- Work out which gap you are buying first: seniority, execution hours, or a permanent internal capability. They point to different answers.
- Percentage-of-spend pricing pays your adviser more when you spend more. Flat fees remove the conflict; ask for one.
- Ask who will be in the account on a Tuesday, then compare the answer to who is in the pitch.
- Own your ad accounts, your tag container and your data. Never launch on someone else's account.
- At 90 days you should have clean tracking, a documented allowable CAC and three implemented fixes. Not a deck.
First, work out which gap you are buying
Most hiring goes wrong before anyone is interviewed, because the brief is "we need help with our ads" rather than a description of the specific thing that is missing. There are three gaps, they look similar from the outside, and they point to different answers.
A seniority gap means someone is running the accounts and the work is competent, but nobody is making the decisions above the campaign: what a customer is worth, what the budget should be, which channels deserve to exist. This is the most common gap and the one a consultant fits best, because the constraint is judgement rather than hours.
An execution gap means the direction is clear and there is simply more work than people. Creative production, feed management, multi-market builds, constant campaign maintenance. This is what agencies are structured for, and a single consultant is a poor substitute for a team.
A permanence gap means the programme is mature enough that the knowledge should live inside the business. That is an in-house hire, and it works well when there is already a senior person to hire into rather than a role someone has to invent alone.
| Consultant | Agency | In-house | |
|---|---|---|---|
| Buys you | Judgement and accountability | Capacity and coordination | Permanence and context |
| Who does the work | The person you hired | Often a junior, day to day | One person, learning as they go |
| Time to output | Days | Weeks, after onboarding | Months, including recruitment |
| Real monthly cost | Flat fee | Retainer or % of spend | Salary, on-costs, tools, training |
| Scales badly when | You need 40 hours a week of production | Your account is small relative to their book | The one person leaves |
| Honest best fit | Seniority is the constraint | Volume is the constraint | A senior lead already exists |
The fourth option
Keep execution in-house and buy senior oversight: a standing review, a measurement framework and someone accountable for the allocation, without replacing the team you already have. It is cheaper than either full alternative and it is the right answer more often than it gets chosen, largely because nobody advertises it.
When the answer is nobody, yet
Below roughly five thousand dollars a month in ad spend, management fees consume a large enough share of the total that the maths rarely works. At that level you are usually better served by learning the fundamentals yourself, buying a one-off audit and a build, and revisiting when spend justifies ongoing management. A consultant who tells you this is more useful than one who takes the engagement.
Pricing models, and the conflict each one creates
Every pricing model creates an incentive. None of them is dishonest, but you should know which way yours pulls before you sign.
| Model | How it works | Which way it pulls |
|---|---|---|
| Flat monthly fee | Fixed amount for a defined scope | Toward efficiency of effort. Watch scope creep and under-servicing |
| Percentage of ad spend | Typically 10–20% of media | Toward higher spend, whether or not spending more is right |
| Hourly | Billed against time | Toward hours rather than outcomes; hard to budget |
| Performance-based | Fee tied to revenue or leads | Toward whatever the metric is, including gaming it |
| Hybrid | Base fee plus a performance element | Reasonable if the metric is margin-aware and jointly agreed |
Percentage of spend is worth a closer look, because it is still the default in much of the market. It sounds aligned: the adviser does well when you grow. What it actually does is pay them more to recommend increasing budget and less to recommend the thing that is frequently correct, which is to spend the same money in fewer places. It also penalises you for scaling, since your fee rises with spend whether or not the work does.
Performance-based pricing has the opposite problem. Tie a fee to leads and you will get leads, including the cheap ones that never close. Tie it to revenue measured in the ad platform and you are paying against a number the person being paid also configures. If you go this route, the metric must come from your own systems, be margin-aware, and be agreed in writing before the first month.
Ask what happens if you pause
A useful and revealing question: if we paused media spend for a month to run a holdout test, what happens to your fee? Under percentage-of-spend the honest answer is that they earn almost nothing, which tells you how likely they are to ever recommend that test.
