How to choose
an SEO agency.
The questions worth asking, what a good answer sounds like, and the two checks that tell you more than any pitch. We answer all of them about ourselves at the bottom.
Written to be useful whether or not you hire us
Who wrote this, and why that matters
How were your case study results actually produced?
A screenshot of a graph going up says nothing about what caused it. Traffic rises when a competitor closes, when a brand runs television advertising, or when someone counts branded searches as an SEO win.
- A good answer
- They talk about specific pages, specific terms and what changed, and they can tell you which parts did not work. They will say where the client already had an advantage.
- A bad answer
- Percentages with no baseline. "Traffic up 400%" from a site that had eleven visitors is true and meaningless.
Can you show me an example from a business shaped like mine?
Ecommerce SEO, local service SEO and B2B SEO are close to different jobs. Being excellent at one says little about the others, because what moves the ranking is different in each.
- A good answer
- A relevant example, or an honest admission that they have not done your kind of business before and what they would do about that.
- A bad answer
- A logo wall where every case is a sector other than yours, and a claim that the principles are all the same.
Are my terms actually winnable, and how do you know?
This is the question almost nobody asks and the one that decides whether the spend was ever going to work. If the pages ranking for your term have hundreds of other websites linking to them and you have three, no amount of writing closes that gap in a year. The agency can tell before starting, using tools they already pay for.
- A good answer
- They check before quoting, name the measure they used, and are willing to say a term is not realistic and suggest a narrower one.
- A bad answer
- Enthusiasm about a competitive term with no discussion of what is behind the pages already ranking for it.
What exactly happens in the first 90 days?
SEO retainers are easy to sell and hard to inspect. Without a stated plan the first quarter can be absorbed by audits and setup, and by the time you notice, you have paid three invoices for a document.
- A good answer
- Named deliverables with rough dates, and something visible on your website inside the first month.
- A bad answer
- "Discovery, then strategy, then implementation", with no artefact you could point at.
Who is doing the work, and will I speak to them?
The people in the pitch are frequently not the people doing the work. That is not automatically wrong, but you should know before signing rather than after.
- A good answer
- You meet whoever will do it, and you can reach them without going through an account manager.
- A bad answer
- You cannot get a straight answer about who will be on it, or the answer changes after you sign.
What do I own if I leave?
Some arrangements leave you with nothing transferable: content on the agency's platform, links that vanish when the retainer stops, a rented site.
- A good answer
- Everything is on your domain and in your account, and stops nothing if you leave.
- A bad answer
- Pages hosted by the agency, or links that are rented monthly. Rented links stop working when you stop paying, which is a subscription rather than an asset.
What does the contract actually commit me to?
Twelve-month minimums are common and often defended on the basis that SEO takes time. SEO does take time; that is an argument for patience, not for being locked in with someone who turns out to be wrong for you.
- A good answer
- A clear term you understood before signing, with a stated way out.
- A bad answer
- A long minimum introduced late in the process, or an auto-renewal you have to remember to cancel.
How will we know whether it worked?
If success is not defined in advance, any outcome can be presented as one. Impressions and "visibility scores" both go up without a single extra customer.
- A good answer
- A recorded baseline before anything starts, and reporting on terms you agreed were worth winning.
- A bad answer
- Monthly reports full of numbers that only ever go up, none of which are enquiries.
Are you doing anything that could get my site penalised?
Some tactics still work for a while and then stop working badly. You carry that risk, not the agency, and the damage lands long after the invoices stop.
- A good answer
- A plain description of how links are earned, and a refusal to buy them.
- A bad answer
- Vagueness about link sources, or a private network described as "our publisher relationships".
What are you doing about AI answers?
A growing share of buyers now get an answer rather than a list of links. Being quoted in that answer is a different job from ranking, and most of the market is not doing it yet.
- A good answer
- They can explain how being cited differs from ranking, and what on the page affects it.
- A bad answer
- Treating it as the same work under a new name, or as a fad worth ignoring.
Two answers that should end the conversation
A guaranteed ranking
Nobody controls the ranking, so the guarantee is either meaningless or the terms have been chosen because nothing was competing for them. Ask what happens if the guarantee is missed. The answer is usually more free months of the thing that did not work.
A price quoted before anyone looked at your terms
The cost of ranking depends almost entirely on what is already ranking. An agency that quotes before checking is either not checking, or the price is unrelated to the work. Both are worth knowing before you pay.
The same ten questions, answered about us
A guide whose author would not survive it is an advert. So here are ours, including where we come off badly.
| The question | Our answer |
|---|---|
| How were your results produced? | We are a young agency and our public proof is our own site, not a client graph. We publish our starting numbers, including a Domain Rating of zero, and update them as they move.Not a strong answer |
| An example shaped like my business? | We work with UK B2B and local service businesses. If you are ecommerce, we are not the right fit and will say so on the call.Not a strong answer |
| Are my terms winnable? | We check before quoting and tell you either way. It is the first thing we do and the reason we sometimes turn work down. |
| What happens in the first 90 days? | A brief you approve, then a page built and published, then off-site work supporting it. Something is live on your site in the first month. |
| Who does the work? | Two founders. You speak to whoever is doing it, because there is nobody else to speak to. |
| What do I own if I leave? | Everything. Pages are on your domain, in your accounts, and nothing stops working when you stop paying. |
| What does the contract commit me to? | Monthly rolling, no minimum term. |
| How will we know it worked? | A recorded baseline before anything starts, then the same measures reported against it, including the months they do not improve. |
| Anything that could get my site penalised? | We do not buy links. Off-site work is making the business genuinely findable elsewhere, which is slower and does not stop working. |
| What about AI answers? | It is half of what we build for, and we can show you exactly which parts of a page affect it. |
Two of those answers are weak, and they are weak in the same way: we are new. If a track record with businesses exactly like yours is what you need most, an established agency is a reasonable choice and we would rather you made it deliberately than found out later.
Questions people ask while choosing
Straight answers. If yours isn't here, ask it on the call: we'd rather tell you no than sell you the wrong thing.
If you want the same questions answered about a specific term of yours, that is what our SEO service starts with. Local businesses can also read how this works in Middlesbrough.
Ask us the ten questions
Book 15 minutes and put this list to us. We will also pull the live results for a term you care about and tell you whether it is winnable, which is the one question you cannot answer on your own.
No minimum term · we say when a term is not worth chasing