Nearly every agency page says the same six things, and none of them are checkable. Data-driven. Results-focused. Transparent. Partners, not vendors. You cannot buy on any of that, so people fall back on the two things they can compare, which are the price and how much they liked the call. Both are terrible signals, and here is what to use instead.
What should I ask an SEO agency before hiring them?
Three questions, and you want the answers in writing before the first call. What does it cost. What happens when the work does not work. Who actually does the work. An agency that cannot answer those three in writing is an agency whose price depends on what you look like you can afford.
Notice that none of them are about strategy. Strategy is what a good discovery call is for, and you cannot evaluate one from the outside anyway. These three are about whether the thing you are buying has a shape. If it does not, no strategy conversation will save it.
- Ask for the price list, not a proposal A proposal is a number invented for you. A price list is a number that exists whether or not you called. If the answer is "it depends on your goals", ask what the cheapest thing they sell is and what it produces. If that also cannot be answered, the price is being set by what you look like.
- Ask what happens when it does not work Not a guarantee, because nobody can guarantee a ranking. A written replacement or remedy policy: what happens if a link is removed, if an article misses the brief, if a deliverable arrives late. If there is nothing in writing, the answer is nothing.
- Ask who does the work and where The senior person on the call is often not the person on the account. That is normal and fine, as long as you are told. It is only a problem when it is a surprise.
- Ask to see one real deliverable Redacted is fine. A sample audit or link sheet tells you more in two minutes than an hour of case studies, because you can see whether it is a document a human wrote or a tool's export with a logo on it.
- Ask what they will not sell you Anybody who sells everything to everybody is not making judgements. The useful answer names sectors, tactics and situations they turn down, and why.
What are the warning signs of a bad SEO agency?
A guaranteed ranking, a price that only exists after a call, a contract that locks you in before anything is delivered, and reporting built on metrics that no search engine uses. Any one of those is a reason to slow down. Two together and you are looking at the sales process, not the service.
The guarantee one is worth spelling out, because it is sold so confidently. Nobody controls a search engine. Nobody has a switch. An agency that guarantees position one for a term is either picking a term nobody searches, which is easy and worthless, or is about to buy links that will cost you later. Both are common and both are sold with the same certainty.
| What they say | What is usually happening | What to ask |
|---|---|---|
| "Guaranteed first page" | A term with no search volume, or bought links. | Which exact terms, and what is their monthly search volume? |
| "Custom pricing for your goals" | The price is set by your budget, not by the work. | What is the smallest thing you sell, and what does it cost? |
| "We increased traffic 400%" | Often true and often irrelevant. Traffic is not revenue. | What happened to enquiries and sales in the same period? |
| "Twelve month minimum" | The first deliverable is a long way from the first invoice. | What arrives in month one, and can I stop after it? |
A long commitment is not automatically bad. Authority genuinely takes months. What matters is whether anything checkable arrives before the commitment is locked.
Why is cheap SEO expensive?
Because the cheapest way to move a ranking is to buy links at scale from sites that sell them to everybody, and that footprint is exactly what a search engine is built to find. The ranking arrives in weeks, the collapse arrives in months, and removing the links costs more than the links did.
This is not a moral argument, it is an arithmetic one. A cleanup involves auditing a link profile, contacting site owners, documenting refusals, and filing a disavow, which is more hours than the original campaign took. People find their way to that service every week, and they almost always arrive holding an invoice from somebody who promised the opposite.
- Earned editorial placements
- Bought at volume
Illustrative, not a client result. Bought links win the first six months and that is precisely why they sell. The line that matters is the one still standing at month twelve, and the recovery work after the drop is billed to you, not to the agency that sold it.
Growth is a process. Nothing here happens overnight, and anybody promising you overnight results is lying to you.
See how we place linksShould an agency publish its prices?
There is no good reason not to, and one bad reason to hide them. Published prices mean the number is the same for a founder and for a procurement department. Hidden prices mean somebody looks at your website, your funding and your job title, and decides.
The usual defence is that every engagement is different. Some are. A twelve-country enterprise programme genuinely has to be priced by hand. But the entry work is not different: an audit is an audit, a link is a link, a tester is a tester. If none of it can be priced in the open, the variable being priced is you.
Is it cheaper to do SEO in-house?
Cheaper per hour, usually. Cheaper per outcome, only if the person doing it has done it before. The expensive version of in-house is a capable generalist learning on your site for six months, because the mistakes take longer to undo than they took to make.
A reasonable split, if you have somebody good: keep strategy, keyword decisions and publishing in-house, where context matters most and nobody outside knows your customers. Buy the parts that need either specialist judgement or relationships you do not have, which in practice means technical audits, outreach and anything requiring a panel of real people. That is also the cheapest way to find out whether you need an agency at all.
How long should I commit to an agency?
Commit to the first deliverable, not to the year. Buy the smallest real piece of work they sell, read it, and decide from the document rather than from the pitch. If nothing they sell is small enough to test, that is the answer to the question.
The reason to hold this line is that a proposal and a deliverable are written by different parts of an agency. The proposal is the best writing in the building. The deliverable is what you are actually buying for the next twelve months. Spending a hundred pounds to see the second one is the highest-return decision in this whole process.