Start with the problem, not the discipline
Most briefs arrive already naming a solution. People ask for SEO, or ads, or social, when what they actually have is a problem: not enough enquiries, the wrong kind of enquiries, or enquiries that cost more every quarter. Naming the solution first hands the agency the easiest possible job, because now they only have to sell you the thing you already asked for.
Describe the problem instead, and see what each agency proposes. The good ones will sometimes tell you that the thing you asked for is not the thing you need, and occasionally that the answer is not them at all. That single behaviour is more informative than any case study, because it is the one thing an agency has an incentive not to do.
Five questions that are hard to fake
Ask these in the first call. None of them can be answered well from a script, and the hesitation tells you as much as the answer does.
- Who actually does the work, and what is their experience? The person selling is rarely the person delivering. Ask to meet whoever will be doing it.
- What happens in month one, specifically? A vague answer here usually means the first month is discovery, meetings and a strategy document you paid for.
- How do you verify a change actually took effect? Recommending a fix, shipping a fix and confirming a fix on the live page are three different things, and only the third is worth paying for.
- What do I own if I leave? Website, domain, analytics, Search Console, ad accounts, content. Get the answer before you sign, not after.
- Name something you would not do for us. An agency with no limits has no expertise, only capacity.
Red flags worth walking away from
Not all of these are dishonest. Some are just how a business gets built when nobody pushes back. All of them cost you.
- Guaranteed rankings, citations or dates. Search engines and AI assistants are third-party systems nobody controls. Anyone guaranteeing an outcome inside them is guessing, and has decided you will not check.
- Accounts held in the agency's name. If your analytics, ad account or domain sits in their tenancy, leaving is expensive by design.
- Reporting that only goes up. Real work has bad months. A report with no losses in it is marketing, not measurement.
- No named team. If you cannot find out who is doing the work, you cannot assess whether they can do it.
- Auto-renewing fixed terms. A minimum term is fair. A term that silently renews into another term is a retention strategy, not a delivery one.
- A proposal with no line items. If the scope is one paragraph and one number, you cannot compare it to anything, which is usually the point.
What a fair agreement looks like
Contract terms are where good intentions get tested. This is the shape of an agreement that survives a bad quarter without trapping either side.
| Ask for this | Walk away from this | |
|---|---|---|
| Term | A minimum term, then month to month on clear notice | A fixed term that auto-renews into another fixed term |
| Scope | Itemised: what, how much, how often, and who does it | One paragraph and one number |
| Ownership | You own the site, accounts, content and data | Assets held in the agency's own accounts |
| Reporting | A change log of what shipped and what was verified | A monthly deck of traffic charts with no actions in it |
| Approval | A named senior person signs off changes before they ship | Changes appear on your site with no record of who approved them |
| Exit | Documented handover, access transferred on request | Handover quoted as a separate project |
Comparing quotes that are not comparable
Two proposals at the same monthly figure can differ by an order of magnitude in the senior time behind them, and nothing on the page will tell you which is which. The fix is to stop comparing prices and start comparing scope.
Ask every agency to restate their proposal as line items with quantities: how much of each activity, performed by whom, at what frequency. Once both are in that form the comparison is arithmetic rather than instinct. An agency that will not itemise has answered a different and more useful question about how they work.
If you want the mechanics of why agency quotes vary so much, we wrote that up separately in what a marketing agency actually costs in Australia.
Now turn the checklist back on us
A selection guide written by a seller is worth nothing unless it is willing to lose on its own criteria. So, against the list above: we publish a floor rather than a rate card, and scope every engagement individually, which means you cannot price us from the website alone. That is a real friction and some buyers will rule us out for it.
On the rest: every change is approved by a senior SEO before it ships and re-checked on your live page afterwards, and the change log sits in your portal rather than being summarised in a deck. You own your site, accounts and content throughout. We run a minimum term and then month to month on 60 days notice, with no auto-renewal into a new term. We do not sell advertising, so when ads are the right answer we say so and have nothing riding on it.
And the last question, the one we told you to ask: the thing we will not do is promise you a ranking, a citation or a date. The engines are third-party systems and anyone quoting you a date is guessing. What we commit to is the work, shipped weekly and verified on the page.