Most advice about agency pricing gives you bands. A number for SEO, a number for a website, a number for ads. We have written those ourselves, for SEO and for websites, and they are useful.
What they do not tell you is whether the number in front of you is reasonable for what is being promised. Two agencies quote $2,400 a month. One is good value and one cannot possibly deliver what it has listed, and the proposals look almost identical.
There is a way to tell, and it takes about a minute.
Divide the fee by the rate
A retainer is not a price. It is a number of hours wearing a monthly price, and the moment you convert one into the other most proposals answer themselves.
Australian agency rates sit roughly where you would expect. Offshore or junior work runs around sixty to a hundred dollars an hour. A mid-tier Australian agency blends out somewhere between a hundred and twenty and a hundred and eighty, because your account is a mix of junior execution and senior oversight. Genuine specialists charge two hundred and up.
So take the fee and divide. A $1,500 retainer at a $150 blended rate is ten hours a month. A $3,000 retainer is twenty. A $5,000 retainer is a bit over thirty, which is less than one working day a week.
Now read the proposal again. If ten hours a month is supposed to cover keyword research, two blog posts, technical fixes, Google Business Profile maintenance, a paid campaign and a monthly report, it does not. Something on that list is not being done, or is being done in a token way, and you will not find out which for about five months.
This is not an accusation of dishonesty. Agencies write proposals that describe a complete program because that is what wins the work. The arithmetic is simply absent from the conversation, and it is the single most useful thing you can bring to it.
How much of it is actually spent on your account
The second adjustment is less comfortable.
Not all of those hours go on your work. Some go to account management, some to building the report, some to internal meetings about your account, and a share covers the software licences your work consumes. On a well-run account, delivery is somewhere around sixty to eighty per cent of the fee. The rest is the cost of being a client.
That is not waste. Somebody has to read what the specialist did and decide whether it was right, and reporting you can actually check is worth paying for, which is the whole argument in what transparent agency reporting should look like. But it should be visible, not hidden.
So the ten hours is really six to eight hours of work on your business. Run that number against the proposal instead.
A fair question to ask directly: what proportion of this fee is delivery and what proportion is management? An agency that answers plainly is telling you it has thought about it. One that treats the question as hostile is telling you something else.
Why the first three months cost more than they look
Nearly all of the setup sits in the first two or three months and none of it sits in the price.
Access to accounts. Tracking and conversion setup that is usually broken or absent. Research. Writing the actual plan. Getting into the detail of a business the agency did not know existed six weeks ago. That work is real, it is front-loaded, and the monthly fee is flat.
Which is why a minimum term is not automatically a red flag. An agency that lets you walk at month two has done the expensive part and kept none of the return, and the ones that offer no minimum term often compensate by not doing much setup at all.
Three to six months as an initial term is reasonable. Twelve months with no exit is not. The arrangement that signals the most confidence is a short initial term and then month to month, because it says the agency expects the results to hold you rather than the contract.
There is a floor, and below it nothing works
Here is the part that costs us work to say.
Below about $1,500 a month, an ongoing agency engagement struggles. Below $1,000 it cannot function at all, whoever you hire. The reason is the fixed cost of simply having a client: onboarding, reporting, the monthly call, the account admin. That overhead does not shrink with the fee, so at $600 a month it eats almost everything and what reaches your business is a thin, largely automated service.
Plenty of agencies will take that money. The honest advice if that is your budget is to not spend it on a retainer. Buy a one-off audit and a plan, then do the free work yourself, which is genuinely the part that pays best and needs no agency at all. We set out the order in what actually moves the needle and the version you can work through alone in local business marketing.
A business that has done that is a much cheaper client later, because the obvious problems are already solved and the retainer buys progress rather than catch-up.

What being an SME changes
Everything above applies to a sole trader too. Three things change once there are staff, and all three cost hours.
Approvals consume the fee. At an owner-operator, a draft is approved the evening it arrives. At a twenty-person business it goes to a manager, then to someone else, then comes back with changes from two people who disagree. Every round of that is billable agency time doing nothing new. A business with slow approvals genuinely gets less for the same retainer, and it is the one variable entirely within your control.