The questions worth asking
Most interview questions invite a rehearsed answer. These are the ones where the answer is either specific or evasive, and the difference is easy to hear.
About the work itself
- Who will be in my account on a Tuesday afternoon? Name and seniority. Then compare it to who is in this meeting. The gap between those two answers is the single most predictive thing you will learn.
- How would you work out what we can afford to pay for a customer? A good answer starts with gross margin and a payback horizon. A weak one starts with an industry benchmark return on ad spend.
- How do you handle branded search in reporting? If it is not separated from non-brand, every efficiency number you are shown will be inflated and nobody will be lying.
- Our platforms report more revenue than we actually took. Why? Anyone senior will explain double counting and attribution windows in under a minute. Anyone who promises to fix it does not understand it.
- Tell me about an account you could not fix. Twenty years of work contains failures. An unbroken record of success means either a short career or a selective memory.
- What would make you tell us to spend less? A concrete answer, involving saturation or margin, indicates someone managing your money. Vagueness indicates someone managing your account.
About the arrangement
- Who owns the accounts, the tag container and the data? The only acceptable answer is that you do.
- What does the exit look like? Notice period, handover, documentation. Ask before you need it, because the answer changes once you do.
- How many other clients, and how many in our category? Capacity and conflict, in one question. A very large client list and a very personal pitch do not coexist.
- What is in scope, and what triggers a new quote? Get this written down. Scope ambiguity is the most common cause of a relationship souring in month four.
Ask for the audit before the proposal
Anyone worth hiring can look at your accounts and tell you three specific things that are wrong, with evidence, before you have paid them anything. A proposal that arrives full of methodology and free of observations about your account was written before they opened it.
Red flags
| What you hear | What it usually means |
|---|---|
| "We'll run it through our agency account" | You will not own your data, and leaving will cost you your history |
| A guaranteed return on ad spend or lead volume | Either the guarantee has escape clauses, or the number is being manufactured |
| A twelve-month lock-in "so the strategy has time to work" | Ninety days is enough to show direction. The rest is churn insurance |
| Reporting only from inside the ad platform | You will never see whether the numbers reconcile to your bank account |
| Google Premier Partner status offered as the main credential | It largely reflects managed spend and certifications, not client outcomes |
| Case studies with percentages but no absolute numbers | "ROAS up 340%" from a base of almost nothing is a rounding error with a chart |
| Optimisation score presented as a performance metric | It measures adoption of Google's recommendations, some of which raise spend without raising profit |
| Nobody asks about your gross margin | Whatever target they set will have come from somewhere other than your business |
None of these is automatically disqualifying on its own. Two or three together describe an arrangement that will be difficult to leave and difficult to evaluate, which is a combination worth avoiding.
Terms worth insisting on
Four clauses matter more than everything else in the document, and all four are about what happens when the relationship ends rather than while it is working.
- You own the accounts. Google Ads, Meta Business Manager, the tag container, the analytics property. They get access; you hold ownership. Launching inside someone else's account means your conversion history, audience lists and learning periods are theirs, and rebuilding from zero is a genuine cost.
- A short notice period. Thirty days is standard and sufficient. Anything longer is protecting the supplier from the consequences of poor performance, which is precisely the wrong risk to be insuring.
- Documentation as a deliverable. The measurement setup, the naming conventions, the attribution settings and the reasoning behind the current structure, written down and kept current. Without it, a handover is an archaeology project.
- Data portability on exit. Historical reporting, creative assets, feed configurations and audience definitions handed over in usable formats, with a stated timeframe.
The clause to read twice
Watch for anything describing work product, structures or "methodology" as the supplier's intellectual property. Campaign builds and account structures inside accounts you own should not be leaving with them. Ambiguity here is occasionally accidental and always worth resolving before signature.
Paid Media Plus works without lock-in contracts for exactly this reason: an arrangement that has to be legally enforced was not earning its keep. The service agreement sets out the full terms, and the pricing brochure covers how engagements are scoped.
