How to Vet an SEO Agency When You Cannot Assess the Work

The uncomfortable position most buyers are in: you need to hire a specialist, and the thing that makes them a specialist is precisely the thing you cannot evaluate. Every agency has case studies. Every agency says the right words. The pitch is the product they are best at.

You do not need to become an SEO to fix this. You need questions where a bad answer is obvious even if you cannot judge a good one.

Questions that work

“What would you look at first, and what would make you tell me not to hire you?”

The second half is the real question. Someone who has never turned down work is either very lucky or not being straight with you. A good answer names a specific situation: budget better spent elsewhere, a site that cannot convert the traffic it already has, an intake process losing more leads than marketing could add.

Anyone who cannot imagine a circumstance where they are the wrong choice is telling you something.

“Walk me through a campaign that did not work.”

You are listening for whether they diagnose or blame. “The client did not implement our recommendations” is a non-answer that appears constantly. A real answer sounds like: we assumed X, the data said otherwise by month three, here is what we changed and here is what I would do differently now.

“What will you need from my team, and how much of their time?”

Anyone claiming SEO requires nothing from you is either not doing much or has not thought about implementation. Real work needs developer time, content approvals, and access. If they have not scoped that, they have not scoped the engagement.

“How will we know in month three whether this is working?”

The trap answer is rankings. Rankings are an input. A good answer names leading indicators appropriate to your sales cycle — qualified enquiries, form completions, tracked calls, opportunities created — and is honest that revenue attribution lags.

“Who specifically does the work, and will I ever speak to them?”

Agencies frequently sell with senior people and deliver with junior ones. That is not inherently wrong; it is only wrong when it is hidden. Ask for names and ask who is on the calls in month six.

“What happens to everything you have built if we stop working together?”

The right answer is that you keep it: the site, the content, the analytics configuration, the documentation. Watch for anything hosted on their infrastructure, proprietary tracking you cannot export, or content on a domain they control.

What good answers have in common

They contain numbers, timeframes and caveats. They distinguish between what is measured and what is estimated. They occasionally say “I do not know, I would need to look.”

That last one is the strongest signal in the entire process. Certainty about a site they have not examined is not confidence, it is a script.

Red flags

Guaranteed rankings. Nobody controls Google’s results. A guarantee is either meaningless, scoped to terms nobody searches, or describes something other than what you think you are buying.

Percentage-of-spend pricing on paid media. It rewards the vendor for spending more of your money. It is common, it is not automatically disqualifying, but it is an incentive worth naming out loud.

A proposal with no discovery. If they have not looked at your analytics, your site or your competitors before proposing scope, the proposal is a template.

Deliverables measured in volume. “Twenty blog posts a month” and “fifty backlinks” are quantities, not outcomes. Quantity-based deliverables usually mean the work is being produced to hit a number rather than to move anything.

Vagueness about link building. Ask exactly how links are acquired. Some answers are fine — digital PR, editorial placements, genuine partnerships. Some carry real risk. Any reluctance to describe the method is itself the answer.

No exclusivity in a local market. If your agency also works with your closest competitor in the same market and practice area, the work is structurally self-defeating. Ask directly.

Reference calls are worth more than case studies

Case studies are marketing. References are not, if you ask properly.

Ask the reference: what surprised you? What took longer than expected? What do you wish you had asked before signing? Would you hire them again for the same scope, or a different one?

Ask for a reference whose engagement ended. How an agency behaves during offboarding tells you more than any active client will.

The structural fix

None of this fully closes the gap. You are still judging expertise you do not have, with better questions.

The genuine fix is having someone independent in the room who can assess the actual craft — read the technical recommendations, sanity-check the keyword strategy, tell you whether the proposal is substance or theatre. Not to run the engagement, just to help you pick.

That is what hiring advisory is. It applies equally to assessing an agency and to vetting an in-house hire, and it costs a fraction of a mis-hire at specialist level.

If you would rather do it yourself, the questions above are most of the value. Use them in order, take notes, and pay attention to which candidates get more specific under pressure and which get vaguer.