More stakeholders means more meetings. Each additional person in the monthly call is another person’s hour, and the call is paid for out of your delivery time.
There is usually something to protect. Brand guidelines, an existing customer base, a reputation with trade partners. That is legitimate and it is slower than starting clean.
None of this is an argument for being less careful. It is an argument for deciding who signs things off before the engagement starts, and keeping that to one person.
What cheap actually buys
Cheaper is not automatically worse, and the industry is dishonest in both directions about this.
Cheaper because the agency is small, has almost no overhead, and the person who does your work is the person you met is frequently the best value available to an SME. There is no account manager layer because there is no account manager.
Cheaper because the work is done by juniors with no senior review, or templated and run across two hundred accounts, is a different thing entirely. You will not tell these apart from the proposal, because both use the same words.
One question separates them: who specifically will do the work, by name, and how many other accounts does that person carry? The answer to the second half is the one that matters. Somebody carrying forty accounts is not thinking about yours. The broader list is in what to ask before hiring an agency.
And on fee structures, the one to be careful of is a percentage of ad spend, because it rewards the agency for spending more of your money. We made that case in SEM agency or SEO agency.
Frequently Asked Questions
How do I work out if a marketing retainer is good value?
Divide the monthly fee by a realistic blended hourly rate and look at the number of hours you get. At Australian mid-tier agency rates, somewhere around 120 to 180 dollars an hour blended, a two thousand dollar retainer is roughly eleven to sixteen hours a month. Then ask what the proposal promises and whether it fits in that time. If the list includes strategy, content, technical work, paid campaigns and reporting, it does not fit, and something on the list is not happening. The arithmetic is simple and almost nobody does it.
What is the minimum a marketing retainer can sensibly be?
For ongoing work with a real agency, somewhere around fifteen hundred dollars a month is the practical floor, and below a thousand the engagement cannot function regardless of who you hire. At that level the fixed overhead of simply having a client, the onboarding, the reporting, the meetings, the account admin, consumes most of the fee before anybody touches the actual work. Plenty of agencies accept smaller retainers anyway. The result is usually a thin automated service, which is worse than spending nothing and doing the free work yourself.
How much of my retainer is actually spent on my account?
Less than owners expect, and the honest range is roughly sixty to eighty per cent. The rest goes to account management, reporting, internal meetings about your account, and the share of software licences your work consumes. That is not waste, some of it is genuinely necessary, but it should be visible rather than hidden. A reasonable question to ask any agency is what proportion of the fee is delivery and what proportion is management, and a good one will answer it without flinching.
Why do agencies want a minimum term?
Because the first two or three months carry the setup and the fee does not. Research, access, tracking, tooling and the plan itself are front-loaded, so an agency that lets you leave at month two has worked at a loss. A three to six month initial term is reasonable for that reason. Twelve months with no exit is not, and the useful compromise is a short initial term followed by month to month, which tells you the agency expects the work to justify itself.
Is a cheaper agency worth trying?
It depends entirely on what makes it cheaper. Cheaper because it is small, has low overheads and the person doing your work is the person you met is often excellent value. Cheaper because the work is done by juniors with no oversight, or templated and run at volume, generally is not, and you will not be able to tell from the proposal because both describe themselves the same way. The question that separates them is who specifically will do the work and how many other accounts that person carries.
The bottom line
Stop asking what a marketing agency costs and start asking what the fee buys in hours.
Divide the retainer by a hundred and fifty. Take sixty to eighty per cent of the result as actual work on your business. Hold that number against the list of things the proposal promises. If they do not reconcile, the proposal is describing a program the fee cannot pay for, and the gap will show up as things quietly not happening.
Then ask the two questions that cannot be answered with a brochure. What share of this is delivery rather than management, and who by name is doing the work and how many other accounts do they carry.
If you want somebody to run that arithmetic across a proposal you have already been sent, including one of ours, have a chat with us. We will tell you if the numbers do not add up, including when they are our own.