What good looks like in the first 90 days
Set the expectations at the start, in writing, so the first review is a comparison rather than a negotiation. A reasonable shape:
| By | You should have | Warning sign |
|---|---|---|
| Week 2 | An audit naming specific problems with evidence, and the three highest-impact fixes | A generic deck about their process |
| Week 4 | Conversion tracking verified against your own systems; the gap quantified | "Tracking looks fine" with nothing checked |
| Week 6 | A documented allowable cost per acquisition, derived from your margin | A target return on ad spend with no stated source |
| Week 8 | The three fixes implemented, not just recommended | A longer list of recommendations |
| Week 12 | A reconciliation of platform claims against banked revenue, and a defended allocation | Platform-only reporting and a request for more budget |
Note what is absent from that list: dramatic performance improvement. Ninety days is enough to fix foundations and establish direction, and often enough to show early movement, but a transformation inside a quarter is usually a sign that something was badly broken rather than that something clever happened. Judge the first quarter on whether the work was done and the reasoning holds, then judge the second on the numbers.
Give them the uncomfortable information
Real gross margin, the products you actually make money on, the leads that never close, the seasonality that embarrasses you. Consultants working from public information and a dashboard produce advice that reflects public information and a dashboard. The engagements that work are the ones where the awkward numbers arrive in week one.
Frequently asked questions
How much does a paid media consultant cost in Australia?+
Fees vary widely with account size and scope. The more useful question is the pricing model rather than the number: flat monthly fees keep the incentive clean, whereas percentage-of-spend pays your adviser more when you spend more regardless of whether spending more is correct. Whatever the arrangement, include the fee in your cost per acquisition calculation, because media spend alone understates what acquiring a customer actually costs you.
Consultant or agency — which is better?+
Neither, universally. An agency is the right answer when you need volume of production and coordination across several people, and your spend is large enough that you will be a priority account rather than an afterthought. A consultant is the right answer when the constraint is seniority rather than hours, which is the more common situation. Identify which gap you are filling before you compare proposals.
How long should I give a new consultant before judging results?+
Ninety days for direction and foundations, six months for compounding results. Inside the first quarter you should see tracking verified, an allowable cost per acquisition documented, and three specific fixes implemented rather than recommended. If none of that exists at 90 days, the timeline is not the problem.
Should I sign a twelve-month contract?+
There is rarely a good reason to. The argument that strategy needs twelve months to work does not survive contact with the fact that ninety days is enough to demonstrate direction. Long lock-ins protect the supplier against the consequences of underperformance, which is exactly the wrong risk to be underwriting. A thirty-day notice period is sufficient for both sides.
What should I look for in case studies?+
Absolute numbers alongside percentages, the starting position, the timeframe, and what specifically was changed. A 340 percent improvement from a tiny base is arithmetic rather than skill. Be equally wary of results with no mention of what was tried and did not work, because real accounts contain both.
Can a consultant work alongside our in-house team?+
Yes, and it is frequently the best-value arrangement available. The team keeps execution and context, while the consultant supplies the senior decisions: allowable cost per acquisition, channel allocation, measurement framework and a standing review. It costs less than either a full agency retainer or a senior hire, and it develops the internal team rather than sidelining them.
Where this fits
Run the numbers
Before the first meeting, work out your own numbers so you can tell whether their targets come from your margin or from a benchmark. The target CPA calculator and the break-even ROAS calculator take a couple of minutes each.
Hiring well comes down to being specific about the gap, choosing a pricing model whose incentives you can live with, owning your own accounts, and setting a 90-day expectation in writing before anyone starts. Most bad engagements were avoidable at the brief.
If you want to see how this practice works before committing to anything, the free audit gives you the findings and the three highest-impact fixes whether or not we end up working together, and the paid media consulting page sets out how engagements are structured. Worth reading first: the strategy guide, so you can tell whether the person across the table is starting from your margin or from a benchmark.