Marketing agency vs in-house: which should a small business choose?

The question is almost never asked in the abstract. It is asked at a specific moment: the agency invoice has arrived again and the results feel thin, or a founder has been running the ads themselves at midnight and cannot keep doing it, or revenue has grown to the point where marketing clearly needs an owner. At that moment the choice feels binary. It is not. What you are really deciding is which parts of marketing need to live inside the business, and which parts are better rented.

Getting that split right matters more than the agency-or-not question, because the most expensive outcomes come from mismatches. Paying agency rates for work anyone in your team could do is one kind of waste. Hiring a single generalist and expecting them to run paid media, search, email and creative at a professional standard is another. Both are common, and both are avoidable with a clear read of what you actually need.

What does a marketing agency actually cost in Australia?

Comparing an agency to a hire only works when you compare the whole cost of each, not the headline number. Agency pricing generally takes one of three shapes: a fixed monthly retainer, a percentage of ad spend (commonly somewhere in the 10 to 20 per cent range), or a project fee for defined work such as a website build or a campaign launch. Small business retainers frequently run from a few thousand dollars a month for a single channel up to five figures a month for genuinely full-service work.

An in-house hire costs more than the salary. Superannuation, payroll tax where applicable, leave loading, software licences, recruitment cost and the productivity dip while someone learns your business all sit on top. A useful working rule is to add roughly 25 to 35 per cent to a salary figure before comparing it to an annual agency fee. Then add the tools the agency was providing for free out of its own stack, which for a small business can be a meaningful line on its own.

One number makes the comparison honest: the total cost of the marketing function per month, including fees, salaries on-costs, software and the ad budget itself. If you are not sure what you should be spending in total, our guide on how much a small business should spend on marketing in Australia sets the envelope before you decide who spends it.

Agency, freelancer or in-house: how do the three models compare?

Framing this as two options hides the one that fits many small businesses best. A specialist freelancer is a genuine third model with a different risk profile, not a cheaper agency.

Factor	Agency	Freelancer or specialist	In-house
Breadth of skill	Several specialists across channels	Deep in one channel, thin outside it	Usually one generalist at first
Cost shape	Retainer or percentage of spend	Hourly, day rate or small retainer	Salary plus on-costs and tools
Speed to start	Fast, often within weeks	Fastest	Slowest, allow for hiring and ramp-up
Product knowledge	Learned, and lost if you leave	Learned, held by one person	Deepest, and it compounds
Capacity risk	Low, team absorbs absence	High, no cover during leave	Medium, single point of failure
Best when	Multiple channels, unproven mix	One channel needs expert hands	Proven channel needs daily ownership

Read the table as a diagnosis rather than a scorecard. If your honest answer to "which channels matter?" is still "we are not sure", you are in discovery, and renting breadth is cheaper than buying it. If your answer is "search brings us almost everything and we have never worked it properly", you need one expert pair of hands, not a full-service retainer.

When does hiring in-house actually make sense?

There are four signals that the work has earned a salary. The first is proven channel economics: you know which channels acquire customers profitably, so the job is now consistent execution rather than exploration. The second is volume of daily decisions. When someone needs to be inside the ad accounts, inbox and analytics every day, a monthly reporting rhythm becomes the bottleneck.

The third is product depth. If your marketing lives or dies on understanding a technical product, a regulated category or a specific trade, an outsider will always be a step behind, and the cost of that gap shows up as generic messaging. The fourth is simple arithmetic: when the agency fee approaches or exceeds the fully loaded cost of a capable hire, you are paying a premium for flexibility you may no longer need.

Be careful with the reverse case too. Hiring in-house because you are frustrated with an agency often just relocates the problem. If the brief was vague, the measurement unclear, or nobody internally owned the relationship, a new employee inherits exactly the same conditions and struggles for exactly the same reasons.

What should you look for before signing with an agency?

Most bad agency relationships are visible in the sales process if you know what to look for. The strongest signals are structural rather than creative.

• Who does the work. Ask to meet the person on your account, not only the person selling. Senior pitch, junior delivery is the most common gap.
• Scope in specifics. "Manage your social media" is not a scope. Number of campaigns, channels, deliverables and review cadence are.
• Ownership of assets. Ad accounts, analytics properties, pixels, domains and creative files should be in your name from day one, not the agency's. This is the clause that is most expensive to fix later.
• The reporting they choose. If the sample report leads with impressions and reach rather than cost per acquisition and revenue, that tells you what will be optimised.
• Notice period. A long lock-in with a short track record is a poor trade. Favour a reasonable notice period and let performance earn the renewal.

Many of the same tests apply to any external partner, and the broader version of that process is covered in how to choose a business consultant in Australia. It is also worth agreeing which channel is being bought before you buy it, since the Google Ads versus SEO decision changes both the timeline and the skills required.

How do you tell whether it is working?

Set the judging criteria before the engagement starts, not when you are already unhappy. Decide the review window by channel, because they do not move at the same speed. Paid channels usually produce meaningful signal within one to three months. Organic search and content generally need six months or more before the trend is readable, and judging them on a 30-day view will lead you to cancel work that was about to compound.

Then fix the numbers. Two or three outcome metrics, agreed in writing: cost per acquisition, qualified leads or orders, and contribution after costs. Impressions, reach and follower counts are diagnostic at best. Insist on seeing the platform accounts yourself rather than only a formatted summary, because a report can be shaped and an ad account cannot. If spend is flowing to search terms and placements that were never going to convert, that shows up quickly in the raw data, which is exactly the pattern described in our piece on reducing wasted Google Ads spend.

What does the hybrid model look like in practice?

The arrangement most small businesses land on is not a compromise, it is the mature answer. Strategy and ownership stay in-house. Specialist execution is bought. Someone internally owns the numbers, the brief and the customer knowledge. Outside specialists handle the parts that need constant technical currency, such as paid media buying, technical SEO or performance creative.

This works because it puts the compounding assets, product knowledge, customer insight and measurement discipline, inside the business where they accumulate, while renting the skills that would be expensive and fast-decaying to maintain internally. It also gives you a clean exit in either direction: you can change agencies without losing your marketing brain, and you can bring a channel in-house once it is proven without starting from nothing.

A practical sequence for a growing business: start with a specialist on your single most important channel, add an internal owner as soon as someone can be accountable for the numbers full time, then expand outward channel by channel as each one proves out. This is the same staged approach we use in Alpha Vault's growth and performance marketing work, and it keeps fixed costs behind proven returns rather than ahead of them.

How to make the decision

Write down three answers. How many channels do you genuinely need running in the next twelve months? Of those, how many are already proven to acquire customers profitably? And what is the fully loaded monthly cost of each option, with on-costs and tools included? If most of your channels are unproven, rent. If one is proven and demands daily attention, own it. If the honest answer is a mix, build the hybrid deliberately instead of arriving at it by accident.

Whichever way you go, the failure mode is the same: handing over marketing without keeping ownership of the brief, the accounts and the numbers. Keep those three things and either model can work. Lose them and neither will. If you want an outside read on which parts of your marketing should be owned and which should be rented, book a consultation and we will work through the numbers with you.