billing and email only, managed from the account's Access and Security settings. An agency working through a manager account links to your account rather than containing it. Keep at least one admin login of your own, on an email address that is not controlled by anybody you might one day part with.
Read the full breakdown: Who owns what, and what happens if you leaveHide the full breakdown: Who owns what, and what happens if you leave
What this protects is accumulated work. Years of conversion data, negative keyword lists, data segments and learned bidding signals sit inside the ad account. Historical traffic sits inside the analytics property. If those were never yours, a change of agency means starting from nothing while paying somebody to rebuild what you already bought once.
On exit terms, the reasonable position is a stated notice period, no penalty for leaving once it has run, and a handover that transfers access rather than requiring you to ask for it. Long initial commitments are not automatically predatory: some work genuinely cannot show results inside a quarter. But a long commitment combined with ownership sitting on the agency side is a combination worth refusing.
We publish our own prices rather than quoting on request, which you can read on the pricing page. That is a position, not an industry norm, and it exists because the most common complaint in this market is being unable to compare two proposals.
Questions worth asking before you sign anything
None of these are aggressive questions. They are the ones a well run agency expects and answers immediately, which is precisely what makes them useful.
- Who owns the ad account, the analytics property, the tag manager container and the website, in writing?
- What is the notice period, and is there an initial commitment before it applies?
- On the day this ends, what do I keep and what do I have to ask for?
- Exactly what is in scope each month, and what counts as an extra charge?
- Is there a setup fee, and what specifically does it pay for?
See the full checklist: Questions worth asking before you sign anythingHide the full checklist: Questions worth asking before you sign anything
- Which numbers appear in the monthly report, and do they include cost per booked customer rather than only impressions and clicks?
- Who is actually doing the work, and who do I speak to when something is wrong?
- How is our spend separated from your fee on the invoice?
If any answer arrives as a reassurance rather than as a fact, ask again. The vocabulary that recurs in this market, locked in, they went quiet, vanity metrics, do I own the account, is almost entirely made of questions somebody did not ask early enough.
You cannot budget what you cannot measure
Every step in the method above depends on knowing how many enquiries arrived and how many became customers. Most businesses that cannot set a budget cannot set one because those two numbers do not exist yet.
Start with calls, because for most local businesses that is where the majority of enquiries arrive and where the majority of attribution is lost. A call that rings the main line and is recorded nowhere is a lead the business paid for and cannot count.
Then close the loop at the other end. An enquiry logged and never followed through to whether it became a customer gives you a cost per lead and no way to know whether those leads were worth having. That gap is where most disagreements between a business and its agency actually live.
This is cheap work compared with everything else on this page, and it makes every subsequent budget decision an argument about evidence rather than an argument about opinion.
Related reading
These pages cover the pieces the method above depends on.
Tracking where leads come from
Getting calls, forms and chats attributed, which is step zero for any budget you intend to defend.
Tracking where leads come fromWhat counts as a good conversion rate
The multiplier in step three, and why it varies so much by traffic source.
What counts as a good conversion rateAre Google Ads worth it for a small business
Whether paid search is the right place for the media portion of the budget at all.
Are Google Ads worth it for a small businessWhat a website costs
The other half of most first-year marketing budgets, treated with the same arithmetic.
What a website costsHow to choose a marketing agency
The longer version of the checklist above, including how to compare two proposals fairly.
How to choose a marketing agencyWant help running the arithmetic on your own numbers?
Bring your enquiry count and your close rate and we will work the figure out with you, including telling you if the target is not fundable yet.
Straight answers.
What percentage of revenue should a small business spend on marketing?
There is no percentage we can defend for a local business. The widely quoted federal small business figure cannot be traced to any published page, and the one properly sourced number available describes large companies rather than local ones.
The CMO Survey put marketing at 9.0% of company revenue in February 2026, across 308 mostly large US firms. Treat it as context, then build your own figure from your customer value and close rate.
How much should I spend in the first month?
Enough to produce a readable result, which depends on your category's cost per enquiry rather than on a round number. A budget that buys a handful of clicks a week will not tell you anything before it runs out.
If the arithmetic says the minimum useful budget is unaffordable, that is a genuine finding. Narrow the scope to fewer services or a smaller area rather than spreading the same money thinner.
Is it cheaper to hire in-house or use an agency?
It depends almost entirely on how many specialisms the work needs. One competent generalist in-house is often cheaper for a single-channel programme and struggles when the work spans paid media, content, technical site work, design and reporting.
Compare total cost honestly: salary plus employment costs plus software plus the management time, against a fee. Then compare what each option can actually cover.
Should I pay an agency based on results?
It can work where a result is unambiguous and both sides trust the counting. It gets difficult when leads arrive and the business cannot convert them, because the fee is owed for work the client does not value.
If you go this route, agree the definition of a qualified lead in writing, agree which system counts, and agree what happens when the two systems disagree.
What happens to my accounts and data if I leave my agency?
If the accounts were set up correctly, nothing. They belong to you, the agency's access is removed, and the history stays where it is.
If they were not, you can lose years of conversion data, negative keyword lists and analytics history. Establish ownership at the start, when it is an administrative question rather than a dispute.
Why do most agencies not publish their prices?
Usually because scope varies and because a published number invites comparison on price alone. Both are real reasons and neither helps the buyer, which is why the most repeated complaint in this market is being unable to compare two proposals.
We publish ours. If a proposal will not put a figure on paper, ask what would have to be true for it to.
How long should I commit for?
Long enough for the channel to show something, which is a few weeks for paid search and several months for organic work. That is an argument for patience, not automatically for a long contract.
The combination to avoid is a long commitment alongside accounts owned by the agency. Either one on its own is manageable. Together they remove your ability to change your mind.
Where this comes from.
Primary documentation and published research behind the guidance on this page.
- Google Ads Help: Manage access to your Google Ads account (opens in a new tab)How account ownership and access levels actually work.
- FTC: Advertising and marketing basics for business (opens in a new tab)
- Google Ads Help: About conversion measurement (opens in a new tab)
- The CMO Survey, February 2026 (Duke Fuqua, n=308 US marketing leaders) (opens in a new tab)
- LocalIQ, Search Advertising Benchmarks 2026 (opens in a new tab)
Talk to the team
A short call, a look at how the business currently shows up, and a straight answer on what we would do first.