Cost is the wrong axis
Almost every version of this comparison lines a salary up against a monthly fee and declares a winner. Both sides of that comparison are wrong.
The salary side omits the superannuation guarantee, leave entitlements and notice obligations, payroll tax where the threshold applies, recruitment, software and tools, a desk, and the management time of whoever the role reports to. It also omits the risk of a bad hire, which for a first marketing role is substantial because the people hiring usually cannot assess the craft.
The agency side omits what you do not get: someone in the room, someone who knows why the second branch underperforms, someone who can be redirected on a Tuesday without a scope conversation.
Once both sides are stated fully, cost stops separating them and the real question appears — do you need depth in a few things, or breadth across many?
What one in-house hire realistically covers
Modern marketing is at least six disciplines: message and positioning, paid channels, organic search, content production, measurement and analytics, and the systems that hold customer data. They require different training and different temperaments.
One person cannot be a specialist in six disciplines. Any structure that assumes otherwise is a resourcing plan with a person-shaped hole in it.
A capable hire can genuinely own two, coordinate the rest, and be the accountable owner of the outcome. That is a real and valuable role. What they cannot do is execute all six to a standard that competes with businesses buying specialists for each, and pretending otherwise is how good hires end up leaving.
What you do get is the thing no agency sells: context. An employee learns which jobs are profitable, which customers complain, what the crews are hearing on site and what the business is actually good at. That knowledge accumulates and it never appears on an invoice.
If the plan is to lift an existing team rather than replace it, that is its own piece of work — closer to training and capability building than to recruitment.
What you are actually buying from an agency
Not labour. You can buy labour more cheaply almost anywhere. You are buying four things, and it is worth being explicit about them because they are what you should be assessing.
| What you buy | Why it is hard to hire |
|---|---|
| Breadth on demand | Six specialists, none of them needed full time |
| Pattern recognition | Having seen the same failure across many businesses, not one |
| Capacity that flexes | A launch does not require a permanent headcount |
| An outside opinion | An employee who says the offer is the problem is taking a career risk; an agency is not |
The corresponding weaknesses are real. An agency has less context, competes for attention with other clients, and can leave taking the knowledge with it. That last risk is manageable, and the way to manage it is to insist that accounts, data and documentation live in assets you own — a requirement that belongs in the agreement rather than in a conversation later.
It is also worth reading any lock-in, automatic renewal or one-sided termination clause carefully. Unfair contract terms protections exist for small business agreements, and long lock-ins with no corresponding obligation on the supplier are exactly the shape those protections were written for.
One person cannot be a specialist in six disciplines. Any structure that assumes otherwise is a resourcing plan with a person-shaped hole in it.
The hybrid that works, and the one that does not
Most businesses past a certain size end up with both, and the arrangement succeeds or fails on one detail: who owns the outcome.
The version that works. One internal owner who holds the strategy, the customer knowledge and the measurement, buying specialist execution from outside. The employee is accountable for the result. Suppliers are accountable to the employee. There is one throat to clear and one version of the numbers.
Context is the thing an agency cannot buy and an employee cannot avoid acquiring. Everything else is negotiable.
The version that fails. An internal marketer and an agency both doing overlapping execution, with nobody clearly accountable. Two sets of numbers appear, they disagree, and every meeting becomes a reconciliation exercise. Work gets duplicated, the awkward parts get dropped by both sides, and the relationship ends badly regardless of the results.
The distinction is not about seniority or budget. It is entirely about whether one named person owns the outcome, and it is worth settling before either party is engaged.
Five questions that settle it
Answer these honestly and the decision usually makes itself.
- Is the work continuous or intermittent? Continuous work suits an employee. Intermittent specialist work suits a supplier, and hiring for it means paying for idle capacity.
- Can you assess the craft? If nobody internally can judge whether the work is good, a bad hire will not be detected for a long time. An agency at least fails visibly against an agreed measure.
- How much does the work depend on knowing your business? High dependence favours in-house. Low dependence — technical execution against a clear brief — favours an agency.
- Do you know what good looks like yet? If not, buy it before you hire it. Hiring into a discipline you cannot evaluate is the most expensive way to learn.
- What happens if this person leaves in six months? If the answer is that everything stops, you have a concentration risk that a supplier relationship would not have.
Question four is the one most often skipped, and it is the reason advisory engagements exist — establishing what good looks like is a separate job from doing the work, and doing it first makes both later options cheaper.
Context is the thing an agency cannot buy and an employee cannot avoid acquiring. Everything else is negotiable.
Make either model accountable the same way
Whichever you choose, apply identical standards. The mistake is holding an agency to a scoreboard and an employee to a vibe, or the reverse.
- One outcome measure, agreed before work begins, expressed in booked work rather than activity
- Reporting you can read without translation, stating what changed, why, and what it produced
- All accounts, data and documentation owned by the business, not the supplier or the individual
- A review window long enough to be fair and short enough to matter, agreed in advance
The second point is the one that separates working relationships from expensive ones. If you cannot follow the reporting, you cannot manage the work, and the diagnostic for that is set out in the briefing on reporting you cannot follow.
Whether the execution sits inside or outside, the programme it belongs to is the same one — acquisition measured through to booked work rather than to activity. Decide who owns that, then decide who does it. Doing those in the other order is how organisations end up with two marketing functions and one result.
If you would rather see how we work before deciding, the engagement model is written out in full, including what sits with us and what stays with you.
Common questions
Is an agency cheaper than hiring?
Not reliably, and cost is the wrong axis anyway. A salary comparison omits superannuation, leave, tools, recruitment and management time; an agency comparison omits the context and availability an employee gives you. Decide on breadth versus depth, then check affordability.
What should my first marketing hire be?
Whichever discipline your business needs continuously and cannot brief out cheaply. For most Australian service businesses that is someone who owns the message and the customer relationship, with specialist execution bought in around them.
Can one person run everything?
No, and expecting it produces the most common failure in this decision. One person can own strategy and coordinate specialists, or be genuinely deep in one or two channels. Nobody is a specialist in six disciplines at once.
How do I hold an agency accountable?
The same way you would hold an employee accountable. Agree the outcome measure before work starts, require reporting you can follow without translation, and insist on knowing what changed, why, and what it produced.
When is it clearly time to bring work in-house?
When the work has become continuous, when you can evaluate it well enough to manage it, and when the volume is high enough that a full-time person is busy. Until all three are true, hiring buys idle capacity and loses you breadth.